HomeMy WebLinkAboutLake-Cornelia-and-Lake-Edina-Water-Quality-Study_2019Lake Cornelia and Lake Edina
Water Quality Study
Use Attainability Analyses for Lake Cornelia (updated from 2010)
and Lake Edina (first version)
Prepared for
Nine Mile Creek Watershed District
July 2019
4300 MarketPointe Dr • Suite 200
Minneapolis, MN 55435
952.832.2600 • www.barr.com
REPORT SUMMARY
Lake Cornelia and Lake Edina Water Quality Study
Use Attainability Analysis for Lake Cornelia (updated from 2010)
and Lake Edina (first version)
Prepared for
Nine Mile Creek Watershed District
July 2019
Lake Cornelia and Lake Edina are located in the southeast portion of Edina within the Nine Mile Creek
watershed. The shallow, urban lakes suffer from poor water quality. In fact, both lakes are included on the
state’s impaired waters list for excess nutrients. The NMCWD, a local unit of government that works to solve
and prevent water-related problems, conducted a study of Lake Cornelia and Lake Edina in 2018-2019 to
help address the poor water quality. Additional information on the current lake conditions, causes of the
poor water quality, and recommended management strategies are summarized in this project overview.
Protecting and enhancing the water quality of the lakes within the Nine Mile Creek watershed is one of
the primary goals of the Nine Mile Creek Watershed District. The NMCWD’s lake management program
includes data collection (monitoring), assessment (e.g., studies), and implementation of projects and
programs to protect and improve water quality and aquatic habitat. Utilizing monitoring data
collected by NMCWD in recent years (2015 – 2017), the objectives of this study were to assess
or “diagnose” the lakes’ water quality problems, understand the cause or sources of the
problems, and recommend management strategies to improve the water quality
and overall health of the lakes.
LAKE MANAGEMENT GOALS
When assessing the ecological health of a lake, it is
important to take a holistic approach, considering
the factors including chemical water quality
(e.g., phosphorus concentrations), the health
and quality of the aquatic communities,
and water quantity (see Figure 1). How
recreation and wildlife habitat affect and
are affected by overall lake health are
also considered. Numerical goals exist
for some of these factors (e.g., state
water quality standards), however,
other ecological lake health factors are
assessed relative to narrative criteria
(e.g., criteria that describe the desired
condition) without strict numerical goals.
For this study, the primary goals are to
achieve the water quality standards for
shallow lakes, support more diverse, native
macrophyte (aquatic plant) populations, and
promote a more healthy, balanced fishery.
UNDERSTANDING LAKE CORNELIA
AND LAKE EDINA
WORKING TO MEET DISTRICT GOALS
1
THE LAKES AND HOW THEY ARE CONNECTED
Lake Cornelia is a shallow lake with a northern and southern basin, which are connected by a storm
drain. North Cornelia, spanning 19 acres, has a maximum depth of 7 feet, and a mean depth of
approximately 3 feet. South Cornelia, with a water surface area of 33 acres, has a maximum depth of
8 feet, and a mean depth of approximately 4 feet.
North Cornelia receives stormwater runoff from a relatively large watershed (863 acres), shown
in green in the map below. Land use within the highly developed watershed includes a large
commercial area (including the Southdale Shopping Center), portions of Highway 62 and Highway
100, residential areas (high and low density), and Rosland Park. Most of the runoff from the highly
impervious commercial area drains through a series of waterbodies (i.e., Point of France Pond and
Swimming Pool Pond) prior to reaching North Cornelia. In addition to flows from North Cornelia,
South Cornelia receives runoff from a relatively small, residential watershed (112 acres), shown in
orange in the map below.
Runoff that flows through
Lake Cornelia drains to
Lake Edina, which ultimately
discharges into the North
Fork of Nine Mile Creek.
Lake Edina is a shallow, 25-
acre lake, with a maximum
depth of 5 feet and a mean
depth of approximately
3 feet. The Lake Edina
watershed, shown in red
in the map to the left,
encompasses approximately
400 acres. Land use within
the watershed is mainly
low-density residential,
with smaller portions of
high density residential,
commercial, institutional
(Cornelia Elementary
School), and park.
AN IN-DEPTH LOOK
= FLOW DIRECTION
2
LAKE CORNELIA WATER QUALITY CHALLENGES
Water quality in Lake Cornelia is poor, with summer-
average total phosphorus and chlorophyll a
concentrations well above the state standard for
shallow lakes. The poor water quality is primarily
due to excess phosphorus in the lake, which fuels
algal growth and decreases water clarity. The
phosphorus in Lake Cornelia comes from several
sources, including stormwater runoff from the
watershed (external source) and internal sources
such as nutrient-rich sediments and decomposition
of curly-leaf pondweed. Fish activity, specifically the
disruption caused by bottom-feeding species such
as bullhead and goldfish, may also be decreasing
water clarity. The primary sources of phosphorus to
Lake Cornelia are described further below.
Phosphorus in stormwater runoff — Stormwater runoff conveys phosphorus
from streets, lawns, and parking lots to Lake Cornelia via a series of storm drain
pipes. Computer models indicate that stormwater runoff is the major contributor
of phosphorus to Lake Cornelia, ranging from 48% - 76% of the contribution to
North Cornelia in evaluated years (2015, 2016, 2017).
Nutrient-rich sediments — Phosphorus builds up over time in lake bottom
sediments as a result of sedimentation and die-off of vegetation and algae. When
oxygen levels are low at the lake bottom (typically periodically throughout the
summer), some of the phosphorus is released from the sediment into the water
column, contributing to poor water quality conditions. Sediment cores collected
from Lake Cornelia confirmed the potential for internal phosphorus loading.
Analysis of the sediment indicates the amount of phosphorus in the sediment is
similar to other metro lakes that have poor water quality.
Curly-leaf pondweed — The invasive (i.e., non-native) aquatic plant grows
under the ice during the winter and gets an early start in the spring, often
crowding out native species. It dies back in late-June and early-July, much earlier
than native species. As the plants decay, phosphorus is released into the water
column, fueling algal production and causing oxygen depletion.
Bottom-feeding fish — Fish activity, specifically the disruption caused by
bottom-feeding species such as the bullhead and goldfish found in Lake Cornelia,
can influence phosphorus concentrations in a lake. These fish feed on decaying
plant and animal matter found at the sediment surface and transform sediment
phosphorus into phosphorus available for uptake by algae through digestion and
excretion. Bottom-feeding fish can also cause resuspension of sediments, causing
reduced water clarity and poor aquatic plant growth.
Summer average phosphorus concentrations in Lake
Cornelia (North Basin) have historically been well
above the state standard for shallow lakes.
SOURCES OF PHOSPHORUS TO LAKE CORNELIA
3
LAKE EDINA WATER QUALITY
CHALLENGES
Water quality in Lake Edina is poor, with summer-
average total phosphorus and chlorophyll a
concentrations generally not meeting the state
standard for shallow lakes. The poor water quality
is primarily due to excess phosphorus in the lake,
which fuels algal production and decreases water
clarity. Phosphorus in Lake Edina comes from several
primary sources, including stormwater runoff from the
watershed (external sources) and flows from upstream
Lake Cornelia (see pie charts below).
The invasive aquatic plants curly-leaf pondweed and
Eurasian watermilfoil are both present within the lake.
In recent years, curly-leaf pondweed was observed at
low levels in two areas on the west side of the lake.
Eurasian watermilfoil is widespread throughout the
shallow lake. The invasive plants can outcompete
native species, overtaking habitat and lowering
native plant diversity.Summer average phosphorus concentrations in
Lake Edina have historically been above the state
standard for shallow lakes.
Curly-leaf pondweed
in Lake Cornelia
Graphs showing the distribution of phosphorus
sources during 2017, one of the modeled years.
Contributions from the various sources can vary
year to year based on climatic conditions.
North Cornelia:
Model Year 2017
Lake Edina:
Model Year 2017
South Cornelia:
Model Year 2017
Direct Watershed
Upstream Lakes
Internal Loading
Curly-leaf Pondweed
Atmospheric Deposition
48%
11%
19%
19%56%
63%
35%
5%
40%
0.6%
1%
0% (upstream lakes)
0% (groundwater)
0% (groundwater)
0% (groundwater)
1%
1%
PHOSPHORUS SOURCES
4
MANAGEMENT STRATEGIES TO IMPROVE LAKE CORNELIA & LAKE EDINA
Water quality in Lake Cornelia is impacted by both external sources (stormwater runoff from the watershed)
and internal sources of phosphorus (i.e., release from lake bottom sediments and die-off/decay of curly-leaf
pondweed). Because of this, the recommended management strategy is to implement a combination of
in-lake and watershed management practices.
Water quality in Lake Edina is highly influenced by the water quality of Lake Cornelia. Accordingly, the
primary recommended management strategy is to implement the recommendations for upstream Lake
Cornelia. Opportunities to reduce phosphorus from the direct watershed to Lake Edina
should also be considered.
The following section highlights the recommended management practices.
Recommended In-lake Management Practices
Study results indicate the internal management practices described below
will result in the greatest predicted improvements in water quality throughout
the three lakes, as compared to other evaluated management activities. The
predicted improvements in summer-average phosphorus concentrations in
all three lakes as a result of these internal management practices are shown in
Figure 2 on the next page.
Alum treatment in Lake Cornelia — A whole-lake alum treatment in North
and South Lake Cornelia is recommended to bind (or immobilize) the
phosphorus in lake bottom sediments and prevent release into the water column.
Curly-leaf pondweed treatments in Lake Cornelia — Continued city-led spring herbicide
treatments in Lake Cornelia are recommended to reduce the presence of curly-leaf pondweed
and promote a healthy native aquatic plant population.
Although not directly evaluated in the modeling analysis, these other in-lake
management activities should be further considered to promote the health of the lake ecosystems:
Management of bottom-feeding fish in Lake Cornelia —
Installation of a winter oxygen injection system in North and
South Lake Cornelia is recommended for further consideration
to prevent winterkill of beneficial predator fish species and
promote a healthy, more balanced fishery. Other management
activities to reduce the bottom-feeding fish population should
also be considered after collection of additional information on
the migration and movement of these species.
Invasive plant management in Lake Edina — Treatment of
invasive curly-leaf pondweed in Lake Edina is recommended
to prevent it from further threatening the lake’s aquatic plant
community and to minimize the plant fragments conveyed to
Nine Mile Creek and downstream Normandale Lake.
Recommended Watershed Management Practices
Study results indicate that the greatest source of phosphorus
to Lake Cornelia and Lake Edina is stormwater runoff. Because
the watersheds are fully-developed, significantly reducing
the phosphorus inputs from watershed runoff is logistically
challenging and expensive. One BMP that is recommended is
installation of a spent lime/CC17 treatment chamber under
the parking lot in Rosland Park (see image at right). The
innovative spent lime/CC17 treatment chamber would
serve as a “polishing” step, diverting a portion of the
discharge from Swimming Pool Pond through the
WHAT IS SPENT LIME TREATMENT?
Using spent lime, or other calcium carbonate-
based media like crushed limestone (CC17), to
remove phosphorus from stormwater is a relatively
new and innovative approach that several local
watershed management organizations have been
experimenting with in recent years. While still
experimental, benefits of using spent lime to treat
stormwater include:
• Spent lime is considered a “waste material”
from water treatment plants and thus, a green
material with low material costs.
• Rapid chemical substitution reactions between
phosphate and carbonate lead to a high
treatment capacity.
• Unlike iron-sand filters, spent lime
performance is not affected by low oxygen
conditions.
• Spent lime material has high hydraulic
conductivity, so it can treat a large amount of
water in a small treatment area.
• Spent lime treatment can remove
both particulate and dissolved
phosphorus.
Alum treatment barge
spent lime filtration chamber to remove dissolved phosphorus before discharge to Lake Cornelia. Figure
2 shows the predicted improvements in summer-average phosphorus concentrations in all three lakes as a
result of the spent lime/CC17 treatment chamber, in combination with the alum and curly-leaf pondweed
treatments. While the incremental improvement resulting from the watershed BMP is not significant,
reducing the external phosphorus loading will also increase the longevity of the alum treatment (and
therefore the frequency of repeat treatments) and reduce future build-up of phosphorus in lake bottom
sediments.
Another potential watershed management practice for consideration is an expanded street sweeping
program. Study results show that a weekly street sweeping program can result in reliable and consistent
reductions in phosphorus removal from stormwater runoff. However, this nonstructural BMP only showed
moderate improvements in lake water quality, similar to other watershed management practices. An
expanded street sweeping program to target residential streets or streets and/or commercial parking lots
not already treated by BMPs could be considered further.
Other recommended watershed management practices that landowners can implement include rain
gardens, shoreline buffers, redirection of gutter downspouts, clean-up of grass clippings, and participation
in the adopt-a-drain program.
6
Street sweeping can help
reduce the amount of
pollutants that reach the lakes
from stormwater runoff.
Figure 2
Water moves
through the spent
lime filtration
chamber to remove
phosphorus before
discharge to Lake
Cornelia.
NORTHCORNELIA
SWIMMING
POOL POND
Prepared by Barr Engineering Co.
IMPLEMENTING COST-EFFECTIVE IMPROVEMENTS
COST EFFECTIVENESS OF POTENTIAL MANAGEMENT PRACTICES
The management practices evaluated as part of this study span a wide range of treatment scales, costs,
and effectiveness. A cost-benefit analysis was completed for the modeled management practices
to compare the cost effectiveness in terms of dollars per pound of phosphorus reduction achieved.
As shown in the bar graph below, the in-lake management practices (alum treatment and curly-
leaf pondweed management) have the lowest cost per pound of phosphorus reduction achieved,
indicating they provide the greatest overall cost effectiveness. Of the watershed management practices
evaluated, the spent lime/CC17 treatment
chamber is the most cost effective, with a
cost per pound of phosphorus reduction
considerably lower than that of weekly
street sweeping or underground stormwater
infiltration systems on commercial properties.
PLANNING-LEVEL COST ESTIMATES FOR RECOMMENDED
MANAGEMENT PRACTICES
Planning-level cost estimates for the recommended management practices are provided in the table
below. These costs are intended to assist in evaluating and comparing potential management practices
but should not be considered as absolute values. All estimated costs are presented in 2019 dollars and
include costs for engineering and project administration.
Cost comparison of various management
practices in terms of dollars per pound of
phosphorus reduction achieved to North Lake
Cornelia during the time period of
April through September.
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx ix
Lake Cornelia and Lake Edina Water Quality Study
Use Attainability Analyses for Lake Cornelia (updated from 2010)
and Lake Edina (First version)
July 2019
Contents
Report Summary .................................................................................................................................................................................... i
1.0 Introduction ............................................................................................................................................................................... 1
1.1 Purpose and Process of the UAA ............................................................................................... 1
1.2 Scope of UAA Study ................................................................................................................... 2
2.0 Shallow Lake Characteristics and Water Quality ......................................................................................................... 3
2.1 Eutrophication ........................................................................................................................... 3
2.2 Nutrients .................................................................................................................................... 3
2.2.1 Stratification Impacts on Internal Loading ....................................................................................... 4
2.2.2 pH Impacts on Internal Loading ........................................................................................................... 5
2.2.3 Organism Impacts on Internal Loading ............................................................................................. 5
2.2.4 Curly-leaf Pondweed Impacts on Internal Loading ....................................................................... 5
2.3 Climate Change Considerations ................................................................................................ 6
2.3.1 Projected Changes to the Hydrologic Cycle .................................................................................... 6
2.3.2 Projected Changes to Waterbodies (Physical and Chemical).................................................... 7
2.3.3 Projected Changes to Eutrophication ................................................................................................. 7
3.0 Identification of Goals and Expectations ........................................................................................................................ 9
3.1 NMCWD Goals for Lake Management ...................................................................................... 9
3.1.1 Water Quality Goals ................................................................................................................................... 9
3.1.2 Other Lake Health Goals ........................................................................................................................10
3.2 Lower Minnesota River Watershed TMDL Report—Draft ...................................................... 11
3.3 Natural or Background Water Quality Conditions .................................................................. 12
3.4 NMCWD Adaptive Management Approach ............................................................................ 13
4.0 Lake Basin and Watershed Characteristics ..................................................................................................................14
4.1 Lake Cornelia Basin Characteristics ......................................................................................... 14
4.1.1 North Cornelia ...........................................................................................................................................14
4.1.2 South Cornelia ...........................................................................................................................................14
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx x
4.2 Lake Edina Basin Characteristics ............................................................................................. 17
4.3 Watershed Characteristics ...................................................................................................... 20
4.4 Lake Inflows and Drainage Areas ............................................................................................ 22
4.4.1 Natural Conveyance Systems ...............................................................................................................22
4.4.2 Stormwater Conveyance Systems ......................................................................................................22
4.4.3 Southdale Center Cooling System Discharge ................................................................................22
5.0 Existing Water Quality ..........................................................................................................................................................25
5.1 Water Quality .......................................................................................................................... 25
5.1.1 Phosphorus, Chlorophyll a, and Clarity ............................................................................................25
5.1.2 Chlorides ......................................................................................................................................................30
5.2 Sediment Quality ..................................................................................................................... 30
5.2.1 Lake Cornelia (North and South Basin) ............................................................................................30
5.2.2 Lake Edina ....................................................................................................................................................30
5.3 Aquatic Communities .............................................................................................................. 32
5.3.1 Lake Cornelia ..............................................................................................................................................32
5.3.2 Lake Edina ....................................................................................................................................................44
6.0 Water Quality Modeling for the UAA ............................................................................................................................51
6.1 P8 Model Runoff and Phosphorus Loading ............................................................................. 51
6.2 Water Balance Calibration....................................................................................................... 51
6.2.1 Precipitation and Runoff ........................................................................................................................51
6.2.2 Stormwater Volume Calibration (Water Balance) ........................................................................53
6.3 In-Lake Phosphorus Modeling ................................................................................................. 55
6.3.1 Atmospheric Deposition ........................................................................................................................56
6.3.2 Settling and Copper Sulfate .................................................................................................................56
6.3.3 Internal Sediment Loading and Benthivorous Fish .....................................................................56
6.3.4 Curlyleaf Pondweed Die Off and Decay (Lake Cornelia Only) ................................................57
6.3.5 In-Lake Water Quality (Phosphorus) Model Calibration ...........................................................58
6.3.6 In-Lake Water Quality (Phosphorus) Model Calibration Loading Summaries ..................64
6.4 Modeling Chlorophyll a and Secchi Disc Transparency ........................................................... 69
7.0 Evaluation of Management Strategies ..........................................................................................................................73
7.1 Watershed Management Strategies/Scenarios ...................................................................... 73
7.1.1 Infiltration BMPs on Commercial Properties ..................................................................................75
7.1.2 Filtration BMPs on Commercial Properties .....................................................................................77
7.1.3 Spent Lime/CC17 Treatment Chamber ............................................................................................77
7.1.4 Weekly Street Sweeping ........................................................................................................................81
7.1.5 Other Watershed Management Strategies.....................................................................................83
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xi
7.2 Internal Load Reductions......................................................................................................... 83
7.2.1 Curly-leaf Pondweed Management...................................................................................................84
7.2.2 Alum Treatment of Lake Sediments ..................................................................................................84
7.3 Other Lake Management Strategies ....................................................................................... 85
7.3.1 Carp and Goldfish Tracking and Benthivorous (Bottom-feeding) Fish Management ...85
7.3.2 Lake Edina Aquatic Plant Management ...........................................................................................86
8.0 Lake Response to Management Strategies .................................................................................................................89
8.1 Lake Response to Watershed (External) Management Strategies ......................................... 89
8.1.1 Changes in In-lake Phosphorus Concentrations ..........................................................................89
8.1.2 Summer Average Total Phosphorus Concentrations .................................................................90
8.1.3 Phosphorus Loading Reductions ........................................................................................................92
8.1.4 Watershed BMP-specific Results ........................................................................................................94
8.2 Lake Response to Internal Loading Management ................................................................... 96
8.2.1 Changes in In-lake Phosphorus Concentrations ..........................................................................96
8.2.2 Summer Average Total Phosphorus Concentrations .................................................................99
8.2.3 Phosphorus Loading Reductions ..................................................................................................... 101
8.3 Lake Responses to Combined Internal and External Management ...................................... 104
8.3.1 Changes in In-lake Phosphorus Concentrations ....................................................................... 104
8.3.2 Summer Average Total Phosphorus Concentrations .............................................................. 104
8.3.3 Phosphorus Loading Reductions ..................................................................................................... 109
8.4 Management Alternatives Summary .................................................................................... 112
9.0 Cost-Benefit of Management Efforts .......................................................................................................................... 116
9.1 Opinions of Probable Cost for Modeled Scenarios ............................................................... 116
9.1.1 Cost Details for Modeled Scenarios ............................................................................................... 117
9.2 Cost-Benefit Analysis for Modeled Scenarios ....................................................................... 118
9.3 Opinions of Probable Cost for Other Evaluated Strategies ................................................... 123
9.3.1 Winter aeration of Lake Cornelia using direct oxygen injection ......................................... 123
9.3.2 Lake Edina Aquatic Plant Management ........................................................................................ 123
10.0 Conclusions and Recommendations ........................................................................................................................... 124
10.1 Phosphorus Sources .............................................................................................................. 124
10.2 Management Strategies ........................................................................................................ 124
10.2.1 In-lake Phosphorus Management ................................................................................................... 125
10.2.2 External (Watershed) Phosphorus Management ...................................................................... 125
10.2.3 Responses in Chlorophyll a Concentrations and Water Clarity ........................................... 126
10.2.4 Recommended Management Practices ........................................................................................ 128
11.0 References ............................................................................................................................................................................ 134
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xii
List of Tables
Table 3-1 NMCWD water quality goals for shallow lakes ........................................................................................ 10
Table 3-2 NMCWD holistic lake health assessment evaluation factors .............................................................. 11
Table 3-3 Growing season total phosphorus load reductions summary from the Draft Lower
Minnesota River Watershed TMDL Report ................................................................................................. 12
Table 4-1 Stage-storage-discharge relationships for Lake Cornelia .................................................................... 15
Table 4-2 Stage-storage-discharge relationship for Lake Edina ............................................................................ 17
Table 4-3 Land use classifications in the Lake Cornelia and Lake Edina watersheds .................................... 20
Table 5-1 Maximum potential internal loading rate for North and South Cornelia and Lake Edina
compared to other Twin Cities Metro Area lakes. ................................................................................... 31
Table 5-2 Lake Cornelia algal copper sulfate treatments ......................................................................................... 37
Table 5-3 South Cornelia fyke net results showing the number of fish caught by size ............................... 43
Table 5-4 North Cornelia fyke net results showing the number of fish caught by size ............................... 44
Table 5-5 Lake Edina algal copper sulfate treatments ............................................................................................... 47
Table 6-1 Precipitation amounts for 2015, 2016, and 2017 ..................................................................................... 52
Table 6-2 Water balance summary of watershed runoff inflows and discharges ........................................... 55
Table 7-1 Infiltration BMP treatment volumes and infiltration rates ................................................................... 76
Table 7-2 Summary of alum and sodium aluminate application doses for North and South Cornelia . 85
Table 8-1 Total phosphorus (TP) load reductions resulting from watershed management efforts ........ 93
Table 8-2 Total phosphorus (TP) load reductions resulting from internal loading management
efforts ...................................................................................................................................................................... 103
Table 8-3 Comparison of total phosphorus summer average concentrations under existing
conditions to combined management (internal and commercial infiltration BMPs)
conditions .............................................................................................................................................................. 106
Table 8-4 Comparison of total phosphorus summer average concentrations under existing
conditions to combined management (internal and spent lime/CC17 treatment chamber)
conditions .............................................................................................................................................................. 108
Table 8-5 Total phosphorus load reductions summary for combined management (internal and
external) conditions ........................................................................................................................................... 111
Table 8-6 Comparison of total phosphorus summer average concentrations for all modeled
management scenarios .................................................................................................................................... 114
Table 8-7 Comparison of total phosphorus loads for all modeled management scenarios .................... 115
Table 9-1 Planning-level cost estimates for modeled management alternatives ......................................... 116
Table 9-2 Cost-benefit summaries for North Cornelia, South Cornelia, and Lake Edina for modeled
management alternatives................................................................................................................................ 120
Table 10-1 Planning-level cost estimates for recommended management alternatives ............................. 129
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xiii
List of Figures
Figure 2-1 Generalized thermal lake stratification diagram ......................................................................................... 4
Figure 3-1 NMCWD Holistic Lake Health Assessment Factors (NMCWD, 2017, amended 2019) ................ 9
Figure 4-1 Lake Cornelia Bathymetry ................................................................................................................................. 16
Figure 4-2 Lake Edina Bathymetry ....................................................................................................................................... 19
Figure 4-3 Land Use Lake Cornelia and Lake Edina Watersheds ............................................................................. 21
Figure 4-4 Lake Cornelia Stormwater Conveyance ....................................................................................................... 23
Figure 4-5 Lake Edina Stormwater Conveyance ............................................................................................................. 24
Figure 5-1 Summer total phosphorus and chlorophyll a concentrations in North Cornelia from 2004
through 2017. The red crosses indicate the average summer (June through September)
concentrations. ...................................................................................................................................................... 27
Figure 5-2 Summer total phosphorus and chlorophyll a concentrations in South Cornelia from 2004
through 2017. The red crosses indicates the average summer (June through September)
concentrations. ...................................................................................................................................................... 27
Figure 5-3 Summer Secchi disc depth readings in (a) North and (b) South Cornelia from 2004
through 2017. The red crosses indicate the average of summer (June through September)
readings. ................................................................................................................................................................... 28
Figure 5-4 Summer total phosphorus and chlorophyll a concentrations in Lake Edina from 2004
through 2017. The red crosses indicate the average summer (June through September)
concentrations. ...................................................................................................................................................... 28
Figure 5-5 Summer Secchi disc depth readings in Lake Edina from 2004 through 2017. The red
crosses indicate the average of summer (June through September) readings. .......................... 29
Figure 5-6 North Cornelia Macrophyte Species Richness Compared with Plant IBI Threshold for
Species Richness ................................................................................................................................................... 33
Figure 5-7 North Cornelia Floristic Quality Index (FQI) Compared with Plant IBI Threshold for FQI ........ 33
Figure 5-8 South Cornelia Macrophyte Species Richness Compared with Plant IBI Threshold for
Species Richness ................................................................................................................................................... 34
Figure 5-9 South Cornelia Floristic Quality Index (FQI) Compared with Plant IBI Threshold for FQI ........ 34
Figure 5-10 Curly-leaf pondweed growth observed in Lake Cornelia in 2017. .................................................... 36
Figure 5-11 North Cornelia Phytoplankton Data Summary (2004-2017) ............................................................... 37
Figure 5-12 South Cornelia Phytoplankton Data Summary (2004-2017) ............................................................... 38
Figure 5-13 North Cornelia blue-green algae data compared with the World Health Organization’s
Risk of Adverse Health Effects Guidelines .................................................................................................. 39
Figure 5-14 South Cornelia blue-green algae data compared with the World Health Organization’s
Risk of Adverse Health Effects Guidelines .................................................................................................. 39
Figure 5-15 North Cornelia Zooplankton Data Summary (2008-2015)................................................................... 41
Figure 5-16 South Cornelia Zooplankton Data Summary (2008-2015)................................................................... 42
Figure 5-17 Lake Edina Species Richness Compared with Plant IBI Threshold for Species Richness .......... 45
Figure 5-18 Lake Edina Floristic Quality Index (FQI) Compared with Plant IBI Threshold for FQI ................ 45
Figure 5-19 Lake Edina Phytoplankton Data Summary (2008-2017) ....................................................................... 47
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xiv
Figure 5-20 Lake Edina Blue-green Algae Data Comparison with World Health Organization Risks
Guidelines ................................................................................................................................................................ 48
Figure 5-21 Lake Edina Zooplankton Data Summary (2008-2017) ........................................................................... 49
Figure 5-22 Lake Edina average annual cladoceran and blue-green algae numbers comparison .............. 50
Figure 6-1 Photo taken of Lake Edina low water levels in April 2015 ................................................................... 53
Figure 6-2 North Cornelia (2015) Water Balance ........................................................................................................... 54
Figure 6-3 North Cornelia In-Lake Calibration Model for 2015 ............................................................................... 60
Figure 6-4 North Cornelia In-Lake Calibration Model for 2016 ............................................................................... 60
Figure 6-5 North Cornelia In-Lake Calibration Model for 2017 ............................................................................... 61
Figure 6-6 South Cornelia In-Lake Calibration Model for 2015 ............................................................................... 61
Figure 6-7 South Cornelia In-Lake Calibration Model 2016 ...................................................................................... 62
Figure 6-8 South Cornelia In-Lake Calibration Model 2017 ...................................................................................... 62
Figure 6-9 Lake Edina In-Lake Calibration Model 2015 .............................................................................................. 63
Figure 6-10 Lake Edina In-Lake Model 2016 ...................................................................................................................... 63
Figure 6-11 Lake Edina In-Lake Calibration Model 2017 .............................................................................................. 64
Figure 6-12 Loading Summaries from North Cornelia In-Lake Calibration Models........................................... 66
Figure 6-13 Loading Summaries from South Cornelia In-Lake Calibration Models ........................................... 67
Figure 6-14 Loading Summaries from Lake Edina In-Lake Calibration Models ................................................... 68
Figure 6-15 North Cornelia relationships between total phosphorus, chlorophyll a, and Secchi Disc
Transparency .......................................................................................................................................................... 70
Figure 6-16 South Cornelia relationships between total phosphorus, chlorophyll a, and Secchi Disc
Transparency .......................................................................................................................................................... 71
Figure 6-17 Lake Edina relationships between total phosphorus, chlorophyll a, and Secchi Disc
Transparency .......................................................................................................................................................... 72
Figure 7-1 Conceptual design of the double-chamber spent lime/CC17 treatment cell .............................. 80
Figure 8-1 In-lake phosphorus concentrations that resulted from watershed management efforts in
North Cornelia in 2015 ....................................................................................................................................... 90
Figure 8-2 North Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for
Watershed Management Efforts .................................................................................................................... 91
Figure 8-3 South Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for
Watershed Management Efforts .................................................................................................................... 91
Figure 8-4 Lake Edina In-Lake Summer Average Phosphorus Concentration Summary for Watershed
Management Efforts ............................................................................................................................................ 92
Figure 8-5 In-Lake Phosphorus Concentration Changes that resulted from internal management
efforts in North Cornelia, South Cornelia, and Lake Edina in 2017 .................................................. 98
Figure 8-6 North Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for
Internal Management Efforts ......................................................................................................................... 100
Figure 8-7 South Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for
Internal Management Efforts ......................................................................................................................... 100
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xv
Figure 8-8 Lake Edina In-Lake Summer Average Phosphorus Concentration Summary for Internal
Management Efforts .......................................................................................................................................... 101
Figure 8-9 Remaining Total Phosphorus Load to North Cornelia in 2017 with various combinations of
internal and external commercial infiltration management .............................................................. 109
Figure 8-10 Remaining Total Phosphorus Load to North Cornelia in 2017 with various combinations of
internal and external spent lime/CC17 management .......................................................................... 110
Figure 9-1 Annualized cost per unit reduction (µg/L) in summer average total phosphorus
concentration for individual management practices ........................................................................... 121
Figure 9-2 Annualized cost per unit reduction (µg/L) in summer average total phosphorus
concentration for combined management practices ........................................................................... 122
Figure 10-1 Comparison of summer average total phosphorus concentrations (µg/L) for
recommended management alternatives ................................................................................................ 126
Figure 10-2 Summer average chlorophyll a concentrations (µg/L) for recommended management
alternatives ............................................................................................................................................................ 127
Figure 10-3 Summer average Secchi disc depths (m) for recommended management alternatives ....... 127
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xvi
List of Appendices, Attachments, or Exhibits
Appendix A Existing Pond Information—Lake Cornelia and Lake Edina
Appendix B Lake Cornelia System Fisheries Assessment (2018)
Appendix C In-Lake Model Water Balance Results
Appendix D External Loading Management Concentration Plots
Appendix E Internal Loading Management Concentration Plots
Appendix F Combined (Internal + External) Management Concentration Plots
Appendix G Combined (Internal + External) Management Loading Bar Plots
Appendix H Opinions of Probable Cost
P:\Mpls\23 MN\27\2327634\WorkFiles\Lake Cornelia\2018 UAA\2018 UAA Report\Lake Cornelia_LakeEdina_UAA_Final_July 2019.docx xvii
Acronyms
Acronym Description
AACE Association for the Advancement of Cost Engineering
AIS Aquatic Invasive Species
BMP Best Management Practice
Chl a Chlorophyll a
EWM Eurasian watermilfoil
FiN Fishing in the Neighborhood
FQI Floristic Quality Index
HSG Hydrologic Soil Groups
MDNR Minnesota Department of Natural Resources
MnLEAP Minnesota Lake Eutrophication Analysis Procedure
MPCA Minnesota Pollution Control Agency
NCHF North Central Hardwood Forests
NMCWD Nine Mile Creek Watershed District
RPBCWD Riley Purgatory Bluff Creek Watershed District
SSURGO Soil Survey Geographical Database
TMDL Total Maximum Daily Load
TP Total Phosphorus
TSS Total Suspended Solids
UAA Use Attainability Analysis
USGS U.S. Geological Survey
WHO World Health Organization
WiLMS Wisconsin Lake Modeling Suite
1
1.0 Introduction
This report describes the results of the Use Attainability Analysis (UAA) for Lake Cornelia and Lake Edina in
Edina, Minnesota. Both lakes are summarized in this report due to the significant influence of Lake
Cornelia on Lake Edina’s water quality. An UAA provides the scientific foundation for a lake-specific best
management plan that will permit maintenance of, or attainment of, the intended beneficial uses of a
waterbody. The UAA is a scientific assessment of a water body’s physical, chemical, and biological
condition. This study includes both a water quality assessment and prescription of protective and/or
remedial measures for Lake Cornelia and Lake Edina and their tributary watersheds. The work presented in
this report follows analyses that were previously completed for an UAA developed for North Cornelia in
2010.
The conclusions and recommendations presented in this report are based on historical water quality data,
a fisheries survey conducted in 2018, several years of aquatic plant surveys, and the results of intensive
lake water quality monitoring in 2015, 2016, and 2017. Lake models were developed and calibrated to the
2015, 2016, and 2017 data sets to gain a better understanding of the influence of various phosphorus
sources on lake water quality. The models were also used to quantify the expected outcome of different
management actions on lake water quality. Water quality goals for the lakes were identified based on the
lakes’ designated beneficial uses (e.g., runoff management, curly-leaf pondweed control). In addition, best
management practices (BMPs), including both external and internal phosphorus controls, were evaluated
to compare their relative effect on total phosphorus concentrations in the lakes. Management scenarios
were then assessed to determine attainment or non-attainment of the lake goals.
1.1 Purpose and Process of the UAA
The Nine Mile Creek Watershed District (NMCWD) has historically used a process referred to as Use
Attainability Analysis (UAA) to assess the water quality condition of its lakes relative to the desired
beneficial uses that can be reasonably achieved and maintained with implementation of management
recommendations. The UAA process addresses a wide range of goals (e.g., water quantity, aquatic
communities, recreational use, and wildlife), with the primary focus being achievement of water quality
goals. As part of the Nine Mile Creek Watershed District Water Management Plan (Plan) adopted in 2017
and amended in 2018 and 2019 (NMCWD, 2017, amended 2019), the NMCWD has expanded its emphasis
on the role of ecological indicators (aquatic plants, phytoplankton, fish, etc.) in overall lake health, as well
as the feedback mechanisms between these indicators. A properly functioning ecosystem supports the
attainment of good water quality. The NMCWD has also adopted the Minnesota eutrophication standards
as part of their 2017 Plan.
The UAA employs a watershed runoff model and an in-lake water quality model to quantify the benefits
of management efforts. The in-lake water quality model predicts changes in lake water quality based on
the results of the watershed runoff model (external inputs) as well as internal processes such as sediment
phosphorus release due to anoxia and bioturbation (carp/goldfish), curly-leaf pondweed death and decay,
and enhanced phosphorus settling rates due to phytoplankton bloom treatments. Using these models,
various watershed and lake management strategies can be evaluated to determine their likely effects on
2
lake water quality. The resulting lake water quality can then be compared with the water quality goals to
see if the management strategies are able to produce the desired changes in the lake. Using the tools of
the UAA, the cost-effectiveness of the management strategies can also be evaluated.
1.2 Scope of UAA Study
This UAA evaluates current and various proposed conditions for Lake Cornelia and Lake Edina. Several
steps are necessary for the evaluation of the watersheds, lakes, and management initiatives. Those steps,
briefly summarized below, are described in detail in the following sections.
Identification of Goals and Expectations- To evaluate lake management strategies, it is first necessary
to establish the criteria against which outcomes can be measured.
Assessment of Current Conditions- The conditions of Lake Cornelia’s and Lake Edina’s watershed,
biological communities, water quality, and in-lake response to nutrient (phosphorus) inputs were
evaluated for this study. Sources of phosphorus to Lake Cornelia and Lake Edina were identified and
quantified through modeling analyses.
Evaluation of Management Strategies- A variety of watershed (external) loading and internal loading
reduction scenarios were evaluated. The in-lake model was used to predict the lakes’ responses to these
changes. Costs of the management strategies were estimated so that those costs could be compared to
the in-lake benefits that the management initiatives are expected to provide.
3
2.0 Shallow Lake Characteristics and Water Quality
Lake Cornelia and Lake Edina can both be classified as shallow lake ecosystems. Shallow lakes are lakes
that generally have well mixed water columns throughout most of the year and have depths that allow for
light penetration to reach the entire sediment surface (i.e., potential for macrophyte growth over the
entire lake). Shallow lakes classically exist in two states: (1) clear water with submerged and emergent
macrophytes; and (2) turbid water with phytoplankton. The concentration of nutrients entering the
shallow water system, the biovolumes of benthivorous fish per unit lake area, and the presence or absence
of invasive species such as curly-leaf pondweed are primary drivers that determine the state of shallow
lakes.
There are a number of concepts and terminology that are necessary to describe and evaluate a lake’s
water quality. This section is a brief discussion of those concepts.
2.1 Eutrophication
Eutrophication, or lake degradation, is the accumulation of sediments and nutrients in lakes. As a lake
naturally becomes more fertile, biological production enhances and sediment inflow accumulates filling
the lake’s basin. Over a period of hundreds to thousands of years, a lake can successively become a pond,
a marsh and, ultimately, a terrestrial site. This process of eutrophication is natural and results from the
normal environmental forces that influence a lake. Cultural eutrophication, however, is an acceleration of
the natural processes and is caused by human activities. Nutrient and sediment inputs from wastewater
treatment plants, septic tanks, agriculture, and stormwater runoff can far exceed the natural inputs to the
lake. Nutrient enrichment in lakes often intensifies primary production resulting in the manifestation of
algal blooms. Enhanced sediment loadings can attenuate light and reduce lake transparency, which can
limit macrophyte growth. Since macrophytes assist in creating a stable water state, especially in shallow
lakes, high suspended sediment and enhanced nutrients can often lead to impaired water quality.
2.2 Nutrients
Biological production in an aquatic ecosystem is limited by the concentrations of essential nutrients. The
“limiting nutrient” concept is a widely applied principle in ecology and in the study of eutrophication. It is
based on the idea that phytoplankton and plants require many nutrients to grow, but the nutrient with
the lowest availability, relative to the amount needed by the phytoplankton or plant, will limit growth. It
follows then, that identifying the limiting nutrient will point the way to controlling aquatic plant and algal
growth. Nitrogen (N) and phosphorus (P) are generally the two growth-limiting macronutrients in most
natural waters. Thus, efforts to improve water quality typically focus on reducing the growth-limiting
nutrient concentration in the waterbody; however, it is often difficult to identify and control all of the
nutrient loadings to a specific waterbody.
Two primary sources, external and internal loads, are responsible for elevated nutrient concentrations in
lakes. Nutrients that enter lakes through watershed runoff, groundwater inputs, or atmospheric deposition
are considered external loads. As urbanization has occurred, more areas of impermeable surfaces have
been developed causing increased stormwater runoff and pollutant transport during storm and spring
4
thaw events. In urbanized areas, stormwater runoff typically flows through storm sewer systems to the
downstream waterbody, which generally results in faster velocities than natural channel flow and can
result in higher suspended loadings. Implementation of the NMCWD’s stormwater management rules for
new development and redevelopment and efforts to install retrofit best management practices (BMPs) are
helping to reduce external loads to nearby waterbodies. However, for many shallow lakes, internal load
reduction measures (e.g., alum treatment, aquatic plant management, fish management) are also required
to meet water quality goals.
Once external nutrient loads enter a lake, over time, the nutrients accumulate in the sediment through the
settling of particulates and through organism decay. Natural lake processes such as sediment
resuspension, chemical dissolution, or microbial reduction can reintroduce these nutrients to the overlying
water body resulting in internal loading. This is specifically common for phosphorus, which can be found
bound to the sediment under oxidized conditions. The binding of phosphorus to iron in sediments allows
the sediment to act as a sink or source depending on the lake’s physical and chemical conditions.
Therefore, understanding the chemical and physical conditions and the timing of these conditions will be
important considerations when developing an internal loading management plan.
2.2.1 Stratification Impacts on Internal Loading
Lake stratification, the separating of an upper, well mixed warm layer (epilimnion) from a cool, bottom
layer (hypolimnion) (Figure 2-1), can lead to low oxygen concentrations in lake bottom waters and trigger
internal phosphorus loading. For shallow lakes like Lake Cornelia and Lake Edina, stratification is typically
irregular and can happen on a daily, weekly, or longer timescale. Mixing likely occurs regularly in Lake
Cornelia and phosphorus released from sediments is then made available to phytoplankton during these
frequent mixing events. Because Lake Edina is very shallow, mixing likely occurs daily and this may help to
prevent internal loading.
Figure 2-1 Generalized thermal lake stratification diagram
5
2.2.2 pH Impacts on Internal Loading
The pH of the water column can also play a vital role in affecting the phosphorus release rate under
conditions when oxygen is present in the water column (oxic conditions). Photosynthesis by macrophytes
and algae during the day tend to raise the pH in the water column, which can enhance the phosphorus
release rate from the oxic sediment. Enhancement of phosphorus release at elevated pH (pH > 7.5) is
thought to occur through replacement of the phosphate ion (PO4-3) with the excess hydroxyl ion (OH-) on
the oxidized iron compound (James, et al., 2001). Large increases in pH are often the consequence of
phytoplankton blooms.
2.2.3 Organism Impacts on Internal Loading
Benthivorous fish, such as carp, bullhead and goldfish, can have a direct influence on the phosphorus
concentration in a lake (LaMarra, 1975). These fish typically feed on decaying plant and animal matter and
other organic particulates found at the sediment surface. The fish digest the organic matter, and excrete
soluble nutrients, thereby transforming sediment phosphorus into soluble phosphorus available for
uptake by algae at the lake surface. Benthivorous (bottom-feeding) fish can also cause resuspension of
sediments in shallow ponds and lakes, transporting phosphorus from sediment into the water column,
causing reduced water clarity and poor aquatic plant growth, as well as high phosphorus concentrations
(Cooke, Welch, Peterson, & Newroth, 1993). In some cases, the water quality impairment caused by
benthivorous fish can negate the positive effects of BMPs and lake restoration.
The critical difference between biological (e.g., benthivorous fish feeding) and physical (e.g., wind and
waves) sediment resuspension is the area and the frequency to which these components can induce
impacts. The volume of sediment impacted by physical resuspension is largely influenced by the geometry
of the lake (e.g., size, fetch, bathymetry) and wind events (e.g., direction, velocity). For example, a wind
event may develop wave induced sediment resuspension along a portion of the shoreline. However,
biological resuspension from feeding or mating activities can occur over a much larger area and is
impacted by the number of organisms in the aquatic ecosystem. Additionally, while physical resuspension
occurs in a periodic, episodic-based fashion, benthivorous fish resuspension can be more continuous.
2.2.4 Curly-leaf Pondweed Impacts on Internal Loading
Another potential source of internal phosphorus loading is the die-off of curly-leaf pondweed. Curly-leaf
pondweed is an invasive (i.e., non-native) aquatic plant that is common in many of the lakes in the Twin
Cities metropolitan area. Curly-leaf pondweed grows under the ice during the winter and gets an early
start in the spring, crowding out native species. It releases a small reproductive pod that resembles a small
pine cone in late-June, and then begins its die-back in late-June and early-July. The biomass sinks to the
bottom of the lake and begins to decay, releasing phosphorus into the water column and causing oxygen
depletion, exacerbating the internal sediment release of phosphorus. This cycle typically results in an
increase in phosphorus concentrations in the lake in late-June of early-July.
6
2.3 Climate Change Considerations
Considerable studies have been devoted to predicting the impacts of a warming climate on the
hydrologic cycle. Of particular concern are the changes to atmospheric moisture content, evaporation,
precipitation intensity, and the possibility of increased risk for drought and flooding extremes (Trenberth,
1999; Trenberth, Smith, Qian, Dai, & Fasullo, 2003; Giorgi, et al., 2011; Trenberth, 2011).
Alterations to the hydrologic cycle will consequently impact freshwater ecosystems. Observational records
and climate model projections show evidence of freshwater vulnerability to a warming climate (Dokulil &
Teubner, Eutrophication and climate change: Present situation and future scenarios, 2011). Freshwater
characteristics such as lake stratification and mixing, ice coverage, and river flow could see discernable
changes by the end of the 21st century (Dokulil & Teubner, 2011; Dokulil, 2013). Increases in nutrient
loadings and water temperatures, changes to water levels, and amplified eutrophication could impact
aquatic organisms and influence biodiversity.
2.3.1 Projected Changes to the Hydrologic Cycle
Larger concentrations of greenhouse gases in the atmosphere, such as carbon dioxide and methane,
create an increased downwelling of longwave radiation to the earth’s surface (Trenberth, 1999). This
enhanced downwelling not only escalates surface temperature warming, but also induces changes to the
atmospheric moisture content and evaporation. Higher atmospheric temperatures allow for an expanded
water holding-capacity of the atmosphere and enhanced radiation causes elevated rates of evaporation.
This results in increases to the atmospheric moisture content, which, consequently, will impact
precipitation (Trenberth, 1999; Trenberth, Smith, Qian, Dai, & Fasullo, 2003; Kharin, Zwiers, Zhang, &
Wehner, 2013).
While changes to precipitation amounts and intensity are expected on a global scale, the changes will be
geographically disproportionate. Shifts in the natural modes of atmospheric circulation have been
documented. The increase in evaporation and atmospheric moisture content causes more moisture to be
transported from divergence regions (i.e., subtropics) to convergence zones (i.e., tropics and
mid-latitudes). This causes wet areas to become wetter and dry areas to become drier (Trenberth, 1999).
In recent decades annual average precipitation has risen in the Northeast, Great Plains, Pacific Northwest,
and Alaska, while decreases have been observed for parts of the Southwest United States and in Hawaii
(Cayan, 2013; Walsh, 2014; Dettinger, Udall, & Georgakakos, 2015).
According to the National Oceanic and Atmospheric Administration’s (NOAA’s) 2013 assessment of
climate trends for the Midwest (NOAA, 2013), upward trends in annual and summer precipitation amounts
have been observed. The frequency of higher intensity storms have also been noted. Specifically in
Minnesota, climatologists have identified four significant climate trends (MDNR, 2017):
• Increasing annual precipitation
• Increasing frequency and size of extreme rainfall events
• Increasing temperatures, with winter temperatures warming the fastest
• Decline in severity and frequency of extreme cold weather
7
Overall, the changes to precipitation induced by atmospheric warming pose difficult challenges. The shift
to more frequent, high intensity precipitation events in Minnesota indicates a risk for extreme flood
events. Higher intensity precipitation events typically produce more runoff than lower intensity events
with similar amounts of precipitation because higher intensity rainfall can overwhelm the capacity of the
land surface to infiltrate and attenuate runoff.
Not only do these hydrologic changes pose challenges for agriculture, infrastructure, and human safety;
but also has the potential to induce changes to aquatic environments. The subsequent section describes
the anticipated impacts to aquatic ecosystems if atmospheric warming trends continue.
2.3.2 Projected Changes to Waterbodies (Physical and Chemical)
In freshwater lakes, one of the most important atmospheric variables influencing the lake’s physical and
chemical parameters is temperature. Due to enhanced air temperatures and the projected increasing
trends, lake water temperature and the number of ice free days are projected to change in most inland
waters globally. Increases in lake temperature will affect mixing regimes, the length and depth of summer
stratification in deep lakes, and the oxygen concentration in the hypolimnion (Dokulil, 2013; Dokulil, 2014;
Dokulil, 2016). As water temperature rises, lake stability enhances, which results in longer thermal
stratification and shorter mixing periods (Dokulil, 2013). Resistance to mixing between the nutrient rich
hypolimnion and nutrient poor epilimnion across the thermocline increases considerably at temperature
gradients of only a few degrees Fahrenheit (Sahoo, et al., 2016).
Prolonged lake stability and a lower thermocline enhances the risk of oxygen depletion in the
hypolimnion (Jeppesen, et al., 2009; Sahoo, et al., 2016). Anoxic conditions in the hypolimnion can cause
nutrient release from the sediments raising the potential for algal blooms. Additionally, overall oxygen
concentrations in the lake will be reduced as solubility decreases when the water temperature warms
(Dokulil & Teubner, 2011).
In the tropics and mid-latitudes where precipitation is likely to increase, with the heighted chance for
extreme events, other concerns are warranted. Intense rainfalls resulting in flooding could raise the
loading of suspended sediments associated with larger areas experiencing soil erosion (Dokulil &
Teubner, 2011; Dokulil, 2016). The combination of longer dry periods and extreme precipitation events
could create episodic and intense pulse flows affecting aquatic habitats, bank stability, and species
(Dokulil, 2016). Additionally, the increase in the number of extreme, high intensity rain events is likely to
increase the runoff driven phosphorus transfers from the land to the water (Jeppesen, et al., 2009).
2.3.3 Projected Changes to Eutrophication
The potential for increased erosion and nutrient inputs from large runoff rates combined with higher
water temperatures and prolonged lake stratification in summer could lead to widespread, climate-related
eutrophication based on the results of existing studies (Dokulil & Teubner, 2011; Dokulil, 2013). Nutrient
enrichment, whether through external or internal loading, stimulates the development of phytoplankton
biomass. This resulting surface biomass absorbs light, can shade out benthic algae or macrophytes, and
can produce negative lake aesthetics (Dokulil & Teubner, 2011). Unfortunately, not only has previous
8
research projected larger biomasses of phytoplankton in a warmer climate, but research also predicts that
a higher proportion of these phytoplankton biomasses will consist of potentially toxic cyanobacteria
assemblages (Jeppesen, et al., 2009; Dokulil & Teubner, 2011; Jeppesen, et al., 2014; Dokulil, 2016).
Multiple regression analyses on data from 250 Danish lakes sampled during the month of August
indicated higher dominance of cyanobacteria with a warming climate. Studies during heat waves in the
northern hemisphere also showed that higher percentages of cyanobacteria correlated with rises in
temperature (Huisman, Matthijs, & Visser, 2005).
Changes in the seasonal pattern and dynamics of freshwater productivity could also be a consequence of
a changing climate. With the earlier onset of warmer air temperatures in the spring, the timing of the
phytoplankton peak is likely to shift forward. If the phytoplankton blooms contain a larger percentage of
cyanobacteria species or if the timing of algal production falls out of synchrony with the food demands of
zooplankton and fish, then upper levels of the food chain could be negatively impacted (Dokulil, 2016).
Enhanced phytoplankton biomasses can also induce thermal feedback mechanisms for lakes. The thermal
structure of lakes can be influenced by phytoplankton via light attenuation. The area of biomass at the
surface of a lake affects vertical short-wave radiation. Thus, a large area of phytoplankton biomass can
result in greater surface temperatures and stronger stratification by influencing the temperature gradient
with depth (Dokulil, 2013). Additionally, increased light attenuation at the surface will reduce light
availability at the lake bottom influencing macrophyte growth (Jeppesen, et al., 2014).
This UAA study did not directly assess potential impacts to lake responses due to a changing climate.
However, any current and/or future management efforts for waterbodies will be affected by changing
climate conditions. Continued monitoring of lake conditions will be important as management efforts are
implemented and as changing climate conditions progress. Long-term studies of waterbodies will be
essential in order to create the most effective plans to overcome climate-induced impacts.
9
3.0 Identification of Goals and Expectations
3.1 NMCWD Goals for Lake Management
The NMCWD’s approach to assessing and improving lake health is illustrated in Figure 3-1. The primary
factors identified as affecting lake ecological health include chemical water quality (e.g., nutrient
concentrations), aquatic communities, and water quantity (groundwater and surface water). The effects of
recreation and wildlife habitat on overall lake health are also considered.
Figure 3-1 NMCWD Holistic Lake Health Assessment Factors (NMCWD, 2017, amended 2019)
3.1.1 Water Quality Goals
One of the primary goals of the District is to “ensure the water quality of the lakes and streams of the
NMCWD is protected and enhanced.” In 1996, the NMCWD established lake water quality management
goals based on designated uses for a waterbody (i.e., full-contact recreational activities such as swimming;
non-full body contact recreational activities such as boating, canoeing, or water skiing; fishing and
aesthetic viewing; runoff management). In 2008, the MPCA adopted eutrophication water quality
standards for Minnesota lakes, which vary by ecoregion and include criteria for both shallow and deep
lakes. The MPCA defines “shallow” lakes as having a maximum depth of 15 feet or less or having at least
80% of the lake area shallow enough to support aquatic plants (referred to as “littoral area”).
10
In their 2017 Plan, the NMCWD adopted the state’s lake eutrophication standards as their lake water
quality goals, as well as the state water quality standards for Escherichia coli and chloride. The water
quality goals for shallow lakes (including Lake Cornelia and Lake Edina) are presented in Table 3-1.
Table 3-1 NMCWD water quality goals for shallow lakes
Water Quality Parameter Water Quality Standard for Shallow Lakes1, 2
Total Phosphorus (summer average, μg/L) 60
Chlorophyll a (summer average, μg/L) 20
Secchi Disc Transparency (summer average, m) 1.0
Total Suspended Solids (mg/L) NA
Daily Dissolved Oxygen Flux (mg/L) NA
Biological Oxygen Demand (5 day) (mg/L) NA
Escherichia coli (# per 100 mL) 126 3
Chloride (mg/L) 230
1 NMCWD goals are based on MPCA standards included in MN Rules 7050. Revisions to MN Rules 7050 will supersede
NMCWD standards. Note that MN Rule 7050.0220 includes standards for additional parameters that are enforced by the
MPCA.
2 Shallow lakes have a maximum depth less than 15 feet or littoral area greater than 80% of the total lake surface area.
3 126 organisms per 100 mL as a geometric mean of not less than five samples within any month, nor shall more than 10%
of all samples within a month exceed 1,260 organisms per 100 mL.
3.1.2 Other Lake Health Goals
In addition to the water quality goals presented in Table 3-1, the NMCWD’s 2017 Plan expresses the
desire to establish holistic lake health targets for District-managed lakes. The holistic lake health targets
consider a wide range of factors, with an increased emphasis on the role of ecological factors in overall
lake health and the interrelated nature of these factors.
Table 3-2 lists the evaluation factors used by the NMCWD to holistically assess lake health. Numerical
goals exist for some of the factors presented in this table (e.g., MPCA water quality standards), while other
holistic health factors are assessed qualitatively by comparing to narrative criteria. The NMCWD
collaborates with stakeholders and regulatory agencies (MPCA, MDNR) to develop lake-specific numerical
goals for ecological indicators where appropriate.
11
Table 3-2 NMCWD holistic lake health assessment evaluation factors
Lake Health Assessment
Factors Evaluation Factors
Chemical Water Quality
• Nutrients
• Sediment
• Clarity
• Chlorophyll a
• Chloride
Aquatic Communities
• Aquatic Plant IBI1- species richness and floristic quality
• Invasive Species Presence
• Phytoplankton Populations
• Blue-green Algae Presence
• Zooplankton Populations
Water Quantity
• Water Levels
• Water Level Bounce
• Groundwater Levels
Recreation
• Shore Access
• Navigation Potential
• Aesthetics
• Use Metrics
Wildlife • Upland biodiversity
• Buffer extent/width
1 Lake plant eutrophication Index of Biotic Integrity (IBI) methodology developed by the MDNR and MPCA
3.2 Lower Minnesota River Watershed TMDL Report—Draft
Both Lake Cornelia and Lake Edina were first listed on the 303(d) Impaired Waters list in 2008 for impaired
aquatic recreational use due to excess nutrients. For all water bodies listed on the 303(d) Impaired Waters
list, the MPCA is required to conduct a Total Maximum Daily Load (TMDL) study for each pollutant that
causes the water body to not meet the state water quality standards. A TMDL study for the Lower
Minnesota River Watershed, which included Lake Cornelia and Lake Edina, was completed in 2017-2018.
The draft TMDL report was submitted to the MPCA in June 2018 (Barr Engineering Co. & MPCA, 2018).
The Lower Minnesota River Watershed TMDL report was developed as part of a larger effort to address
impaired waters in the northern urban portion of the watershed in the Twin Cities Metropolitan area (Barr
Engineering Co. & MPCA, 2018). The areas investigated in the report include portions of Carver and
Hennepin Counties, specifically the Riley Purgatory Bluff Creek Watershed District (RPBCWD) and Nine
Mile Creek Watershed District (NMCWD). The report provides TMDLs for 13 lakes impaired for nutrients
(which include North Cornelia, South Cornelia, and Lake Edina), two streams impaired for bacteria, and
one stream impaired for total suspended solids (TSS) and impaired biota.
The modeling efforts for the Lower Minnesota River Watershed TMDL report indicate that the percent
reductions in total phosphorus loadings for North Cornelia, South Cornelia, and Lake Edina needed to
12
reach the MPCA’s water quality total phosphorus standard of 60 µg/L are 59%, 61%, and 34%,
respectively, during the growing season. These total phosphorus load reductions were also divided
among watershed, upstream lakes, and internal loads. Table 3-3 summarizes the estimated load
reductions required for each lake to reach water quality goals. The TMDL analysis for North Cornelia
indicates that total phosphorus reductions should be focused on external as well as internal management
efforts. The results for South Cornelia indicate that management efforts in upstream lakes (North Cornelia)
will have a large impact on reducing lake total phosphorus concentrations. Internal management efforts
were also recommended. For Lake Edina, the TMDL analysis suggests that upstream management efforts
in North and South Cornelia will assist in reducing in-lake total phosphorus concentrations in Lake Edina.
External management efforts in the direct Lake Edina watershed are also recommended for load
reductions. Refer to the Draft Lower Minnesota River Watershed TMDL for further details.
Table 3-3 Growing season total phosphorus load reductions summary from the Draft Lower Minnesota River Watershed TMDL Report
Water
Body
Watershed Load
Reductions
Internal Load
Reductions
Upstream Lake
Reductions Total Load Reductions
lbs/growing
season
% of
Total
lbs/growing
season
% of
Total
lbs/growing
season
% of
Total
lbs/growing
season
% of
Existing
North
Cornelia 110 51% 104 49% - - 214 59%
South
Cornelia 0 0% 150 60% 100 40% 250 61%
Lake
Edina 38 42% 0 0% 52 58% 90 34%
3.3 Natural or Background Water Quality Conditions
There were three major baseline water quality prediction methods that were used to assess natural,
background water quality in Lake Cornelia in the past and these methods have been referenced in
previous UAAs. The first baseline water quality prediction method is the Minnesota Lake Eutrophication
Analysis Procedure (MnLEAP), which is intended to be used as a screening tool for estimating lake
conditions and for identifying “problem” lakes (Heiskary & Wilson, 1990). Results of the MnLEAP modeling
completed for Lake Cornelia in the 2010 UAA identified the lake as one that could achieve “better” water
quality than is currently observed by showing estimated natural total phosphorus concentrations ranging
from 55-97 µg/L.
The second baseline water quality prediction method is based on the Vighi and Chiaudani (MEI) model
(1985). The Vighi and Chiaudani MEI model provides reasonable estimates of pre-European settlement
total phosphorus concentrations for lakes, based on current alkalinity or conductivity water quality
measurements. The modeling results completed for the Lake Cornelia 2010 UAA indicated that pre-
European settlement (natural background) total phosphorus concentrations ranged from 27-66 μg/L.
13
The third baseline water quality prediction method is the Wisconsin Lake Modeling Suite (WiLMS) model
(WI-DNR, Wisconsin Lake Modeling Suite (WILMS), 2004). The WiLMS model (Lake Total Phosphorus
Prediction Module) uses an annual time step and predicts spring overturn, growing season mean, and
annual average total phosphorus concentrations in lakes. Natural, background total phosphorus
concentrations found during the Lake Cornelia 2010 UAA were estimated at 53-133 μg/L.
For additional information on these baseline water quality prediction methods, please refer to the Lake
Cornelia UAA developed in 2010. These methods were not revised for this UAA update as the previous
estimates were adequate. These baseline water quality prediction methods were also not estimated for
Lake Edina for this UAA effort.
3.4 NMCWD Adaptive Management Approach
The NMCWD implements an adaptive management approach to improve lake health based on water
quality and assessment of the other holistic lake health factors. While striving to achieve the state
standards for shallow lakes, the NMCWD recognizes that achieving the water quality goals may not be
feasible for some lakes or may require a timeframe that extends several decades. For these situations, the
NMCWD’s objective it to make reasonable and measureable progress towards meeting the water quality
goals and other holistic lake health targets.
The NMCWD reviews lake monitoring data annually to assess progress toward lake management goals.
For lakes that are meeting the goals, the NMCWD continues periodic monitoring to track variations in
water quality and potential trends. If water quality declines or if water quality does not meet NMCWD
goals, a lake-specific Use Attainability Analysis is developed or updated to identify additional protection
and improvement measures, as is being completed in this report for Lake Cornelia and Lake Edina.
14
4.0 Lake Basin and Watershed Characteristics
The following sections describe the unique characteristics of the Lake Cornelia and Lake Edina basins and
watersheds. Lake Cornelia and Lake Edina form a chain of lakes; water that flows through North Lake
Cornelia drains to South Lake Cornelia, then Lake Edina, before ultimately discharging into the North Fork
of Nine Mile Creek.
4.1 Lake Cornelia Basin Characteristics
Lake Cornelia is located in the central portion of the City of Edina. The lake was a natural marsh area prior
to the 1960s when a control structure was installed by the City of Edina. Lake Cornelia is comprised of
North (North Cornelia) and South (South Cornelia) basins, connected by a 12-inch culvert under
66th Street (with an invert elevation of 859.0 feet MSL) on the south side of the North Cornelia, and a
secondary 12-inch pipe located on the southeast side of North Cornelia (with an invert elevation of
860.2 feet MSL). Ultimately water levels in North Cornelia are controlled by the outlet structure at South
Cornelia. The outflow from South Cornelia discharges directly over a 14-foot long weir structure with a
control elevation of 859.1 feet MSL. Discharge from South Cornelia are conveyed to Lake Edina through
an extensive storm sewer network. Due to limited storm sewer capacity downstream of Lake Cornelia,
stormwater runoff from portions of Cornelia Drive and Dunberry Lane backs up into the lake during large
storm events, which provides temporary storage of the flood volumes.
4.1.1 North Cornelia
North Cornelia has a water surface area of approximately 19 acres, a maximum depth of 7 feet, and a
mean depth of approximately 3 feet at a normal water surface elevation of 859.1. At this elevation the lake
volume is approximately 75 acre-feet. The water level in the lake is controlled mainly by weather
conditions (snowmelt, rainfall, and evaporation), by the outlet capacity of the pipe on North Cornelia, and
by the elevation and capacity of the outlet structure located on South Cornelia. The stage-storage-
discharge relationship that was used in this study for North Cornelia is shown in Table 4-1. The
bathymetric information used to inform this stage-storage-discharge relationship is shown on Figure 4-1
and is based on surveys completed by the City of Edina in 2017.
4.1.2 South Cornelia
South Cornelia has a water surface area of approximately 33 acres, a maximum depth of 8 feet, and a
mean depth of 4.2 feet at a normal surface elevation of 859.1. At this elevation the lake volume is
approximately 166 acre-feet. The water level in the lake is controlled by the elevation of the weir structure
at the south side of the lake. The stage-storage-discharge relationship that was used in this study for Lake
Cornelia is shown in Table 4-1. The bathymetric information used to inform this stage-storage-discharge
relationship is shown on Figure 4-1 (City of Edina, 2017).
15
Table 4-1 Stage-storage-discharge relationships for Lake Cornelia
Elevation Area (acres) Cumulative
Storage (ac-ft) Discharge (cfs) Comment
North Cornelia
852.00 0.2 0.0 0.0
Wet Detention Storage Volume
853.00 0.7 0.4 0.0
854.00 2.3 1.9 0.0
855.00 10.5 8.3 0.0
856.00 15.6 21.3 0.0
857.00 16.7 37.5 0.0
858.00 17.6 54.7 0.0
859.00 18.7 72.8 0.0 Invert of outlet pipe
859.1 18.9 74.7 0.0 Normal Water Level
859.25 19.3 77.6 0.1
Available Live Storage for Flood
Control
859.50 19.8 82.5 0.5
860.00 20.9 92.6 1.5
860.50 31.1 105.6 2.5
862.00 32.7 153.5 4.3
863.00 36.3 188.0 15.0
South Cornelia
851.00 0.4 0.0 0.0
Wet Detention Storage Volume
852.00 4.2 2.3 0.0
853.00 11.5 10.1 0.0
854.00 17.9 24.8 0.0
855.00 23.4 45.5 0.0
856.00 27.7 71.0 0.0
857.00 29.9 99.8 0.0
858.00 31.3 130.4 0.0
859.00 33.2 162.7 0.0
859.10 33.2 166.1 0.3 Normal Water Level
859.25 33.3 171.0 2.2
Available Live Storage for Flood
Control
859.50 33.5 179.4 6.5
859.75 33.6 187.8 10.0
860.00 33.8 196.2 10.8
861.00 34.7 230.4 23.5
861.10 35.0 233.9 23.9
862.10 36.7 269.8 27.8
863.10 39.0 307.7 31.7
864.00 40.2 343.3 35.2
865.80 41.2 416.5 60.4
868.00 49.5 516.3 92.1
3'
7'
7'
6'
5'6'
5'5'6'7'6'5'7'4'
4'
3
'6'
4
'
6'
2'
8'
5'
5'
4'
1'
3'2'
2'
5'3'
5'
1'
1'
North Lake Cornelia
South LakeCornelia
Point DrCornel
ia
DrWooddale AveWes
t
S
h
o
r
e
D
r
Oaklawn Ave
W 66th St
Creston RdBulfanz Rd
Laguna Dr
Cornelia DrW 64t
h
S
t
Barr Footer: ArcGIS 10.6.1, 2019-04-30 14:19 File: I:\Client\Nine_Mile_Creek_WD\Work_Orders\23270634_Project\2018_Cornelia_Edina_UAA\Maps\Reports\UAA\Figure 4-1 Lake Cornelia Bathymetry.mxd User: RCS2
LAKE CORNELIABATHYMETRY
FIGURE 4-1
0 150 300
Feet
!;N
1-ft Bathymetric Contours*
Digital Globe Imagery Circa June 2017
* Source: City of Edina, 2017
17
Since Lake Cornelia’s two basins are shallow, the lake is prone to frequent wind-driven mixing of the lake’s
shallow waters during the summer. Therefore, one would expect Lake Cornelia to be polymictic (mixing
many times per year) as opposed to lakes with deep, steep-sided basins that are usually dimictic (mixing
only twice per year). Daily monitoring of the lake would be necessary to precisely characterize the mixing
dynamics of a lake, but the limited data gathered from Lake Cornelia strongly suggests that the lake is
polymictic.
4.2 Lake Edina Basin Characteristics
Lake Edina is located south of Lake Cornelia in the City of Edina. The lake is a natural marsh area. North
Cornelia and South Cornelia are located upstream of Lake Edina and discharges from South Cornelia are
conveyed to Lake Edina through an extensive storm sewer network. The control structure from Lake Edina
is a 12-foot weir structure, with a control elevation of 822.0 feet MSL. Outflow from Lake Edina discharges
to the North Fork of Nine Mile Creek via a 36-inch storm sewer under Trunk Highway (TH) 100.
Lake Edina has a water surface area of approximately 25 acres, a maximum depth of 5 feet, and a mean
depth of approximately 3 feet at a normal water surface elevation of 822.0 feet MSL. At this elevation the
lake volume is approximately 68 acre-feet. The water level in the lake is controlled mainly by weather
conditions (snowmelt, rainfall, and evaporation), by the outlet capacity of the pipe, by the volume of flow
received from South Cornelia, and through groundwater interactions with Nine Mile Creek. The stage-
storage-discharge relationship that was used in this study for Lake Edina is shown in Table 4-2.
Table 4-2 Stage-storage-discharge relationship for Lake Edina
Elevation Area (acres) Cumulative
Storage (ac-ft) Discharge (cfs) Comment
Lake Edina
817.00 0.0 0.0 0.0
Wet Detention Storage Volume
818.00 0.1 0.1 0.0
819.00 11.5 5.9 0.0
820.00 20.6 21.9 0.0
821.00 23.6 44.0 0.0
822.00 24.6 68.1 0.0 Normal Water Level
822.20 24.8 73.0 1.6
Available Live Storage for Flood
Control
822.50 25.1 80.5 4.3
823.00 25.6 93.2 9.3
824.00 27.1 119.5 21.8
826.00 34.6 181.3 57.0
827.00 37.3 217.2 100.0
828.50 410.3 552.9 115.0
18
The bathymetric information used to inform this stage-storage-discharge relationship is shown on
Figure 4-2 (City of Edina, 2017).
Since Lake Edina is shallow, the lake is prone to frequent wind-driven mixing of the lake’s shallow waters
during the summer. One would therefore expect Lake Edina to be polymictic (mixing many times per year)
as opposed to lakes with deep, steep-sided basins that are usually dimictic (mixing only twice per year).
Daily monitoring of the lake would be necessary to precisely characterize the mixing dynamics of a lake,
but the limited data gathered from Lake Edina strongly suggests that the lake is polymictic.
Gilford Dr WooddaleAveSedumLa
Poppy La
Phlox La
Hib
i
s
c
u
s
A
v
eMonardo LaWes
t
Sho
re
D
r
100
Lake Edina
3'2'1'4'3'3'3'
Barr Footer: ArcGIS 10.6.1, 2019-04-30 14:33 File: I:\Client\Nine_Mile_Creek_WD\Work_Orders\23270634_Project\2018_Cornelia_Edina_UAA\Maps\Reports\UAA\Figure 4-2 Lake Edina Bathymetry.mxd User: RCS2
LAKE EDINABATHYMETRY
FIGURE 4-2
0 250
Feet
!;N
1-ft Bathymetric Contours*
Digital Globe Imagery Circa June 2017
* Source: City of Edina, 2017
20
4.3 Watershed Characteristics
Land use practices within a lake’s watershed impact the lake and its water quality by altering the volume
of stormwater runoff, sediment load, and nutrient load (namely phosphorus) that reaches the lake from
the lake’s watershed. Each land use contributes a different amount of runoff and phosphorus to the lake,
thereby impacting the lake’s water quality differently. As land use changes over time, changes can be
expected in downstream water bodies as a result.
Historically, the Lake Cornelia and Lake Edina watersheds were primarily comprised of basswood, sugar
maple, and oak forests. There were also numerous wetlands located throughout the watersheds. The
terrain varies from relatively flat to rolling.
Lake Cornelia’s watershed is 986 acres, including the surface area of the lake (52 acres). Runoff from the
watershed enters both North and South Cornelia through overland flow and from several storm sewer
outfalls at various points along the lakeshore, although the majority of the watershed flows through North
Cornelia before entering South Cornelia. Lake Edina’s watershed is 1,380 acres, including the 986-acre
watershed of Lake Cornelia and the surface area of the lake (25 acres). Runoff from the watershed enters
Lake Edina through overland flow and from several storm sewer outfalls along the lakeshore.
Based on the 2016 Generalized Land Use Inventory Dataset developed by the Metropolitan Council
(Metropolitan Council, 2016) and analysis of aerial imagery, the watersheds of Lake Cornelia and Lake
Edina are near fully-developed. Table 4-3 provides a summary of the land use classifications within each
watershed. The major land use classification in each watershed is low-density residential (single family
detached). The watershed also includes some proportions of commercial, highway, and open water. To a
lesser extent, the land use consists of high density residential, developed park, high impervious
institutional, open space, and office space. Figure 4-3 shows a map of the land use classifications within
each watershed.
Table 4-3 Land use classifications in the Lake Cornelia and Lake Edina watersheds
Land Use Classification North Cornelia South Cornelia Lake Edina
Institutional 3% 0% 6%
Major Highway 12% 0% 3%
Mixed Use Commercial 1% 0% 0%
Mixed Use Residential 1% 0% 0%
Multi-Family 7% 0% 2%
Office 8% 0% 2%
Open Water 5% 29% 6%
Park/Recreational 7% 3% 5%
Retail/Commercial 13% 0% 1%
Single Family Attached 3% 0% 1%
Single Family Detached 39% 68% 72%
Undeveloped/Open Space 1% 0% 2%
Total Watershed Area (ac) 873.7 112.6 393.1
LakeEdina
North LakeCornelia
South LakeCornelia
456731
456717
456717
456753
100
62
62
LE_44
SC_3
SC_2
NC_6
NC_130NC_135
LE_54
LE_54
LE_38
LE_51
NC_78
NC_72
NC_30
NC_88
NC_5
NC_2
NC_4SC_1
NC_62 NC_3
LE_1
NC_4A
Source: Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS,USDA, USGS, AeroGRID, IGN, and the GIS User Community
Barr Footer: ArcGIS 10.6.1, 2019-04-30 13:54 File: I:\Client\Nine_Mile_Creek_WD\Work_Orders\23270634_Project\2018_Cornelia_Edina_UAA\Maps\Reports\UAA\Figure 4-3 Land Use Lake Cornelia and Lake Edina Watersheds.mxd User: RCS2
LAND USELAKE CORNELIA ANDLAKE EDINA WATERSHEDS
FIGURE 4-3
0 500 1,000 1,500
Feet
!;N
Subwatersheds
Land Use (Met Council, 2016)
Industrial and Utility
Institutional
Major Highway
Mixed Use Commercial
Mixed Use Residential
Multifamily
Office
Open Water
Park, Recreational, or Preserve
Retail and Other Commercial
Single Family Attached
Single Family Detached
Undeveloped
Digital Globe Imagery Circa June 2017
22
4.4 Lake Inflows and Drainage Areas
Watershed modeling depends on the evaluation of the watershed conditions as they relate to stormwater
runoff. Therefore, the hydrology of the Lake Cornelia and Lake Edina watersheds is discussed in the
following sections.
4.4.1 Natural Conveyance Systems
Under existing conditions, Lake Cornelia and Lake Edina receive natural surface water inflows only from
their direct watersheds.
4.4.2 Stormwater Conveyance Systems
The stormwater conveyance system in the Lake Cornelia watershed is comprised of a network of storm
sewers, ditches, ponds, and wetlands. The ponds and wetlands provide water quality treatment of the
runoff prior to discharge to Lake Cornelia. Storm sewer and ditches convey stormwater runoff to and from
the ponds and wetlands, and ultimately convey the runoff from the watershed to Lake Cornelia. The
locations of the major stormwater conveyance features are shown on Figure 4-4.
There are about 15 ponding areas (wet and dry detention ponds, infiltration basins, and wetlands) in the
Lake Cornelia watershed. There are also two constructed underground storage areas. A detailed listing of
existing pond information is located in Appendix A.
Similar to Lake Cornelia’s conveyance features, Lake Edina’s conveyance system is comprised of a network
of storm sewer, dry detention ponds, and wetlands within the tributary watershed. The locations of the
major stormwater conveyance features are shown in Figure 4-5.
There are four ponding areas (dry detention ponds, infiltration basins, and wetlands) in the Lake Edina
watershed. A detailed listing of existing pond information is located in Appendix A.
4.4.3 Southdale Center Cooling System Discharge
Prior to 2011, the Southdale Shopping Center pumped groundwater through their heating and cooling
system and discharged the wastewater into Point of France Pond, which is upstream of North Cornelia.
Based on historic records, the Southdale cooling water discharge resulted in approximately 30 to
40 million gallons of water per year entering the North Cornelia watershed (ranging from 1.0 million to
6.8 million gallons per month). In 2011, the Minnesota Department of Natural Resources (MDNR) did not
renew the Southdale groundwater appropriation permit, which required Southdale to abandon the use of
groundwater in their heating and cooling system and eliminated the discharge of cooling water to Lake
Cornelia and downstream Lake Edina.
Melody
North LakeCornelia
South LakeCornelia
LakeNancy
PamelaPond
Birchcrest
LakeOtto
SwimmingPool Pond
Point ofFrancePond
456717
456753
456731
456731
456731
456753456717
62
62
62
100
100
62
LE_44
SC_3
SC_2
NC_6
NC_130
NC_135
LE_54LE_38
LE_38 LE_51
NC_78
NC_72
NC_30
NC_88
NC_5
NC_2
NC_4SC_1
NC_62
NC_3
LE_1
NC_4A
Barr Footer: ArcGIS 10.6.1, 2019-07-09 10:35 File: I:\Client\Nine_Mile_Creek_WD\Work_Orders\23270634_Project\2018_Cornelia_Edina_UAA\Maps\Reports\UAA\Figure 4-4 Lake Cornelia Stormwater Conveyance.mxd User: RCS2LAKE CORNELIASTORMWATERCONVEYANCE
FIGURE 4-4
0 400 800
Feet
!;N
#Storm Sewer Outlet
Storm Sewer
Subwatersheds
Major Watersheds
Digital Globe Imagery Circa June 2017
LakeEdina
456717
456717
456717
100
100
LE_44
SC_3
SC_2
LE_54
LE_54
LE_38
LE_51
NC_72
NC_4
SC_1
NC_62
LE_1
Barr Footer: ArcGIS 10.6.1, 2019-07-09 12:16 File: I:\Client\Nine_Mile_Creek_WD\Work_Orders\23270634_Project\2018_Cornelia_Edina_UAA\Maps\Reports\UAA\Figure 4-5 Lake Edina Stormwater Conveyance.mxd User: RCS2LAKE EDINASTORMWATERCONVEYANCE
FIGURE 4-5
0 300 600
Feet
!;N
#Storm Sewer Outlet
Storm Sewer
Subwatersheds
Major Watersheds
Digital Globe Imagery Circa June 2017
25
5.0 Existing Water Quality
5.1 Water Quality
The NMCWD conducted intensive water quality monitoring in North and South Cornelia in 2015, 2016,
and 2017 in support of this UAA. The NMCWD also collected data in 2004 and 2008. Intensive monitoring
was conducted for Lake Edina in 2008, 2012, 2015 and 2017. Monitoring data was also collected through
the Metropolitan Council Citizen Assisted Monitoring Program (CAMP) from North Cornelia in 2003, 2005,
2006, 2007, and 2008 and from Lake Edina in 2004 and 2005.
5.1.1 Phosphorus, Chlorophyll a, and Clarity
The NMCWD intensive monitoring included the lake eutrophication parameters of total phosphorus (TP),
chlorophyll a, and Secchi disc depth to assess water clarity. Data are presented using box plots. The box
plots show averages (red cross), median values (straight horizontal line), maximum and minimum values
(blue dots), as well as the region where 50 percent of the data lie (the area within the boxes). Box plots
show on Figure 5-1 and Figure 5-2 display the summer average TP and chlorophyll a concentrations for
North and South Lake Cornelia from 2003 through 2017. Figure 5-3 shows the summer average Secchi
disc transparency depths for North and South Cornelia from 2003 through 2017. Figure 5-4 shows box
plots of historic summer average TP and chlorophyll a concentrations from 2004 through 2017 for Lake
Edina. The historic summer average Secchi disc transparency depths for Lake Edina are shown on
Figure 5-5.
5.1.1.1 Lake Cornelia (North and South Basins)
There is significant variability in total phosphorus and chlorophyll a concentrations in North and South
Cornelia from year to year, as well as within a given year. The significant variability can be a reflection of
numerous factors, including climatic variability, changing aquatic plant populations, curly-leaf pondweed
treatments, copper sulfate treatments to control phytoplankton blooms, and likely periodic winter fish
kills. There does not seem to be a clear or consistent trend in TP and chlorophyll a concentrations in the
lake over the monitoring period presented (both North and South Cornelia). The TP and chlorophyll a
concentrations in North and South Cornelia are well above the 0.06 mg/L (60 µg/L) and 20 µg/L shallow
lake criteria, respectively.
5.1.1.2 Lake Edina
It can be seen that for Lake Edina there is significant variation in total phosphorus and chlorophyll a
concentrations from year to year, as well as within a given year (i.e., the individual points and the size of
the box are an indication of change or data variability). Summer average TP and chlorophyll a
concentrations in Lake Edina were generally higher than the shallow lake criteria of 0.06 mg/L (60 µg/L)
and 20 µg/L, respectively.
It appears that TP and chlorophyll a concentrations in Lake Edina have declined in recent years; however,
there is an insufficient amount of data to determine if it is a statistically-significant trend. In the spring of
2015, the water level dropped to the point that the lakebed was dry for an extended period. This may
26
have had an effect on internal phosphorus release from lake sediments or the biota may have been
altered in some way that has altered the phosphorus and phytoplankton (measured as chlorophyll a)
growth dynamics. The 2017 summer average TP and chlorophyll a concentrations were the lowest
measured since 2004 (for data collected by the District and Metropolitan Council Environmental Services).
A comparison of aquatic plant surveys conducted in 2012 and 2017 indicate that the aquatic plant
coverage was notably greater in 2017; this increased abundance may be responsible for the lower TP and
chlorophyll a concentrations in 2017.
27
Figure 5-1 Summer total phosphorus and chlorophyll a concentrations in North Cornelia from 2004 through 2017. The red crosses indicate the average summer (June through
September) concentrations.
Figure 5-2 Summer total phosphorus and chlorophyll a concentrations in South Cornelia from 2004 through 2017. The red crosses indicates the average summer (June through September) concentrations.
28
Figure 5-3 Summer Secchi disc depth readings in (a) North and (b) South Cornelia from 2004 through 2017. The red crosses indicate the average of summer (June through September) readings.
Figure 5-4 Summer total phosphorus and chlorophyll a concentrations in Lake Edina from 2004 through 2017. The red crosses indicate the average summer (June through September) concentrations.
29
Figure 5-5 Summer Secchi disc depth readings in Lake Edina from 2004 through 2017. The red crosses indicate the average of summer (June through September) readings.
0.39 0.30
0.48
0.29
0.54
0.77
2004 2005 2008 2012 2015 2017
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Secchi Disk Depth (m)Year
30
5.1.2 Chlorides
Because high concentrations of chloride can harm fish and plant life, MPCA has established a chronic
exposure chloride standard of 230 mg/L or less, and considers two or more exceedances of the chronic
standard in 3 years to be an impairment. In April of 2015 and 2016, chloride concentrations in North
Cornelia exceeded the MPCA standard of 230 mg/L (254 mg/L on April 14, 2015) and June 12 (314 mg/L
on April 6, 2016). Although North Cornelia has not been listed by the MPCA as impaired for chlorides, the
data indicate the lake meets the MPCA criterion for impairment. Following the high concentrations
recorded in April of 2015 and 2016, chloride concentrations in North Cornelia decreased to below the
standard throughout the summer months. Spring chloride concentrations were not recorded in 2017.
South Cornelia and Lake Edina also showed patterns of high chloride concentrations in the spring, with
concentrations lowering throughout the late-spring and summer months. However, the chloride
concentrations in these lakes did not exceed the MPCA standard of 230 mg/L.
5.2 Sediment Quality
5.2.1 Lake Cornelia (North and South Basin)
Phosphorus in lake bottom sediments is often bound to a range of different elements such as iron and
manganese (often referred to as mobile phosphorus), aluminum, or calcium. Phosphorus can also be
found incorporated into organic matter (organically-bound phosphorus). It is the mobile phosphorus
fraction that releases from sediment during low oxygen conditions (this is often called internal loading).
Organically-bound phosphorus also releases phosphorus from lake sediment but at a slow rate.
Phosphorus composition data from cores taken from lake-bottom sediments in 2008 were used to
estimate the maximum potential phosphorus release rate (internal loading) in North and South Lake
Cornelia and to estimate doses and costs for an alum treatment. Mobile phosphorus in the top
10 centimeters of cores taken in North Lake Cornelia was 55 µg/cm3 while in South Cornelia it was
17 µg/cm3. Hence, it can expected that internal loading will be greater in North Lake Cornelia. Table 5-1
provides the maximum potential internal loading rates for North and South Lake Cornelia and a
comparison with other lakes in the metro area. It can be seen that the maximum potential release rate for
North and South Cornelia are similar to other metro lakes that are eutrophic and experience mid-summer
phytoplankton (algae) blooms.
5.2.2 Lake Edina
Sediment cores were collected at four locations in Lake Edina in 2018 and analyzed for mobile
phosphorus, aluminum-bound phosphorus, calcium-bound phosphorus, and organically-bound
phosphorus. The average mobile phosphorus in the top 10 centimeters of the four cores taken in Lake
Edina was 3.6 µg/cm3. This is significantly lower than the mobile phosphorus concentrations in North and
South Lake Cornelia. This mobile phosphorus concentration is essentially “background” and it can be
expected that there is essentially no release of the mobile phosphorus fraction in the bottom sediments of
Lake Edina during the summer months or any other season. There were typical concentrations of
organically-bound phosphorus in the Lake Edina sediments. Organically-bound phosphorus has a slow
31
rate of decay and release from lake bottom sediments and as a result has the potential to be a minor
source of phosphorus. However, UAA in-lake modeling efforts indicate phosphorus release from lake
bottom sediments is limited (see Section 6.3). Overall, it can be concluded that there is no need to treat
Lake Edina sediments for phosphorus internal loading.
Table 5-1 Maximum potential internal loading rate for North and South Cornelia and Lake Edina compared to other Twin Cities Metro Area lakes.
Lake Maximum Potential Internal P Load (mg/m2/d)
Kohlman1 17.0
Isles (pre-alum, deep hole)2 14.1
Harriett (pre-alum, deep hole) 2 11.1
Calhoun/Bde Maka Ska (pre-alum, deep) 2 10.8
Fish E3 10.5
Cedar (pre-alum) 2 9.3
Fish W3 8.1
Como3 7.6
North Cornelia 7.6
Calhoun/Bde Maka Ska (pre-alum, shallow) 3 5.6
Keller1 3.5
Parkers3 3.5
Phalen3 2.3
McCarrons3 2.0
Bryant3 1.5
South Cornelia 1.3
Nokomis3 1.0
Minnewashta3 0.2
Edina 0.0
Christmas3 0.0
______________________
Sources:
1 (Barr Engineering Co., 2007)
2 (Huser & Pilgrim, 2014)
3 (Pilgrim, Huser, & Brezonik, 2007)
32
5.3 Aquatic Communities
The fish, zooplankton, phytoplankton, and aquatic plants residing in Lake Cornelia and Lake Edina are all
linked and the composition and abundance of biota observed in the lakes provide indications of their
impaired conditions and how biological management could improve water quality. The biota in Lake
Edina are also reflective of the shallow lake ecology that results in low lake levels during dry conditions as
well as influences from Lake Cornelia such as nutrient loading and biological inputs.
5.3.1 Lake Cornelia
5.3.1.1 Aquatic Plants
Macrophytes, also called aquatic plants, grow in aquatic systems such as streams and lakes. There is a
wide range of macrophytes including species attached to the lake bottom, species unattached and
floating, submerged species, and emergent species (e.g., cattails). Macrophytes are an important part of a
shallow lake ecosystem and provide critical habitat for aquatic insects and fish. A healthy native plant
community contributes to the overall health of the lake. However, a dense non-native plant community
can create problems, including recreational use impairment, fluctuating water quality, and a less than ideal
fisheries habitat, which has adverse impacts on the fish community. The dense growth makes it difficult
for invertebrates and other organisms that fish eat to survive. So, with less to eat and less open water, fish
populations decrease (MPCA, Eurasian Water Milfoil, 2019). The dense growth makes it hard for fish to
catch food. When fish are less effective at controlling prey species, an unbalanced fishery results (Indiana
Department of Natural Resources, 2019)
The Minnesota Department of Natural Resources (MDNR) developed the Lake Plant Eutrophication IBI in
recent years to assist the MPCA in assessing lake impairment based on the plant community. The Lake
Plant Eutrophication IBI includes two metrics to assess the viability of aquatic life. The first metric is
species richness—the estimated number of species in a lake. The second metric is floristic quality index
(FQI), which distinguishes the quality of the plant community and can be a reflection of the quantity of
nutrients in the lake.
The MDNR’s Lake Plant Eutrophication IBI was used to assess the health of the North and South Lake
Cornelia plant communities. Aquatic plant data collected by NMCWD from 2004 through 2017 was used
to determine species richness and FQI scores. The scores were then compared with MDNR Lake Plant
Eutrophication IBI impairment thresholds (a minimum of 11 species and an FQI score of at least 17.8) to
determine whether the Lake Cornelia plant community would be considered impaired.
The Lake Cornelia plant community has generally failed to meet the MDNR Lake Plant Eutrophication IBI
criteria since 2004, a reflection of the lake’s poor water quality. The number of species observed in North
Cornelia ranged from 2 to 7 which is less than the plant IBI impairment threshold of at least 11 species
(Figure 5-6). The FQI values from the north basin ranged from 6.4 to 12.7 which is less than the plant IBI
impairment threshold for FQI of at least 17.8 (Figure 5-7). The number of species observed in South
Cornelia ranged from 3 to 12 (Figure 5-8). This metric met the MDNR plant IBI impairment threshold
during August 2015 and June 2016, but was less than the plant IBI impairment threshold of at least
11 species during all other surveys. The FQI values from South Cornelia ranged from 6.9 to 18.1. FQI met
33
the MDNR plant IBI during August 2015, but was less than the plant IBI threshold for FQI of at least 17.8
during all other surveys (Figure 5-9). The plant IBI has not yet been used by the MPCA/MDNR to
determine impairment. However, it is expected to eventually be used as an assessment tool to determine
biological impairment.
Figure 5-6 North Cornelia Macrophyte Species Richness Compared with Plant IBI Threshold
for Species Richness
Figure 5-7 North Cornelia Floristic Quality Index (FQI) Compared with Plant IBI Threshold for
FQI
34
Figure 5-8 South Cornelia Macrophyte Species Richness Compared with Plant IBI Threshold for Species Richness
Figure 5-9 South Cornelia Floristic Quality Index (FQI) Compared with Plant IBI Threshold for FQI
35
Three non-native aquatic invasive species (AIS) are present in Lake Cornelia: hybrid cattail, purple
loosestrife, and curly-leaf pondweed.
In 2017, hybrid cattail was prevalent in North Cornelia. The hybrid cattail spreads aggressively through
underground roots and can become very dense, crowding out native species (Wisconsin Wetlands
Association, 2017).
Purple loosestrife has been observed in North and South Cornelia since 2004. Purple loosestrife was
observed along the northwest shoreline of North Cornelia during 2004, 2008, and in June of 2015. In
August of 2015, purple loosestrife was again present along the northwest shoreline of North Cornelia, but
was also observed along the southeast shoreline. Purple loosestrife has also been observed in South
Cornelia since 2004. During the 2004, 2008, and 2015 surveys, purple loosestrife was found sporadically
along the entire shoreline. In 2017, purple loosestrife was common in both North and South Cornelia and
moderately dense around much of the undeveloped lakeshore.
Curly-leaf pondweed was not observed in North or South Cornelia during the 2004 survey; however, curly-
leaf pondweed was present in North and South Cornelia in a few small patches in 2008 and 2013. In 2015,
curly-leaf pondweed was problematic throughout North and South Cornelia. In June 2015, the density was
light in the center of North Cornelia, but denser along the northern, eastern, and southern shores. Curly-
leaf pondweed was also found throughout North Lake Cornelia in August 2015. Although less dense than
in June, a moderate to heavy density was observed near shore and a light density at the center indicating
a late die-off of the species in 2015. Additionally, curly-leaf pondweed was observed throughout South
Lake Cornelia in both June and August, 2015.
In 2016, the density of curly-leaf pondweed surveyed in Lake Cornelia was significantly more than what
was observed in previous years. The June plant survey documented dense curly-leaf pondweed growth
throughout the entire lake. It is hypothesized that the increased phosphorus pulse from senescence of the
dense curly-leaf pondweed community increased the severity of late-summer blue-green algal blooms.
Curly-leaf pondweed normally dies near the end of June and then decays, adding a pulse of phosphorus
to the lake. When the 2016 dense growth of curly-leaf pondweed in the Lake Cornelia senesced, the lake’s
phosphorus concentrations rapidly increased, which may have fueled the severe blue-green algal blooms
that occurred that year.
On April 23, 2017, a point-intercept plant survey was completed by the City of Edina. The survey found
curly-leaf pondweed forming a solid mat over the entire lake with the exception of a few meters (about
6 feet) buffer near the immediate shoreline (Figure 5-10). The City completed an herbicide (endothall)
treatment in May 2017 to manage the curly-leaf pondweed. A post-treatment survey was completed in
early-June 2017 and curly-leaf pondweed was not observed in North Cornelia and was only observed at
one location in South Cornelia. Thus, the herbicide treatment effectively managed the lake’s curly-leaf
pondweed infestation.
36
Figure 5-10 Curly-leaf pondweed growth observed in Lake Cornelia in 2017.
In 2017, the number of plant species observed in North (4 species) and South Cornelia (4 species) were
lower than the number of species observed during 2013 through 2016 (Figure 5-6 and Figure 5-8).
Similarly, FQI was lower in 2017 than 2013 through 2016 for North (8.0) and South Cornelia (8.0)
(Figure 5-7 and Figure 5-9). In June 2017, 5 species were absent from North Cornelia that had been
observed in June 2016. Similarly, in the June 2017 survey, nine species were absent in South Cornelia that
had been observed in June 2016. The species that were not observed between June 2016 and June 2017
in North and South Cornelia include curly-leaf pondweed, targeted by the herbicide treatment, three
native pondweed species (Potamogeton foliosus, Potamogeton nodosus, and Stuckenia pectinata), four
native bulrush species (Bulboschoenus fluviatilis, Schoenoplectus acutus, Schoenoplectus tabernaemontani,
and Schoenoplectus sp.), muskgrass (Chara sp.), and coontail (Ceratophyllum demersum). It is possible that
the later treatment of curly-leaf pondweed that occurred in May 2017 effected the native plant species of
North and South Cornelia. For future curly-leaf pondweed herbicide treatments, it is recommended that
the herbicide be applied before the lake’s average water column temperature reaches 60°F, prior to the
native plant growing season. Completing the herbicide application prior to the start of the native plant
growing season should protect the native plants.
5.3.1.2 Phytoplankton
Blue-green algae numbers were lower during 2004 through 2013 than 2015 through 2016. Changes in the
Lake Cornelia phytoplankton community since 2004 are hypothesized to have been influenced to some
extent by the curly-leaf pondweed within the lake. Curly-leaf pondweed was not observed in 2004, was
first observed in 2008, and remained at low levels during 2008 through 2013. A rapid increase in curly-leaf
pondweed extent during 2015 may have resulted in a rapid increase in phytoplankton growth, especially
blue-green algae. In 2016, the lake’s curly-leaf pondweed infestation was more severe than 2015 and the
resultant blue-green algal bloom in 2016 was more severe than the lake’s 2015 blue-green algal bloom.
The Lake Cornelia phytoplankton community during 2013 through 2018 was impacted by algaecide
treatments. The City of Edina started treating Lake Cornelia with copper sulfate to control algal
37
populations in 2013. Table 5-2 shows the approximate dates of the algal treatment efforts based on past
records since 2013.
Table 5-2 Lake Cornelia algal copper sulfate treatments
Algal Copper Sulfate Treatments
July 19, 2013
August 21, 2013
June 18, 2014
July 25, 2014
August 18, 2015
August 3, 2016
August 9, 2017
September 7, 2017
July 11, 2018
The changes to the phytoplankton species composition and abundance from 2004 through 2017 can be
seen in Figure 5-11 for North Cornelia and Figure 5-12 for South Cornelia.
Figure 5-11 North Cornelia Phytoplankton Data Summary (2004-2017)
38
Figure 5-12 South Cornelia Phytoplankton Data Summary (2004-2017)
Blue-green algae are associated with water quality problems and can be a source of health concerns. The
World Health Organization (WHO) has established the following guidelines for assessing the risk posed to
lake users by exposure to blue-green algae (World Health Organization, 2003):
• No Risk: Lakes with blue-green algae densities less than 20,000 cells per milliliter pose no risk to
the health of humans or pets.
• Low Risk: Exposure to lakes with blue-green algae density levels between 20,000 and
100,000 cells per milliliter poses a low risk of adverse health impacts (i.e., skin irritation or
allergenic effects such as watery eyes).
• Moderate Risk: Exposure to lakes with blue-green algae densities greater than 100,000 cells per
milliliter poses a moderate health risk (i.e., long-term illness from algal toxins is possible).
The WHO guidelines were applied to the data observed in North and South Cornelia. During the late
summer periods of 2015 and 2016 when higher numbers of blue-green algae were present, comparison
with the WHO guidelines indicated a moderate risk of adverse health effects from exposure to blue-green
algae in North and South Lake Cornelia (Figure 5-13 and Figure 5-14). Low risk of adverse health effects
from exposure to blue green algae was also present earlier in the growing season for 2013, 2015, and
2017 in both North and South Cornelia when lower numbers of blue-green algae were observed
39
Figure 5-13 North Cornelia blue-green algae data compared with the World Health Organization’s Risk of Adverse Health Effects Guidelines
Figure 5-14 South Cornelia blue-green algae data compared with the World Health Organization’s Risk of Adverse Health Effects Guidelines
40
The City of Edina performed an algaecide chemical treatment of Lake Cornelia on August 3, 2016 to
manage the lake’s algal bloom. Although the numbers of blue-green algae were reduced by the
treatment, the District and City were concerned that algal toxins could still be present in the lake and pose
health risks to lake users. The District collected algal toxin samples on September 7 and September 21,
2016 and verified the presence of high levels of algal toxin (Microcystin) that exceeded the public health
advisory level.
In August 2017, blue-green algal scum was observed in both North and South Cornelia and was
sampled/tested for three algal toxins. The tests verified high levels of two algal toxins, microcystins and
anatoxin-a, both exceeding public health advisory levels.
5.3.1.3 Zooplankton
Zooplankton are microscopic animals that are a source of food for fish (e.g., bluegills, crappies). They were
sampled 8 times in 2008 and five times during 2013, 2015, and 2016. During 2008 through 2016, all three
groups of zooplankton (rotifers, copepods, and cladocerans) were generally present in Lake Cornelia. The
relative abundance of each group during each sampling period can be viewed in Figure 5-15 for North
Cornelia and Figure 5-16 for South Cornelia. Small rotifers and copepods and smaller cladocerans
dominated the community; because they do not graze as heavily on algae as the larger cladocerans, they
generally have limited impact on the lake’s water quality.
The data suggest that changes in the Lake Cornelia zooplankton community in 2015 may have been
influenced by the rapid increase in blue-green algae. Blue-green algae are a poor food source for
zooplankton. In addition, smaller-bodied zooplankton species are unable to consume the larger-sized
blue-green algae. The total number of zooplankton sampled in 2015 were much lower than 2008 and
2013. Cladocerans were consistently present throughout 2008 and 2013, but were absent during June and
July of 2015 and present at very low numbers during August and September. A few large-bodied
Cladocerans were present in 2008 and 2013, but only small-bodied cladocerans were present in 2015. It is
difficult to discern between the impact of fish predation and the impact of changes in food source on the
lake’s zooplankton. It is likely that the changes in the 2015 zooplankton community were influenced by
increased blue-green algae species.
In 2016, rapid increases in zooplankton, especially cladocerans, occurred following treatment of Lake
Cornelia with algaecide in early August. The algaecide reduced blue-green algae in the lake. The data
suggest the reduction in blue-greens was followed by rapid increases in zooplankton numbers.
Cladoceran numbers increased more than copepods and rotifers.
41
Figure 5-15 North Cornelia Zooplankton Data Summary (2008-2015)
42
Figure 5-16 South Cornelia Zooplankton Data Summary (2008-2015)
5.3.1.4 Fish
To gain a more complete understanding of the ecology of Lake Cornelia, the District commissioned a fish
survey in 2018 with the primary intent to quantify the common carp population (Maxwell (RPBCWD),
2018). The full survey report is included as Appendix B.
Fish caught in fyke nets set in North and South Lake Cornelia were generally small pan fish with black
bullhead being very abundant. Nearly all of the fish (see Table 5-3 and Table 5-4) were less than 8 inches
suggesting that periodic winter kill is preventing fish from maturing to larger size classes. These fish are
planktivores (they eat zooplankton) and benthivorous (bottom, sediment feeders), the black bullhead in
particular. The zooplankton data collected in 2015 suggest that the Cladocera population may have been
heavily grazed upon by these small fish. It is likely that the fish population in Lake Cornelia is having an
adverse effect on water quality. Zooplankton eat phytoplankton, and if the zooplankton population is
reduced they will be less able to reduce the phytoplankton population, which in turn affects lake clarity.
The black bullhead are benthivorous and in the process of feeding the large population of bullheads
could cause increased sediment resuspension, turbidity, and the transport of phosphorus from lake
bottom sediments into the lake.
43
Very few common carp were found in Lake Cornelia with electrofishing, which is the preferred approach
to quantify carp populations. Conversely, a significant number of goldfish were caught while electrofishing
in Lake Cornelia and in upstream detention ponds. Goldfish are omnivores and are not strictly
benthivorous, and hence they may not have the same effect on water quality as common carp. However,
they could be significant zooplankton grazers. Electrofishing was also conducted in Swimming Pool Pond,
Point of France Pond, and the waterbodies north of Highway 62 known as Lake Otto and Lake Nancy. A
significant number of goldfish were identified in Swimming Pool Pond and connected waterbodies north
of Trunk Highway (TH) 62. Point of France Pond had a high number of common carp with an estimated
biomass of 196 pound per acre. To manage the goldfish and common carp, it will be important to better
understand the movement of the fish between these ponds and Lake Cornelia.
Overall the fish sampled in the Lake Cornelia system were small in size and species richness was limited.
This is most likely a result of the 2017-2018 winterkill and past winterkills that have occurred. The low
number of bluegill and other sunfish species captured from the surveys reflect a limited population that
may not be able to control common carp and goldfish recruitment effectively. The frequency of winterkills
and the availability of connected shallow waterbodies that winterkill which act as nurseries, are most likely
preventing bluegills from effectively controlling carp and goldfish within the system.
It should also be noted that the MDNR, as part of its Fishing in the Neighborhood (FiN) program, has
been stocking Lake Cornelia on a near annual basis (stocking records provided on LakeFinder up to the
year 2014) with bluegill, black crappie, hybrid sunfish, and pumpkinseed sunfish.
Table 5-3 South Cornelia fyke net results showing the number of fish caught by size
Species Number of Fish Caught in Each Size Category (inches body length)
0-5 6-8 9-11 12-14 15-19 20-24 25-29 30+ Total Fish/Net
Black bullhead 75 82 1 159 53
Bluegill sunfish 1 2 3 1
Goldfish 16 16 5.3
Golden shiner 16 1 17 5.7
44
Table 5-4 North Cornelia fyke net results showing the number of fish caught by size
Species Number of Fish Caught in Each Size Category (inches body length)
0-5 6-8 9-11 12-14 15-19 20-24 25-29 30+ Total Fish/Net
Black bullhead 148 161 1 676 225
Back crappie 2 2 0.67
Bluegill sunfish 31 31 10
Common carp 1 1 0.33
Goldfish 9 9 3
Golden shiner 63 23 90 30
Green sunfish 20 20
Hybrid sunfish 12 12
Pumpkinseed 12 12
5.3.2 Lake Edina
5.3.2.1 Aquatic Plants
As mentioned for Lake Cornelia, macrophytes (aquatic plants) are an important part of a shallow lake
ecosystem and provide critical habitat for aquatic insects and fish. The MDNR developed a Lake Plant
Eutrophication IBI to assist the MPCA with determining lake impairment based on the plant community.
Lake Edina plant survey data from 2004 through 2017 were assessed to determine species richness (the
number of species) and FQI scores. The scores were compared with MDNR Lake Plant Eutrophication IBI
impairment thresholds (a minimum of 11 species and an FQI score of at least 17.8) to determine whether
the Lake Edina plant community would be considered impaired.
As shown in Figure 5-17 and Figure 5-18, the Lake Edina plant community has generally failed to meet the
MDNR Lake Plant Eutrophication IBI criteria since 2004, a reflection of the lake’s poor water quality. The
2017 plant community in Lake Edina was poor, but had improved from previous years. The number of
plant species observed in 2017 was higher than the number of species observed during 2008 through
2015—6 to 8 in 2017 compared with 3 to 4 in previous years. Similarly, Figure 5-18 shows that FQI was
higher (better) in 2017 than 2008 through 2015. The plant IBI has not yet been used by the MPCA/MDNR
to determine impairment. However, it is expected to eventually be used as an assessment tool to
determine biological impairment.
45
Figure 5-17 Lake Edina Species Richness Compared with Plant IBI Threshold for Species Richness
Figure 5-18 Lake Edina Floristic Quality Index (FQI) Compared with Plant IBI Threshold for FQI
Three non-native aquatic invasive species (AIS) are present in Lake Edina: purple loosestrife, curly-leaf
pondweed, and Eurasian watermilfoil.
Purple loosestrife was observed along the perimeter of the lake during the 2008, 2012, 2015, and 2017
sampling periods.
46
Curly-leaf pondweed has been observed at low levels in Lake Edina since 2008. In 2008, curly-leaf
pondweed was found at a single location in the southeast area of the lake. In 2012 and 2015, curly-leaf
pondweed was observed at a single location in the central western area of the lake. In June of 2017, the
species was observed at two locations, both in the western area of the lake. In August, it was observed at
a single location in the central western area of the lake.
Although Eurasian watermilfoil was first observed in Lake Edina during 2017, it was widespread and
increased in extent between June and August. Unlike many other plants, Eurasian watermilfoil does not
rely on seed for reproduction. Its seeds germinate poorly under natural conditions and it generally
reproduces by fragmentation—each fragment can grow into a new plant. The plant produces fragments
after fruiting at least once or twice during the summer. These fragments can be carried downstream by
water currents or spread by waves or boaters throughout a waterbody (WI-DNR, Eurasian Watermilfoil -
Beaver Dam Lake, 2012). Eurasian watermilfoil fragments in Lake Edina could be carried downstream to
Normandale Lake.
Once established in an aquatic community, Eurasian watermilfoil reproduces from fragments and stolons
(runners that creep along the lake bed). Stolons, lower stems, and roots persist over winter and store the
carbohydrates that help Eurasian watermilfoil claim the water column early in spring, photosynthesize,
divide, and form a dense leaf canopy that shades out native aquatic plants. Eurasian watermilfoil’s fast
growth rate (up to 2 inches per day in spring and summer), its ability to spread rapidly by fragmentation,
and its ability to effectively block out sunlight needed for native plant growth often results in monotypic
stands. Monotypic stands of Eurasian watermilfoil provide only a single habitat, and threaten the integrity
of aquatic communities in a number of ways. For example, dense stands disrupt predator-prey
relationships by fencing out larger fish, and reducing the number of nutrient-rich native plants available
for waterfowl. Eurasian watermilfoil spreads rapidly and can grow to dominance in as little as 2 years (WI-
DNR, Eurasian Watermilfoil - Beaver Dam Lake, 2012; WI-DNR, Aquatic Plant Eurasian Watermilfoil, 2012).
5.3.2.2 Phytoplankton
The Lake Edina phytoplankton community has generally reflected the lake’s poor water quality. During
2008 through 2015, average phytoplankton numbers increased annually and blue-greens dominated the
phytoplankton community (Figure 5-19). In 2017, phytoplankton numbers declined and the lake’s average
2017 phytoplankton number was the lowest on record. The composition of the 2017 phytoplankton
community was different from previous years. Green algae dominated the phytoplankton community and
numbers of blue-greens were very low.
The change in numbers of phytoplankton observed in 2017 reflects a change from planktonic algae within
the water column to mats of filamentous algae residing on the lake’s surface. Algal mats were observed
during both the June 13, 2017 plant survey and the August 21, 2017 water quality monitoring event.
Surface algal mats outcompeted planktonic algae that reside in the water column by consuming nutrients
and creating shade which limited light beneath the lake’s surface. Samples from the algal mats were
collected and analyzed in the laboratory to determine whether they were harmful blue-greens or harmless
47
greens. The algal species comprising the mats in the lake were green algal species—Rhizoclonium in June
and Spirogrya in both June and August.
Figure 5-19 Lake Edina Phytoplankton Data Summary (2008-2017)
It is hypothesized that the changes to the Lake Edina phytoplankton community since 2008 are partially
due to treatment efforts. The City of Edina started treating Lake Edina with copper sulfate to control algal
populations in 2013. Table 5-5 shows the approximate dates of the algal treatments based on past
records since 2013.
Table 5-5 Lake Edina algal copper sulfate treatments
Algal Copper Sulfate Treatments
August 6, 2013
August 21, 2013
August 18, 2014
July 16, 2015
August 18, 2015
August 3, 2016
July 25, 2017
August 23, 2017
June 8, 2018
48
Blue-green algae are associated with water quality problems and can be a source of health concerns. The
World Health Organization (WHO) has established guidelines for assessing the risk posed by exposure to
blue-green algae (World Health Organization, 2003). Blue-green algae numbers observed in Lake Edina
were compared with WHO guidelines to assess risk of adverse health impacts from the blue-greens
(Figure 5-20). During mid-July 2015, comparison with the WHO guidelines indicated a moderate risk of
adverse health effects from exposure to the blue-green algae species in Lake Edina based on the number
of blue-green algae observed. Algaecide treatments were then applied in July and August to reduce the
numbers of blue-green algae and the associated health risks. Users also had a low risk of adverse health
effects from exposure to blue-green algae species in Lake Edina in 2008, 2012, and 2015. The blue-green
algal numbers were low throughout 2017 and, according to WHO criteria, posed no risk to lake users
during this year.
Figure 5-20 Lake Edina Blue-green Algae Data Comparison with World Health Organization Risks Guidelines
5.3.2.3 Zooplankton
Zooplankton are microscopic animals that are a source of food for fish (e.g., bluegills, crappies). During
2008 through 2017, all three groups of zooplankton (rotifers, copepods, and cladocerans) were generally
present in Lake Edina (Figure 5-21). Small rotifers and copepods and smaller cladocerans dominated the
community; because they do not graze as heavily on algae as the larger cladocerans, they generally have
limited impact on the lake’s water quality.
49
The data suggest that changes in the Lake Edina zooplankton community in 2015 and 2017 may have
been influenced by changes in numbers of blue-green algae (Figure 5-22). The increase in blue-green
algae during 2015 (from 47,757 units/ mL in 2012 to 79,914 units/mL in 2015) was associated with a
decline in cladocerans (from 92,105 per m2 in 2012 to 3,678 per m2 in 2015). A few large-bodied
Cladocerans were present in 2008 and 2012, but only small bodied cladocerans were present in 2015.
Blue-green algae are a poor food source for zooplankton. In addition, smaller-bodied zooplankters are
unable to consume the larger sized blue-greens.
The decline in blue-green algae in 2017 (from 79,914 units/mL in 2015 to 3,515 units/mL in 2017) was
associated with an increase in cladocerans (from 3,678 per m2 in 2015 to 198,864 per m2 in 2017). In
addition, some large-bodied cladocerans were again observed in the lake during 2017. The phytoplankton
community in 2017 was dominated by green algae, a good food source for cladocerans. We are unable to
discern between the impact of fish predation and the impact of changes in food source on the lake’s
zooplankton. If we assume the fish community remained stable, the changes in the 2015 and 2017
cladoceran numbers would likely be due to changes in numbers of blue-green algae within the lake.
Figure 5-21 Lake Edina Zooplankton Data Summary (2008-2017)
50
Figure 5-22 Lake Edina average annual cladoceran and blue-green algae numbers comparison
5.3.2.4 Fish
Fisheries information was not available for Lake Edina.
51
6.0 Water Quality Modeling for the UAA
Phosphorus levels in Lake Cornelia and Lake Edina are high, and will continue to be greatly affected by the
amount of phosphorus loading received. For this UAA study, water and phosphorus loading to Lake
Cornelia and Edina were evaluated for a 3-year period (2015, 2016, and 2017). Modeling was used to link
water and phosphorus loading to Lake Cornelia and Lake Edina to observed phosphorus concentrations in
the water column of these lakes. Somewhat unique to this study was the inclusion of internal lake
processes such as phosphorus release from lake sediments (internal loads), curly-leaf die-off, copper
sulfate treatments, and potential effects of benthivorous fish (included as part of internal loading). Model
years 2015, 2016, and 2017 were typical of the variability that these lakes may experience from a
watershed loading, in-lake management, and biological variability perspective.
6.1 P8 Model Runoff and Phosphorus Loading
Central to a lake water quality analysis is the use of a water quality model that has the capacity to predict
the amount of runoff and pollutants that reach a lake via stormwater runoff (external loading). The P8
(Program for Predicting Polluting Particle Passage through Pits, Puddles and Ponds) modeling software
was used to estimate watershed loads to the lakes (I.E.P, Inc., 1990). The P8 model incorporates hourly
precipitation and daily temperature data; long-term climatic data can be used so that watersheds and
BMPs can be evaluated for varying hydrologic conditions. The P8 model was used to calculate the daily
water volume and phosphorus loads introduced from each tributary subwatershed in the Lake Cornelia
and Lake Edina watersheds. P8 model inputs included:
• Climate Data: hourly precipitation and daily temperature (source: National Weather Service gage
at Minneapolis-St. Paul International Airport, MSP)
• Watershed: storm sewer network, tributary land areas (both pervious and impervious)
• Best management practices: ponds, including the water storage and solids and phosphorus
settling functionalities
The P8 model was run for the Lake Cornelia and Lake Edina watersheds for water years 2015, 2016, and
2017.
Since no data has been collected regarding the inflow water quantity or quality for Lake Cornelia or Lake
Edina on a subwatershed-scale, detailed calibration of the P8 model was not feasible. The P8 model
outputs, used as inputs for the in-lake models (described below) is thought to be best-suited for
considering relative changes in loading under varying watershed conditions.
6.2 Water Balance Calibration
6.2.1 Precipitation and Runoff
The annual water and watershed phosphorus loadings to Lake Cornelia and Lake Edina under existing
land use conditions were estimated for three different water years, each having distinctly different in-lake
52
phosphorus concentrations, water clarity, and biota (e.g., type of phytoplankton, zooplankton, and aquatic
plants). The precipitation totals during the three modeled water years are summarized in Table 6-1. Of the
three modeled water years, 2016 had the greatest amount of precipitation for the entire water year and
for the growing season (May 1 through September 30). Water year 2015 had the lowest amount of
precipitation when looking at the entire water year. Water year 2015 was also a year with lower-than-
average snowpack, which yielded less snowmelt runoff than typical years and resulted in significant
reductions of water levels in upstream stormwater ponds and within the lakes. Water levels were so low
on Lake Edina in spring 2015 that portions of the lake essentially became a mudded, open area
(Figure 6-1). By mid-June the water levels in Lake Edina rebounded due to heavier precipitation events
that occurred throughout May and June. Water year 2017 had the lowest amount of precipitation
throughout the growing season of the 3 years analyzed in this study (approximately 3 inches less than
2015 and 5.4 inches less than 2016).
Annually and seasonally varying climatic conditions affect watershed runoff and subsequently lake volume
and hydraulic residence time (lake volume divided by flow through the lake). In some cases where
phosphorus in stormwater is greater than phosphorus in the receiving lake, a shorter hydraulic residence
time is associated with higher in-lake phosphorus. Hence, loading and flushing (e.g., hydraulic residence
time) can have an effect on annual and seasonal phosphorus concentrations in the lake.
Table 6-1 Precipitation amounts for 2015, 2016, and 2017
Model Year Water Year (Oct 1 through Sept 30)
Precipitation (inches)
Growing Season (May 1 through Sept 30)
Precipitation (inches)
2015 30.2 22.9
2016 41.2 25.3
2017 35.2 19.9
53
Figure 6-1 Photo taken of Lake Edina low water levels in April 2015
6.2.2 Stormwater Volume Calibration (Water Balance)
The changes in water volumes of the lakes over time were calibrated by matching the modeled surface
elevations of North and South Cornelia and Lake Edina to monitored data during the period of October
2014 through September 2017. To translate the water loadings into water surface elevations, a water
balance model was utilized. The model uses estimated daily watershed runoff inflows (predicted by the
P8 model), daily precipitation, daily evaporation, daily discharge (estimated with outlet rating curves),
estimated groundwater inflow or outflow, and observed lake levels to estimate changes in the water levels
of the lakes. The South Cornelia water balance model also included the daily discharge predicted by the
North Cornelia water balance model. Similarly, the Lake Edina water balance model included the daily
discharge predicted by the South Cornelia water balance model.
Figure 6-2 shows an example of the water balance calibration that was completed for North Cornelia for
model year 2015. Appendix C displays the water balance calibrations for North and South Cornelia and
Lake Edina for each modeled year. The predicted water levels, shown by the green line on the plots, were
calibrated to match as closely as possible to the observed monthly water levels, indicated by the blue
diamonds. Groundwater outflow (red line) was used to calibrate the modeled water surface elevations to
the observed data. Groundwater outflow was used as a calibration parameter in the fall, winter, and early
spring periods of the 2015 and 2016 North and South Cornelia water balance models. The water balance
modeling for 2015, 2016, and 2017 included ground water interactions in Lake Edina. Previous studies
54
have indicated that the water levels in Lake Edina are impacted by groundwater interactions with
downstream Nine Mile Creek (Barr Engineering Co., 2015).
Figure 6-2 North Cornelia (2015) Water Balance
In 2017, continuous observed water level data was available for Lake Edina from May 28, 2017 through
October 1, 2017. From this continuous dataset, average daily water levels were calculated in order to
compare the monitored data to the predicted daily water levels from the in-lake model. The daily data
developed from the continuous monitored dataset are represented by an orange line in the Lake Edina
2017 plot (Appendix C). As shown on the figure, the modeled water levels of Lake Edina in 2017 match
fairly well with the continuous observed dataset. The timing of the peaks and the slopes of the water level
declines correlate well.
Overall, the water balance calibrations for all three water bodies and for each modeled year correlate well
with the observed monitored data. Table 6-2 provides a summary of the modeled surface runoff water
loads that were simulated for North Cornelia, South Cornelia, and Lake Edina for each modeled year. The
table also provides the modeled lake discharges. Comparing South Cornelia’s direct watershed runoff load
to the water volume discharging from North Cornelia into South Cornelia, one can see that a large portion
of South Cornelia’s water volume is determined by the outflow from North Cornelia. Similarly, comparing
Lake Edina’s direct watershed runoff load to the load entering from South Cornelia, one can see that the
outflow from South Cornelia comprises a large portion of the Lake Edina water volume.
55
Table 6-2 Water balance summary of watershed runoff inflows and discharges
Water Body Modeled
Parameter
Model Year 2015
(acre-feet)
Model Year 2016
(acre-feet)
Model Year 2017
(acre-feet)
North Cornelia Watershed Runoff 750 1,050 925
Outlet Discharge 640 1,041 953
South Cornelia Watershed Runoff 38 54 48
Outlet Discharge 600 1,122 1,019
Lake Edina Watershed Runoff 187 262 230
Outlet Discharge 570 1,186 1,189
6.3 In-Lake Phosphorus Modeling
The focus of the in-lake phosphorus modeling effort was to assess phosphorus inputs and outputs and
estimate the in-lake concentration changes of phosphorus in North and South Lake Cornelia and Lake
Edina over time. Once the in-lake phosphorus dynamics under existing conditions are understood and
quantified, the effectiveness of phosphorus reduction approaches can be evaluated using the model.
From a very simplistic basis, the concentration of phosphorus in a lake is the mass of phosphorus in the
lake divided by the lake volume. However, the mass of phosphorus in the lake and the volume of the lake
are consistently changing over time. The water balance modeling allowed for tracking water volume
changes of the lakes. An in-lake phosphorus mass balance model tracks the movement of phosphorus
into and out of the lake over time.
There are several processes that dynamically increase or reduce the concentration of phosphorus in the
lake water column, including (the “-“ or “+” indicates that the mechanism either reduces or increases
phosphorus):
• Atmospheric Deposition (+): Phosphorus deposits into the water body from the atmosphere
• Settling (-): Phosphorus in phytoplankton and attached to particles settles out of the lake water
column to the sediments.
• Flushing (-): Phosphorus that leaves through a lake outlet.
• Internal Sediment Loading (+): Phosphorus from lake bottom sediments may release into the
water column during low oxygen conditions.
• Copper sulfate treatment (- and +): In the short term, phytoplankton die and begin to settle
and the phosphorus in the phytoplankton cells also settles. However, the dead phytoplankton
cells decay over time and release phosphorus. When the phytoplankton decay they can also
consume oxygen which then leads to increased internal loading from the sediment.
• Benthivorous (bottom-feeding) fish (+): Although not modeled as a separate internal load,
benthivorous fish are presumed to cause additional internal phosphorus loading during certain
modeling periods due to stirring of the bottom sediments
• Curly-leaf pondweed die-off and decay (+): Phosphorus in the plant tissue is released into the
water column when curly-leaf pondweed dies and decays.
56
The model integrates these phosphorus loads and losses as part of a daily time-step used in a finite
difference approach. Each of these processes occur during different periods and hence they are quantified
(e.g., calibrated) by matching the in-lake phosphorus concentration with the field-measured phosphorus
concentration. Additional detail is provided below for several of these processes.
6.3.1 Atmospheric Deposition
An atmospheric wet and dry deposition rate of 0.42 kg/ha/year was applied to the surface area of the lake
to determine annual phosphorus loading. An annual total phosphorus load from atmospheric deposition
of approximately 4-5 pounds was estimated for North Cornelia for each modeling year. For South
Cornelia, the annual total phosphorus load from atmospheric deposition was approximately 6 pounds.
The estimated annual total phosphorus load from atmospheric deposition in Lake Edina was
approximately 4-5 pounds.
6.3.2 Settling and Copper Sulfate
Phosphorus attached to particles or incorporated into phytoplankton were given a constant settling rate
expressed in meters per day. For periods when documented application of copper sulfate occurred, this
settling rate was increased to account for a period of enhanced phytoplankton death and settling.
Following the copper sulfate settling period, the settling rate was returned to the constant settling rate
used under “normal” conditions. For both Lake Cornelia and Lake Edina, the settling rate was used to
assist with model calibration (e.g., match the model-predicted and observed phosphorus concentrations).
6.3.3 Internal Sediment Loading and Benthivorous Fish
In shallow lakes it is challenging to identify the extent of internal loading by just evaluating the in-lake
monitoring data. Internal phosphorus loading from the sediments can occur when oxygen is depleted
during microbial decomposition. When oxygen is depleted and can no longer be used by microbes as an
electron acceptor, microbes will start to use other compounds as electron acceptors. Specific species of
microbes can use iron in the sediments as electron acceptors and in doing so, will release the phosphate
bound to the iron. This process is complex and is affected by numerous factors including the species of
microbes (over space and time), lake stratification, water temperature, sediment composition, and the
amount and type of organic matter, just to name a few. It is difficult to measure all of the factors that are
involved in internal phosphorus release from the sediments. Therefore, the available monitored data,
along with model equations were used to estimate the internal loading potentials of the lakes in this UAA
process.
Internal phosphorus release for Lake Cornelia was estimated based upon the maximum potential
phosphorus release rate computed from lake sediment cores. The rate of phosphorus release was then
modified depending upon the oxygen concentration in the water column. This was accomplished by using
a Michaelis-Menten equation to modify the phosphorus release rate, V0, (i.e., the internal loading rate) in
accordance with the oxygen concentration in the lake water column. The equation is as follows:
𝑉𝑉0 = 𝑉𝑉𝑚𝑚𝑚𝑚𝑚𝑚1 −[𝐷𝐷0]𝐾𝐾𝑀𝑀+[𝐷𝐷𝐷𝐷]
57
where Vmax is the maximum potential phosphorus release rate, [DO] is the measured concentration of
dissolved oxygen, and KM is the Michaelis-Menten constant. KM indicates the dissolved oxygen
concentration when the total phosphorus release rate is equal to one half the maximum potential
phosphorus release rate. Vmax was estimated for each lake based on sediment core data. For the majority
of the modeled periods, measured DO concentrations were used in the Michealis-Menten equation;
however, for certain periods (mostly during late-summer and early-fall), adjustments were made to the
DO concentrations to calibrate the model to observed in-lake phosphorus concentrations. DO
concentrations were measured in the lakes on approximately a monthly basis and all measurements
occurred during the day. Since microbial activity and phytoplankton respiration, which influence the rate
of oxygen consumption and production, can change dramatically over time and space, monthly
measurements in one location in the lake can only provide a glimpse into a lake’s internal loading
potential. Continuous monitoring of DO concentrations in multiple locations of a lake would be needed to
better quantify the internal loading potential. However, the costs and logistics of completing this effort
may be significant.
Section 5.2 discusses the internal loading potential of North and South Cornelia and Lake Edina based on
sediment core analyses. As mentioned previously, of the three lakes, North Cornelia has the greatest
internal loading potential with a maximum internal loading rate equal to 7.6 mg of phosphorus per meter-
squared per day. South Cornelia was found to have a maximum internal phosphorus loading rate equal to
1.5 mg P/m2/d. The concentration of mobile phosphorus in Lake Edina sediment was at background levels
and based upon an equation developed by Pilgrim et. al. (2007), it is expected that the internal release
rate is near zero.
The effect of benthivorous (bottom-feeding) fish was not explicitly included in the models, however, it was
implicitly included in the internal loading rate, as well as the settling rate. More benthic disturbance
equates to a reduced overall settling rate and enhanced internal sediment loading through an increased
maximum potential phosphorus release rate (Vmax). The effect of benthivorous fish was included in the in-
lake modeling process for South Cornelia for water years 2016 and 2017 by modifying the maximum
internal loading rate during specific periods in the model to calibrate to observed in-lake total
phosphorus concentrations. This is discussed further in Section 6.3.5.
6.3.4 Curlyleaf Pondweed Die Off and Decay (Lake Cornelia Only)
The concentration of phosphorus in curly-leaf pondweed is on average 4.3 grams per kilogram of dry
plant material (measured based on samples from Kohlman Lake in Maplewood, Minnesota). When curly-
leaf dies and decomposes, phosphorus is released from the plant tissue, adding phosphorus to the water
column over time. Curly-leaf pondweed typically begins to die-off in June. Like most biological
populations, die-off of the entire curly-leaf pondweed population does not happen all at once but rather
occurs over time. Model calibration using curly-leaf pondweed die-off and decay was necessary to match
observed and modeled data in the summer. The mortality rate was estimated to be 10 to 15 percent of
the population per day. As the remaining population gets smaller, so too does the mass of plant tissue
that dies until there is only a very small population remaining. A total of 10 to 15 percent of the dead
pondweed population was assumed to decompose each day. This approach was used to calibrate the
58
North Cornelia in-lake models for 2015 and 2017 and for South Cornelia in-lake models for 2015, 2016,
and 2017.
6.3.5 In-Lake Water Quality (Phosphorus) Model Calibration
Calibration is a process in which model parameters and coefficients are reasonably adjusted such that the
model predictions are similar to in-lake measurements. Results of the in-lake phosphorus model
calibrations for North Cornelia are provided in Figure 6-3, Figure 6-4, and Figure 6-5 for modeled years
2015, 2016, and 2017, respectively. The results of the model calibrations for South Cornelia are presented
in Figure 6-6, Figure 6-7, and Figure 6-8 for 2015, 2016, and 2017, respectively. Lake Edina was calibrated
for the 2 years with available monitoring data (2015 and 2017) and the results of those model calibrations
are presented in Figure 6-9 and Figure 6-11, respectively.
An in-lake model was also developed for Lake Edina for 2016 after completion of the two calibration
models. The Lake Edina calibration models showed that approximately 99% of the in-lake phosphorus
concentrations are influenced from upstream discharges from Lake Cornelia, direct watershed runoff
contributions, and atmospheric deposition. Less than 1% of the in-lake phosphorus contributions were
due to internal loading. Since the North and South Cornelia models were calibrated for 2016 and there
was enough data to model Lake Edina watershed runoff in 2016, an in-lake model was also developed to
represent Lake Edina existing in-lake conditions for 2016. Those results are presented in Figure 6-10.
For all plots, the blue line represents the modeled in-lake phosphorus concentrations of the water bodies.
The pink squares represent the monitored phosphorus concentrations. On the secondary axes, the relative
phosphorus loads from the watershed, internal sediment load, and curly-leaf pond weed death/decay are
shown.
Calibration was challenging for North and South Cornelia with several highly-variable and sometimes
human-induced loadings or losses such as variable internal sediment loading and curly-leaf die-off and
decay (where start times of decay were influenced by herbicide treatments depending on the modeled
year), copper sulfate treatments, and benthivorous fish feeding influences. However, calibration was
conducted using measured data (e.g., phosphorus concentration in sediments, in-lake water column
phosphorus concentrations, in-lake dissolved oxygen, and temperature) and consistent modeling
approaches and coefficients throughout the modeling period and between the different model years.
An important calibration parameter to note is that curly-leaf pondweed loading was not included for
North Cornelia for 2016. While macrophyte surveys did show curly-leaf pond weed growth in North
Cornelia in 2016, observed in-lake phosphorus concentrations did not indicate a lake response to curly-
leaf death and decay, as was evident in other modeled years. One explanation is that there were a few
large storm events in 2016 at the typical time for curly-leaf pondweed die-off. Larger flows could have
carried some of the curly-leaf pondweed plants downstream to South Cornelia. However, the exact cause
for the missing in-lake response to curly-leaf pondweed die-off and decay for 2016 in North Cornelia is
unknown.
59
Furthermore, it is important to note that a calibration parameter was introduced for South Cornelia for
modeling years 2016 and 2017 that was not introduced for the other lakes or modeling periods. From
mid-August through the end of September, a bioturbation factor was included for South Cornelia. The
monitored phosphorus concentrations of South Cornelia during these periods showed high phosphorus
loadings which could not be explained by internal loading from lake sediment, curly-leaf pondweed-, or
external loading. As mentioned previously, the effect of benthivorous fish was not explicitly included in
the model, however, it was implicitly included in the internal loading rate. More benthic disturbance
equates to enhanced internal sediment loading through an increased maximum potential phosphorus
release rate (Vmax). During discussions with Dr. Prezemyslaw Bajer from the Department of Fisheries,
Wildlife, and Conservation Biology at University of Minnesota, it was discovered that enhanced internal
loading from benthivorous fish disturbances in August and September have been noted in other water
bodies. Bajer’s experience with carp baiting experiments suggest that carp forage quite actively in August
through September when [the male fish] re-grow their reproductive organs for the next year. Additionally,
Bajer mentioned that there is evidence in literature that carp’s digging in the bottom sediments can
increase during periods when the availability of benthic invertebrates decreases; however, the exact
seasonal cycles of benthic invertebrate availability is unknown for South Cornelia. Furthermore, Bajer
noted that some carp migrate out of lakes from May through June and migrate back in late summer. The
migration patterns of carp and other fish species in Lake Cornelia are unclear. As mentioned in
Section 5.3.1.4, fish surveys of North and South Cornelia and upstream detention ponds were taken in
summer 2018. While the fish surveys provided a summary of the fish species located within the monitored
areas, the surveys did not address fish migration patterns and how migration patterns can effect lake
water quality over time. To manage the Lake Cornelia fisheries, it will be important to better understand
the movement of the fish between the upstream ponds and North and South Cornelia.
The in-lake model calibration for Lake Edina was less complicated compared to Lake Cornelia. The major
phosphorus inputs included discharge from Lake Cornelia, runoff from the direct Lake Edina watershed,
and atmospheric deposition. The primary calibration parameter was the rate of phosphorus settling. The
settling rate for 2015 and 2017 (the two calibration years) was the same for both Lake Edina calibration
models except for periods when algal copper sulfate treatments were applied to the lake and a notable
change in settling rate was apparent from the observed in-lake phosphorus concentrations. Two copper
sulfate treatments were applied in 2015 (July and August) and two copper sulfate treatments were applied
in 2017 (July and August). Modeled settling rates were increased around these periods such that the
model-predicted and in-lake measured phosphorus concentrations matched. While no observed in-lake
phosphorus concentrations were available for 2016, the in-lake model included a period of increased
settling during the known period of algal copper sulfate treatments from the treatment record. The
copper sulfate treatment settling rate used for 2016 was an average of the 2015 and 2017 modeled
copper sulfate settling rates.
60
Figure 6-3 North Cornelia In-Lake Calibration Model for 2015
Figure 6-4 North Cornelia In-Lake Calibration Model for 2016
61
Figure 6-5 North Cornelia In-Lake Calibration Model for 2017
Figure 6-6 South Cornelia In-Lake Calibration Model for 2015
62
Figure 6-7 South Cornelia In-Lake Calibration Model 2016
Figure 6-8 South Cornelia In-Lake Calibration Model 2017
63
Figure 6-9 Lake Edina In-Lake Calibration Model 2015
Figure 6-10 Lake Edina In-Lake Model 2016
64
Figure 6-11 Lake Edina In-Lake Calibration Model 2017
6.3.6 In-Lake Water Quality (Phosphorus) Model Calibration Loading Summaries
After the in-lake water quality model calibrations were finalized for each calibration year, loading
summaries were developed. Pie-charts were developed to show the relative loading contributions for each
lake and for each modeled year. While each lake has some variability in the phosphorus loading
contributions from year to year, general trends can be noted for the lakes.
Figure 6-12 shows the phosphorus loading summaries for North Cornelia. External watershed loading is a
major contributor of phosphorus to North Cornelia (48% to 76%). In 2017, a lower proportion of the
phosphorus loading came from watershed contributions, which can largely be explained by the reduced
precipitation that fell during the growing season (May 1 through September 30). In 2017, an estimated
precipitation depth of 19.9” fell during the growing season. Comparatively, approximately 22.9” and 25.3”
of precipitation fell during the 2015 and 2016 growing seasons, respectively. Internal sediment loading
and curly-leaf pondweed die-off/decay are also major contributors of phosphorus to North Cornelia. As
mentioned previously, the exact reason for the absence of an in-lake response to curly-leaf pond weed in
2016 is unknown. However, having variations in relative loads will provide a conservative estimate of
management success on lake water quality changes. Water bodies are dynamic over space and time, so
having a range of management outcomes will provide a more realistic outlook on probable lake
responses.
65
Figure 6-13 shows the phosphorus loading summaries for South Cornelia. All three calibration model
years show similar results, with the main contribution of phosphorus to South Cornelia from North
Cornelia (54% to 56% of the total phosphorus load). The second major contribution of phosphorus to
South Cornelia is due to the die-off and decay of curly-leaf pondweed (19% to 23%) and the third is
sediment internal loading (14% to 19%). As noted in Section 6.3.3, the internal sediment loading was
enhanced in 2016 and 2017 due to bioturbation of the sediments from biota activity. The inclusion of
bioturbation in the 2016 and 2017 in-lake models resulted in increased proportions of internal loading
compared to what was observed in 2015. For South Cornelia, direct watershed loading still plays a role in
phosphorus additions to South Cornelia, but to a much smaller extent than the other parameters. The
direct watershed to South Cornelia (omitting open water areas) is approximately 13% the size of the direct
watershed to North Cornelia.
Figure 6-14 shows the phosphorus loading summaries for Lake Edina. The two main sources of the
phosphorus loading to Lake Edina are the upstream lakes (North and South Cornelia) and the direct
watershed runoff. Internal phosphorus loading from sediments is minimal in Lake Edina. Therefore, any
management efforts focused on North and South Cornelia (whether internal or external) will have an
impact on the water quality of Lake Edina.
The loading summaries developed from the calibrated in-lake models provided direction for proposed
management efforts. Section 7.0 describes the management efforts modeled for each lake and Section 8.0
describes the projected in-lake responses.
66
Figure 6-12 Loading Summaries from North Cornelia In-Lake Calibration Models
67
Figure 6-13 Loading Summaries from South Cornelia In-Lake Calibration Models
68
Figure 6-14 Loading Summaries from Lake Edina In-Lake Calibration Models
69
6.4 Modeling Chlorophyll a and Secchi Disc Transparency
The P8 model used for the analysis predicts watershed phosphorus loads to North and South Cornelia and
Lake Edina, and the in-lake model is used to determine water quality in the lake itself. However, the in-
lake model only estimates phosphorus concentrations. To estimate the likely chlorophyll a concentrations
and Secchi disc transparencies, lake-specific regression relationships were developed.
Eleven years of water quality data were available to develop relationships between total phosphorus
concentration (TP), chlorophyll a concentration (Chl a), and Secchi disc transparency (SD) for North
Cornelia. Figure 6-15 depicts the numerical water quality models used to estimate the relationships. For
North Cornelia, the equations are: [𝐶𝐶ℎ𝑙𝑙 𝑎𝑎]=0.4003 [𝑇𝑇𝑇𝑇]−6.3171 R2 = 0.4348 [𝑆𝑆𝐷𝐷]=8.6531[𝑇𝑇𝑇𝑇]−0.614 R2 = 0.4154
where the chlorophyll a and total phosphorus concentrations are in µg/L and Secchi disc depth is in
meters.
Six years of water quality data were available to develop relationships between total phosphorus
concentration (TP), chlorophyll a concentration (Chl a), and Secchi disc transparency (SD) for South
Cornelia. Figure 6-16 depicts the numerical water quality models used to estimate the relationships. For
South Cornelia, the equations are: [𝐶𝐶ℎ𝑙𝑙 𝑎𝑎]=0.3589 [𝑇𝑇𝑇𝑇]+7.8432 R2 = 0.3503 [𝑆𝑆𝐷𝐷]=4.5722[𝑇𝑇𝑇𝑇]−0.55 R2 = 0.5043
where the chlorophyll a and total phosphorus concentrations are in µg/L and Secchi disc depth is in
meters.
Six years of water quality data were available to develop relationships between total phosphorus
concentration (TP), chlorophyll a concentration (Chl a), and Secchi disc transparency (SD) for Lake Edina.
Figure 6-17 depicts the numerical water quality models used to estimate the relationships. For Lake Edina,
the equations are: [𝐶𝐶ℎ𝑙𝑙 𝑎𝑎]=0.3605 [𝑇𝑇𝑇𝑇]−0.2411 R2 = 0.3323 [𝑆𝑆𝐷𝐷]=25.552[𝑇𝑇𝑇𝑇]−0.899 R2 = 0.5841
where the chlorophyll a and total phosphorus concentrations are in µg/L and Secchi disc depth is in
meters.
These equations were subsequently used to give indications of what may be expected with respect to
chlorophyll a and transparency, given the P8/in-lake model results for total phosphorus. It should be
noted that the response of chlorophyll a and Secchi depth to total phosphorus is highly variable. Due to
the high variability, the regression equations can be expected only to allow a general indication of the
lake response to changing total phosphorus concentrations, and the predicted chlorophyll a and Secchi
depth values should not be construed as absolute.
70
Figure 6-15 North Cornelia relationships between total phosphorus, chlorophyll a, and Secchi Disc Transparency
71
Figure 6-16 South Cornelia relationships between total phosphorus, chlorophyll a, and Secchi Disc Transparency
72
Figure 6-17 Lake Edina relationships between total phosphorus, chlorophyll a, and Secchi Disc Transparency
73
7.0 Evaluation of Management Strategies
Numerous in-lake and watershed management strategies were considered to evaluate the potential
improvements in lake water quality and assess the extent to which lake water quality goals will be met (or
progress made) through implementation of the management strategies. Watershed management
strategies are designed to reduce external phosphorus loading to the lakes, whereas in-lake strategies are
designed to reduce internal phosphorus loads and/or improve the health of the aquatic communities. The
P8 watershed models and calibrated in-lake models were used to predict the effectiveness of the
individual in-lake and watershed management strategies, and combinations thereof, in reducing
phosphorus concentrations in Lake Cornelia and Lake Edina, and in turn, improving water quality.
Although not quantified through modeling, the progress toward the goals of a healthy native plant
population and fishery resulting from the evaluated in-lake and watershed management strategies was
also considered.
Based on the observed and modeled water quality of North and South Lake Cornelia and information on
the health of the aquatic communities, the following management strategies were considered:
• External load reductions (e.g., infiltration/filtration BMPs through retrofit or redevelopment,
enhancement of existing BMPs, street sweeping, and landowner practices such as leaf and grass
clippings removal, reduced fertilizer use, shoreline restoration and buffers)
• Curly-leaf pondweed treatments
• Alum treatment for lake sediments
• Winter aeration using direct oxygen injection
• Carp and goldfish tracking and benthivorous (bottom-feeding) fish management
Based on the observed and modeled water quality of Lake Edina and information on the health of the
aquatic communities, the following management strategies were considered:
• External load reductions (e.g., infiltration/filtration BMPs through retrofit or redevelopment,
enhancement of existing BMPs, street sweeping, and landowner practices such as leaf and grass
clippings removal, reduced fertilizer use, shoreline restoration and buffers)
• Management efforts in upstream lakes (North and South Cornelia)
A description of the evaluated watershed and in-lake management strategies is provided below. A
summary of the predicted effectiveness of the strategies, and combinations thereof, in improving lake
water quality is included in Section 8.0.
7.1 Watershed Management Strategies/Scenarios
A range of watershed management scenarios were considered to reduce external phosphorus loading to
Lake Cornelia and Lake Edina, and in turn, improve water quality. The watershed (external) management
scenarios evaluated were selected with the following targets in mind:
74
• Maximize benefits to chain of lakes. Runoff from the North Cornelia watershed flows through
South Cornelia and Lake Edina prior to discharging to Nine Mile Creek. Because the North
Cornelia watershed is the largest and most densely developed, runoff from this area also has the
greatest impact on water quality throughout the chain of lakes. To maximize the water quality
improvement benefits throughout the chain of lakes, several management scenarios were focused
on the Lake Cornelia watershed.
• Increase dissolved phosphorus removal. Runoff from most of the commercial, high-impervious
area tributary to Lake Cornelia is already conveyed through a series of ponds, such that much of
the particulate phosphorus (phosphorus attached to sediment particles) in the runoff from this
area has already settled out before reaching Lake Cornelia. Given the existing treatment, there
was a desire to target dissolved phosphorus that is not typically removed by ponds.
• Improve or “build on” effectiveness of existing treatment systems. As mentioned above,
runoff from much of the area tributary to Lake Cornelia already receives treatment from existing
ponds or other BMPs. Opportunities to modify or add on to the existing management systems to
improve effectiveness and/or expand the treatment spectrum were desired.
• Include mix of structural and non-structural BMPs. An optimal stormwater management
program includes a mix of structure and non-structural BMPs. Structural BMPs remove a fraction
of the pollutants and sediment loads contained in stormwater runoff prior to discharge into
downstream water bodies, whereas, non-structural BMPs eliminate pollutants and sediment loads
at the source and prevent them from entering stormwater flows. Examples of structural BMPs
include ponds, bioretention/infiltration basins, filtration systems, or vegetated filter strips/buffers.
Examples of non-structural BMPs include street sweeping, programmatic controls (i.e.,
rules/ordinances), reducing impervious surface, and other better-site-design techniques.
• Provide reliable pollutant removal performance. The watershed management scenarios should
be designed to provide consistent and reliable pollutant removal throughout the growing season
and over the long-term.
• Be reasonably cost effective.
• Land availability.
Based on these selection criteria, the following four external management scenarios were modeled:
1. Infiltration BMPs on commercial property
2. Filtration BMPs on commercial property
3. A spent lime/ CC17 treatment chamber as treatment train for flows from existing Swimming Pool
Pond
4. Weekly street sweeping of all public city roads and all private commercial lots.
75
These targeted BMP scenarios are described in further detail below.
Based on results of the previous studies for these lakes, it was understood that substantial reductions in
watershed phosphorus loading will likely be necessary to see significant improvements in lake water
quality. As such, the watershed management scenarios that were evaluated encompass fairly large-scale
efforts. Actual implementation of these watershed management scenarios could be scaled back or
implemented over a long timeframe.
7.1.1 Infiltration BMPs on Commercial Properties
Infiltration BMPs were selected as a scenario for external management due to their numerous benefits.
Infiltration BMPs, such as bioretention (rainwater gardens), infiltration basins, or underground infiltration
systems, are effective in significantly reducing pollutant loading from developed sites. Stormwater runoff
is captured and infiltrated, reducing the amount of stormwater volume and pollutants leaving the site,
including both particulate and dissolved phosphorus. Infiltration basins remove solids and particulate
nutrients (such as phosphorus) through settling and can reduce dissolved nutrient concentrations through
adsorption, filtration, ion exchange, and decomposition as the water moves through the soil and infiltrates
into the groundwater. Not only do infiltration-based BMPs reduce the amount of stormwater volume and
pollutants leaving a site, they also can reduce downstream flooding by holding back water, assist with
groundwater recharge and reduce peak runoff rates.
Under existing conditions, there are several redevelopment sites within the North Cornelia watershed that
have implemented infiltration-based BMPs on their sites, including portions of the Southdale Shopping
Center. These BMPs were installed in conformance with the NMCWD stormwater rule to achieve the
onsite retention requirement (currently 1.1 inches of runoff from impervious surfaces), as well as sediment
and phosphorus removal and peak flow reduction from the sites.
For this external management scenario, it was assumed that all commercial properties within the Lake
Cornelia and Lake Edina watersheds that have soils conducive to infiltration would implement infiltration-
based BMPs to capture and retain 1.1 inches of runoff from the impervious surface areas of the parcels.
Approximately 194 acres of commercial parcel area was treated under this scenario (approximately 99% of
commercial parcel area) and the infiltration BMPs were sized to treat 147.5 acres of impervious surfaces.
This scenario represents a condition where all commercial parcels are redeveloped in conformance with
NMCWD’s current stormwater rule. While this scenario represents extensive changes in the watershed that
would likely occur over a long period of time, it was evaluated as a benchmark or ”stretch goal” for
external management efforts.
This scenario aligns well with the following target criteria:
• Maximize benefits to chain of lakes. Most of the commercial land area is within the North
Cornelia watershed, so all three lakes within the chain would benefit from the reduction in
phosphorus loading.
76
• Increase dissolved phosphorus removal. Infiltration-based BMPs would increase the amount of
dissolved phosphorus removed from the watershed runoff.
• Include mix of structural and non-structural BMPs. The infiltration-based BMPs are considered
structural BMPs, however, installation of the BMPs would be implemented as a result of the
NMCWD regulatory program (non-structural/programmatic).
• Provide reliable pollutant removal performance. Infiltration-based BMPs would provide
consistent and reliable pollutant removal throughout the growing season (and potentially
throughout the winter during rainfall or snowmelt events). Assuming proper maintenance, it is
expected that the infiltration-based BMPs would continue to provide consistent and reliable
treatment, especially given the sandy soils within this area.
• Be reasonably cost effective. The infiltration-based BMPs would be constructed, funded, and
maintained by the development community, thus minimizing direct costs to the City of Edina or
NMCWD.
• Land availability. BMPs would be installed as part of conformance with the NCMWD stormwater
rule as land within the watershed redevelops. Additional land acquisition would not be required.
To model this external management scenario, commercial parcels with soils classified as Hydrologic Soil
Groups (HSG) A (high infiltration rate) and B (moderate infiltration rate) were identified within the lakes’
subwatersheds. Soil conditions were estimated based on online data resources, including information
from the Soil Survey Geographic Database (SSURGO), historical USGS quad maps, MDNR Public Waters
Inventory, and anecdotal information compiled from historic project data. Impervious areas were
estimated for these commercial parcel areas and infiltration-based BMPs were sized to capture 1.1-inches
of runoff from the commercial impervious surface areas (for consistency with the NMCWD stormwater
rule). The infiltration basins were added to the existing conditions P8 models to assess the impact of the
commercial infiltration BMPs on the phosphorus loading from the watershed. The results from the P8
model were input into the in-lake model to evaluate the changes in phosphorus concentrations in the
downstream lakes. Table 7-1 provides a summary of the infiltration basins modeled under this scenario.
Table 7-1 Infiltration BMP treatment volumes and infiltration rates
Lake Watershed Total Treatment Volume
(ac-ft)1
Infiltration Rate
(in/hour)2
Lake Edina LE-1 0.84 0.8
North Cornelia
NC-3 1.37 0.8
NC-4 9.66 0.8
NC-5 0.32 0.45
NC-6 0.07 0.45
NC-62 1.71 0.6
NC-88 0.39 0.8
1 Treatment volume based on treating 1.1” of runoff from commercial impervious surfaces
2 Infiltration rate based on HSG soil classifications
77
7.1.2 Filtration BMPs on Commercial Properties
For this external management scenario, it was assumed that commercial properties within the Lake
Cornelia and Lake Edina watersheds that have soils conducive to infiltration would implement filtration-
based BMPs to capture and filter 1.1 inches of runoff from the impervious surface areas of the parcels.
Approximately 194 acres of commercial parcel area was treated under this scenario (approximately 99% of
commercial parcel area) and the filtration BMPs were sized to treat 147.5 acres of impervious surfaces.
Filtration basins were considered for treatment due to their numerous benefits. Filtration basins allow for
the removal of solids and particulate nutrients through settling and can reduce dissolved nutrient
concentrations through adsorption, ion exchange, and decomposition as water moves through the initial
soil matrix before being discharged through an underlying drain tile system. Filtration basins can assist
with peak flow reductions, but will have a minimal impact on the total volume of water that reaches
downstream areas during precipitation events. This scenario was selected to analyze a condition that
would reduce phosphorus loadings, but not impact water loadings. The same basins that were developed
for the commercial infiltration scenario were used for this scenario, but modified in the P8 model to
function as filtration basins.
This scenario aligns well with the same BMP target criteria as the Infiltration BMPs scenario, with
exception that the filtration BMPs remove a lower percentage of dissolved phosphorus loads.
Similar to the Infiltration BMPs on Commercial Properties scenario, this scenario represents a condition
where all commercial parcels are redeveloped in conformance with NMCWD’s current stormwater rule.
While this scenario represents extensive changes in the watershed that would likely occur over a long
period of time, it was evaluated as a benchmark or ”stretch goal” for external management efforts.
7.1.3 Spent Lime/CC17 Treatment Chamber
A third watershed management scenario evaluated to reduce external phosphorus loading was
construction of a double-chamber spent lime/CC17 filter at the downstream discharge point of Swimming
Pool Pond. Swimming Pool Pond has a tributary watershed of 410 acres, which is approximately 47% of
the total North Lake Cornelia tributary watershed. Based on the P8 modeling analyses, the waterbody is
effective in settling out sediment and associated particulate phosphorus; however, little to no dissolved
phosphorus is removed as runoff flows through this waterbody into Lake Cornelia. The proposed spent
lime/CC17 treatment chamber would serve as a “polishing” step, diverting a portion of the discharge from
Swimming Pool Pond through the spent lime filtration chamber to remove dissolved phosphorus before
discharge to Lake Cornelia.
The methods to remove dissolved phosphorus from stormwater are limited. To remove dissolved
phosphorus (or phosphate) from runoff, phosphate must be either incorporated into recalcitrant organic
material or be bound to cations (Ca2+, Mg2+, Fe2+) or unreducible trace metals (Al2+). Calcium carbonate is
the primary component of spent lime. A spent lime treatment chamber uses chemical substitution to
exchange phosphate for carbonate. As runoff filters through spent lime, calcium will preferentially bind to
phosphate over carbonate and calcium phosphate will form. This type of “filtration” differs from other
78
methods of dissolved phosphorus removal in that spent lime is not precipitating or flocculating
phosphate (as an iron-enhanced sand filter or alum treatment do).
Using spent lime to treat stormwater is a relatively new and innovative approach that Barr has been
experimenting with at a few locations throughout the Twin Cities metro area in recent years. Benefits of
spent lime treatment to treat stormwater runoff include:
• Spent lime is considered a “waste material” from water treatment plants and thus, a green
material with low material costs
• Rapid chemical substitution reactions between phosphate and carbonate lead to a high treatment
capacity
• Spent lime material has high hydraulic conductivity, therefore allowing for large volumes of
treatment over a relatively small footprint
• Spent lime treatment can remove both particulate and dissolved phosphorus
• Additional water quality benefits are attained through the removal of aluminum, calcium, iron,
zinc, and lead
• Relatively easy maintenance with annual mixing of the lime material to maintain porosity,
maintain hydraulic conductivity, and expose new spent lime surfaces to stormwater
While there are many benefits to spent lime, one drawback is that the material has limited capacity for
prolonged inundation. Therefore, for the proposed treatment chamber located on the downstream end of
Swimming Pool Pond it is recommended that the bottom layer of material be supplemented with CC17
due to the potential for periodic high water levels in North Cornelia. CC17 is a crushed limestone material
that is more soluble that most limestone aggregates and thus, provides calcium that can bind phosphate
and create calcium phosphate. Unlike spent lime, CC17’s lifespan will not be reduced by prolonged
inundation.
To model the effectiveness of this external management BMP, the spent lime/CC17 chamber was sized to
filter 2 cubic feet per second (cfs) of discharge from Swimming Pool Pond prior to discharge into North
Cornelia. A conceptual diagram of the double-chamber spent lime/CC17 treatment cell is depicted in
Figure 7-1. Pilot studies conducted with spent lime and CC17 filters have demonstrated effective removal
of total phosphorus and dissolved phosphorus fractions. For this UAA study, a total phosphorus removal
efficiency of 62% was assumed for the modeling, which was the average removal efficiency for a spent
lime filter located near Wakefield Lake, Maplewood, MN for the sampling period of 2012-2016. Treatment
tests with CC17 have shown that the material is a good filter for particulates and that the material does
not appear to clog as readily as sand. Additionally, recent testing completed on a CC17 treatment cell in
the Ramsey-Washington Metro Watershed District showed that the CC17 material had nearly the same or
more treatment capacity to remove phosphorus than spent lime alone. This indicates that a combined
treatment cell with CC17 and spent lime may have the capacity to effectively treat discharges from
Swimming Pool Pond.
79
This scenario aligns well with the following target criteria:
• Maximize benefits to chain of lakes. The proposed spent lime/CC17 chamber is directly
upstream of North Cornelia, so all three lakes within the chain would benefit from the reduction in
phosphorus loading.
• Increase dissolved phosphorus removal. The spent lime/CC17 chamber would increase the
amount of dissolved phosphorus removed from the watershed runoff.
• Improve or “build on” effectiveness of existing treatment systems. The Point of France Pond
and Swimming Pool Pond provide effective removal of sediment and particulate phosphorus from
the 410-acre tributary watershed, but do not remove dissolved phosphorus. The proposed spent
lime/CC17 would serve as a “treatment train” approach and provide a “polishing” step to a
portion of the water flowing from Swimming Pool Pond prior to discharge to Lake Cornelia.
• Include mix of structural and non-structural BMPs. The spent lime/CC17 chamber would be
considered a structural BMP, however, since the proposed implementation location is within a
public park, signs could be posted near the BMP for educational benefit of the park users (non-
structural/programmatic).
• Provide reliable pollutant removal performance. The proposed spent lime/CC17 would
provide consistent, year-round treatment of low-flows from Swimming Pool Pond. However, high
flows would bypass the system and flow directly to Lake Cornelia without treatment. Use of spent
lime/CC17 to remove dissolved phosphorus from stormwater is still a relatively new and
experimental technique; however, results from other regional applications have been promising.
• Be reasonably cost effective. The spent lime/CC17 chamber uses materials considered “waste”
from previous applications helping to keep the construction costs reasonably cost effective.
• Land Availability: The proposed location for the spent lime/CC17 implementation is underneath
an existing parking lot in a public park. Installation of the proposed BMP would not require re-
purposing of the land area footprint.
80
Figure 7-1 Conceptual design of the double-chamber spent lime/CC17 treatment cell
81
7.1.4 Weekly Street Sweeping
The fourth watershed management BMP scenario to reduce external phosphorus loading was weekly
street sweeping of all public city roads and all private commercial lots. In 2015, the City of Edina
developed a street sweeping management plan which establishes the following measurable goals and
timeframes for using street sweeping as a BMP (Emmons & Olivier Resources, Inc. (EOR), 2015):
“The City will brush or vacuum sweep streets a minimum of twice annually in an effort to reduce the amount
of sediment, trash, and organic material from reaching the storm sewer system and water resources.”
Although this measurable goal identifies biannual sweeping, the city is interested in potentially
implementing changes to the street sweeping program, such as more frequent and/or targeted sweeping,
to further reduce stormwater pollutants. Thus, efforts were made in this UAA process to quantify the in-
lake responses to an enhanced street sweeping program.
The city’s street sweeping management plan investigated the water quality benefits of the current street
sweeping program and the projected benefits of an enhanced street sweeping program (sweeping once a
month, sweeping bi-weekly) (Emmons & Olivier Resources, Inc. (EOR), 2015). Annual pollutant load
recoveries were estimated using the street sweeping planning calculator tool, ‘Estimating Nutrient and
Solids Load Recovery through Street Sweeping’, developed by Kalinosky et. al (2014) as part of a Master’s
Thesis at the University of Minnesota. The calculator was used to predict annual pollutant recovery
benefits for the City of Edina public streets. Watershed-wide tree canopy assessments and curb-mile
summaries were completed to estimate annual total phosphorus reductions in pounds of total
phosphorus per year. Three street sweeping scenarios were analyzed in the study (twice annually,
monthly, bi-weekly) and estimated total phosphorus recoveries for a regenerative air street sweeper
ranged from 48.8 to 181.7 pounds for the Lake Cornelia and Lake Edina watersheds.
The street sweeping load reduction calculator developed by Kalinosky et. al (2014) provides an estimate of
the amount of annual total phosphorus removed from street sweeping, in pounds of total phosphorus
recovered annually, calculated using regression equations based on data collected from nine street
sweeping routes in Prior Lake, Minnesota from August 2010 to July 2012. The calculator does not translate
the estimated amount of annual total phosphorus removed to a load reduction (i.e., the calculator reports
pounds of total phosphorus removed but not a percent removal). Given this, it is difficult to extrapolate
the average annual results to observed time series data used in the lake modeling analysis, due to
differences in precipitation volume and frequency from year to year. Thus, for this UAA study a different
approach was used to estimate the in-lake response to an enhanced street sweeping program.
The street sweeping scenario analyzed for this UAA study assumes public streets and private commercial
parking lots were swept with a high-efficiency vacuum-assisted street cleaner on a weekly basis. It is
recognized that this would be an extensive change to the street sweeping program already in place (twice
annually). However, it sets a benchmark for the greatest potential of nutrient reductions from runoff
through the removal of sediment, leaves, and other organic detritus from urban streets and commercial
lots. A weekly street sweeping scenario would also align better with the BMP target criteria of providing
82
reliable and consistent pollutant removal, as opposed to a bi-annually, monthly, or even bi-weekly street
sweeping program.
For this weekly street sweeping scenario, the modeled efficiency of nutrient removal through street
sweeping efforts was based on a paper by the U.S. Geologic Survey (USGS) William Selbig (Selbig, 2016).
In this study, a paired catchment system was used to quantify the potential for a municipal leaf collection
and street cleaning program in Madison, WI. One catchment area was established as a control with no
effort to remove leaf litter or other organic detritus from streets and a second was established to serve as
the test catchment in which removal of leaf litter and detritus was done through weekly street sweeping,
leaf collection, and leaf blowing. In the period of April through September, weekly street cleaning was the
only form of treatment in the test catchment. Selbig’s results were grouped into seasonal categories
(spring, summer, and fall). For spring samples, Selbig found that weekly street sweeping reduced total
phosphorus loads by approximately 45% (phosphorus loads that reached monitored downstream catch
basins). In the summer, total phosphorus loads were reduced by 36% for weekly street sweeping in the
test catchment area. The extensive street cleaning efforts that occurred in the fall showed that total
phosphorus concentrations could be reduced by 84% during the period when total phosphorus loading
was the greatest. For UAA modeling efforts, a total phosphorus loading reduction of 36% was applied to
inflows throughout the modeled time periods to predict in-lake responses to weekly street cleaning. This
load reduction factor was selected in part because it best reflects the time period of interest (June through
September) for the in-lake modeling. Selection of the 36% total phosphorus load reduction also reflects a
somewhat conservative approach (versus using an increased load reduction assumption during the
springtime). The characteristics of the test catchment included 19% streets, 4% driveways, 19% roofs, 3%
sidewalks, 54% lawns, and 17% street tree canopy. Since not all of the street catchments in the City of
Edina will have these same characteristics, using a more conservative loading reduction factor is
appropriate. Nevertheless, since this study, as well as other studies (Kalinosky, 2015), show that street
sweeping efforts have a greater capacity to reduce loadings in the spring and fall, it should be recognized
that larger load reductions during those time periods than what are presented in this UAA study may be
achievable with weekly sweeping efforts.
The same total phosphorus loading reduction factor found in the USGS study for an urban street was used
to estimate the phosphorus load reductions for private, commercial parking lots. A literature search
yielded no studies investigating the effects of street sweeping of commercial parking lots (versus streets).
A few studies were found that assessed the nutrient characteristics of parking lots compared to
streets/highways. However, these studies are inconclusive. A few studies showed that parking lots have
reduced levels of phosphorus compared to highways or streets (Bannerman, Owens, Dodds, & Hornewer,
1993; Hope, Naegeli, Chan, & Grimm, 2004; Wei, et al., 2010), while others showed that phosphorus
concentrations in parking lots were comparable or greater than streets/highways (Passeport & Hunt,
2009). Due to the limited and inconclusive studies completed for parking lot phosphorus concentrations
and street sweeping efforts, a total phosphorus loading reduction factor of 36% (from impervious areas)
was used for commercial properties for this study. In the future, a pilot study may be warranted to more
accurately reflect the full impact of street sweeping private parking lots.
83
The weekly street sweeping scenario aligns well with the following target criteria:
• Maximize benefits to chain of lakes. The street sweeping scenario would have direct benefit to
all three lakes.
• Improve or “build on” effectiveness of existing treatment systems. Regular street sweeping
would reduce the amount of sediment and nutrients reaching downstream BMPs and/or water
bodies. This reduction in sediment and nutrients could be expected to reduce maintenance
frequency and extend the useful life of these other BMPs and/or water bodies.
• Include mix of structural and non-structural BMPs. Street sweeping would be considered a
non-structural BMP.
• Provide reliable pollutant removal performance. One of the disadvantages of typical street
sweeping programs is the infrequency of sweeping, and therefore the inconsistent treatment
provided by the BMP. The weekly frequency of the modeled scenario reduces the inconsistency of
treatment, but would be resource-intensive.
• Be reasonably cost effective.
• Land availability. One of the benefits of street sweeping is that a dedicated land footprint is not
required. If pursuing sweeping of private land, special access agreements may be necessary.
7.1.5 Other Watershed Management Strategies
The watershed management scenarios considered as part of this study were focused on reducing external
phosphorus loading to both Lake Cornelia and Lake Edina to maximize benefits to the chain of lakes.
However, additional opportunities to reduce phosphorus from the direct watershed to Lake Edina could
also be considered. The Lake Edina watershed is predominantly single-family residential land use;
however, there are several publicly-owned parcels within the watershed that may present opportunities
for retrofitting BMPs to reduce phosphorus loading to Lake Edina. The Cornelia Elementary School (7000
Cornelia Drive, owned by Independent School District 273) and the adjacent Cornelia School Park (owned
by City of Edina) parcels may present future opportunities for installing stormwater BMPs through
partnership projects.
There are numerous watershed management practices that landowners can implement to help reduce
phosphorus loading to Lake Cornelia and Lake Edina, including rain gardens, shoreline buffers, redirection
of gutter downspouts to yards (versus impervious driveways), clean-up of grass clippings, and reduced
use of phosphorus-based fertilizers.
7.2 Internal Load Reductions
A range of in-lake management scenarios were considered to help reduce the phosphorus concentrations
in Lake Cornelia and Lake Edina, and in turn, improve water quality and habitat for aquatic communities.
In-lake BMPs reduce phosphorus already present in a lake or prevent the release of phosphorus from the
84
lake sediments. For this UAA study, two in-lake management scenarios were explicitly modeled for North
and South Cornelia: (1) curly-leaf pondweed management, and (2) alum sediment treatments. No in-lake
management processes were modeled for Lake Edina since the calibration models indicated that internal
loading is not a major factor involved in phosphorus loadings to the lake. Despite Lake Edina having
limited internal phosphorus loading, there are aquatic invasive species present in the lake that require
management considerations. This is discussed further in Section 7.3.
7.2.1 Curly-leaf Pondweed Management
The presence of curly-leaf pondweed and its mid-summer die-off negatively impacts the water quality of
Lake Cornelia and downstream Lake Edina. Modeling results indicate that curly-leaf pondweed contributes
up to 17% of the annual phosphorus loading to North Cornelia and up to 23% of the annual phosphorus
loading to South Cornelia. Accordingly, management of the curly-leaf pondweed is an important
component of a long-term management plan for Lake Cornelia. Curly-leaf pondweed management was
modeled as an in-lake management scenario. Several assumptions were applied to this modeling effort,
including: (1) maximum curly-leaf biomass of 800 kilograms per hectare, (2) phosphorus content of
pondweed tissue was 4.2 grams per kilogram dry plant material, (3) pondweed mortality rate of 0.10 to
0.15 per day, and (4) a decay rate of 0.10 to 0.15 per day. Phosphorus release occurred in the model
during the decay phase.
The City of Edina has been conducting herbicide treatments in Lake Cornelia in 2017 and 2018 to reduce
the impact of curly-leaf pondweed die-back on water quality in Lake Cornelia and downstream Lake Edina
and promote a healthy native aquatic plant population. Continued curly-leaf management efforts would
likely consist of continued herbicide treatments at a treatment dose such that a lethal dose is attained and
sustained for the period of time sufficient to kill the curly-leaf pondweed.
7.2.2 Alum Treatment of Lake Sediments
Modeling results confirm that internal release of phosphorus from lake-bottom sediments is a significant
source of phosphorus to Lake Cornelia, contributing an estimated 14%-40% of the annual phosphorus
loading to North Cornelia during modeled years, and an estimated 14%-19% of the annual phosphorus
loading to South Cornelia. Accordingly, control of the internal phosphorus release is an important
component of a long-term management plan for Lake Cornelia and downstream Lake Edina.
A whole-lake alum treatment (aluminum is the active ingredient, and hence this can be considered an
aluminum treatment) is proposed for both North and South Cornelia. The in-lake modeling scenario
assumes internal phosphorus loading from lake sediments is reduced by 80 percent.
The proposed alum treatment should be conducted across the entire lake surface to depths as shallow as
feasible. The dose in terms of alum (4.4% aluminum by weight) for North Cornelia is 1,539 gallons per acre
for a total application of 29,238 gallons, and for South Cornelia the dose is 530 gallons per acre for a total
application of 16,431 gallons. The doses for both North and South Cornelia were based upon treating the
top 4 centimeters of lake-bottom sediment. Because Lake Cornelia is shallow and there is a potential for
the pH to drop too low if only alum is applied, the aluminum should be applied as a mixture of alum
85
(4.4% aluminum by weight) and sodium aluminate (10.4% aluminum by weight). Sodium aluminate acts as
a buffer and will keep pH in an acceptable range. A summary of the proposed treatment is provided in
Table 7-2. Note that the total gallons of alum and sodium aluminate are less than the total gallons of
alum only as the mass of aluminum is much higher for liquid sodium aluminate, this reduces the total
gallons of sodium aluminate that need to be applied to achieve the desired application mass of
aluminum.
Table 7-2 Summary of alum and sodium aluminate application doses for North and South
Cornelia
Location Dose (g Al/ac) Product Application
Ratio Gallons/Acre Total Gallons
Applied
North
Cornelia 84
Alum 2 673 12,791
Sodium
Aluminate 1 337 6,396
South
Cornelia 29
Alum 2 232 7,188
Sodium
Aluminate 1 116 3,594
An alum treatment can be conducted before the other lake management activities are completed;
however, there are aspects that should be considered. When alum (aluminum) is added to a lake surface,
it settles to the bottom and temporarily there is a layer of aluminum floc (aluminum hydroxide) that
covers the lake bottom. Due to benthic activity from fish and invertebrates, as well as wind and wave
action, aluminum hydroxide mixes readily with benthic sediment and as it is mixed it has an opportunity
to bind phosphorus in the sediment. Mixing is necessary for the aluminum hydroxide to bind phosphorus
in the sediment. However, when conditions are such that mixing is extensive, then the aluminum can
become diluted as it mixes deeper into the sediment. Because of the extensive benthivorous fish in Lake
Cornelia, the longevity of the treatment may be reduced as the aluminum is mixed deeper into the
sediment over time. If treatment is conducted before other activities, it can be expected that a second
treatment will be needed in the relatively near future (5 to 10 years after treatment).
7.3 Other Lake Management Strategies
Several other lake management strategies were considered as part of this UAA study, but not explicitly
included in the modeling analysis. These management strategies are described below.
7.3.1 Carp and Goldfish Tracking and Benthivorous (Bottom-feeding) Fish
Management
Review of the 2018 fishery data indicate that the Lake Cornelia fishery tends to be heavily influenced by
frequent winterkill events, evidenced by a low number of bluegill and other predator fish. The frequency
of winterkills and the availability of connected shallow waterbodies that winterkill which act as nurseries,
are most likely preventing bluegills and other sunfish from effectively controlling carp, bullheads, and
86
goldfish within the system. The abundant benthivorous fish population is likely negatively affecting water
quality by stirring up the lake bottom sediments.
Several in-lake management activities to manage the abundant and unchecked population of
benthivorous fish and promote a healthy and more diverse fishery were considered as part of this UAA
study. One of the proposed management practices is installation of a winter aeration system using direct
oxygen to prevent winterkill and promote survival of predator fish. This management option is discussed
in further detail below. Other potential management activities could include a rotenone treatment of Lake
Cornelia and the upstream water bodies and subsequent fish stocking and/or installation of fish barriers.
Prior to considering these fishery management activities collection of additional information is
recommended regarding the migration and movement of carp and goldfish throughout Lake Cornelia and
the series of connected upstream shallow waterbodies.
7.3.1.1 Winter aeration using direct oxygen injection
The purpose of winter aeration using direct oxygen injection is to prevent winter kill of predator fish,
therefore maintaining a more balanced fishery. Winter aeration would be conducted by injecting oxygen
under the ice. The system would consist of: (1) a unit that generates the oxygen and a structure that
houses the generator, (2) a raft that holds the vertical aeration tubes in place, and (3) a bubbler that
directs the oxygen upwards through the aeration tubes. The aeration tubes consist of an inner tube
surrounded by an outer tube. Air is applied at the bottom of the inner tube, moving the air upward
through the tube. Once the water reaches the surface of the inner tube, the water falls downward in the
outer tube. Pure oxygen is either directed at the bottom of the inner tube or at the top of the outer tube.
The inner tube draws in water at the bottom of the lake. The outer tube is shorter than the inner tube and
aerated water is the delivered laterally across the lake.
The proposed direct oxygen system would be installed in North and South Cornelia at the deep holes of
the lake contingent upon the availability of power. The system would be sized to be operated only in the
winter when there is ice cover. Dissolved oxygen measurements were collected in North and South
Cornelia during the winter of 2019 to inform the potential sizing and placement of winter aeration
systems.
7.3.2 Lake Edina Aquatic Plant Management
Three non-native aquatic invasive species (AIS) are present in Lake Edina: purple loosestrife, curly-leaf
pondweed, and Eurasian watermilfoil.
Purple loosestrife was observed along the perimeter of the lake during the 2008, 2012, 2015, and 2017
sampling periods. The current infestation is not considered problematic. However, the infestation will be
evaluated periodically when aquatic plant surveys are completed. If the infestation becomes problematic
in the future, it can be managed by introducing purple loosestrife eating beetles (Galerucella calmariensis
and/or Galerucella pusilla) to the infested areas. The beetles manage purple loosestrife by inflicting
damage to the plants.
87
7.3.2.1 Curly-leaf Pondweed
The invasive curly-leaf pondweed has been observed at low levels in Lake Edina since 2008. In June of
2017, the species was observed at two locations, both on the west side of the lake. In August, it was
observed at a single location in the central western area of the lake. Although curly-leaf pondweed has
remained at low levels in the lake since 2008, management of curly-leaf pondweed may be warranted to
maintain its low occurrence and prevent the accumulation of turions (i.e., similar to seeds). The goals of
treatment would be to prevent curly-leaf pondweed from establishing dominance to avoid the need for
subsequent long-term annual treatments to reduce an established population that can rebound once
larger numbers of turions are present in the sediments. Management of the current curly-leaf population
would also minimize the potential for turions to be conveyed downstream to Normandale Lake, causing a
resurgence of curly-leaf pondweed after completion of the Normandale Lake water quality improvement
project.
The herbicide selected for the curly-leaf pondweed treatment would depend upon the extent; when
greater than 15 percent of the lake, endothall would be recommended to attain lake-wide control of the
curly-leaf pondweed. When the curly-leaf pondweed extent is less than 15 percent of the lake, a cost and
benefit analysis would be recommended to determine whether Endothall or diquat would be most
appropriate. Based upon experience with other curly-leaf pondweed management projects, the
management of curly-leaf pondweed would be expected to span several years. Management until neither
curly-leaf pondweed nor turions are observed in the lake would be most protective of the Lake Edina
ecosystem as well as downstream Normandale Lake from the problems associated with a curly-leaf
pondweed infestation.
7.3.2.2 Eurasian Watermilfoil
Eurasian watermilfoil was first observed in Lake Edina during 2017, in which it was widespread and
increased in extent between June and August. Unlike many other plants, Eurasian watermilfoil does not
rely on seed for reproduction. It generally reproduces by fragmentation—each fragment can grow into a
new plant. The plant produces fragments after fruiting at least once or twice during the summer. Eurasian
watermilfoil’s fast growth rate (up to 2 inches per day in spring and summer), its ability to spread rapidly
by fragmentation, and its ability to effectively block out sunlight needed for native plant growth often
results in monotypic stands which provide only a single habitat and threaten the integrity of aquatic
communities in a number of ways.
Management of Eurasian watermilfoil in Lake Edina would control its rapidly expanding extent and
prevent Eurasian watermilfoil from further threatening the integrity of the lake’s aquatic community.
Because Eurasian watermilfoil fragments can be carried downstream to Normandale Lake, managing
Eurasian watermilfoil in Lake Edina would also protect the integrity of the Normandale Lake aquatic
community. The herbicide selected for Eurasian watermilfoil treatment in Lake Edina would depend upon
the extent of plant coverage; when greater than 15 percent of the lake, the herbicide 2,4-D would be
recommended to attain lake-wide control of Eurasian watermilfoil. When the extent of Eurasian
watermilfoil is less than 15 percent of the lake, a cost and benefit analysis would be recommended to
determine whether 2,4-D or ProcellaCOR would be the most appropriate herbicide for the treatment.
88
Based upon experience with other Eurasian watermilfoil management projects, the management of
Eurasian watermilfoil in Lake Edina would be expected to span several years, but the extent to be
managed would diminish annually to smaller and smaller levels. As the extent of Eurasian watermilfoil
diminishes, management efforts and associated cost would be expected to change accordingly.
Management until Eurasian watermilfoil is no longer observed in the lake would be most protective of the
Lake Edina ecosystem as well as downstream Normandale Lake from the problems associated with a
Eurasian watermilfoil infestation.
89
8.0 Lake Response to Management Strategies
Section 7.0 discussed the management strategies that were evaluated for North and South Cornelia and
Lake Edina. This section outlines the predicted in-lake responses to these various management strategies,
and combinations thereof.
8.1 Lake Response to Watershed (External) Management Strategies
The four external loading management strategies modeled were (1) infiltration BMPs on commercial
properties, (2) filtration BMPs on commercial properties, (3) a spent lime/CC17 treatment chamber
downstream of Swimming Pool Pond (North Cornelia watershed), and (4) weekly watershed-wide street
sweeping.
The effectiveness of the various management strategies in reducing the predicted in-lake total
phosphorus concentrations in Lake Cornelia and Lake Edina is summarized in the subsections below. It is
important to note that the four watershed management strategies modeled represent a range in scale of
treatments and costs. Accordingly, comparison of the water quality improvements resulting from the four
scenarios should not be considered a like-for-like comparison, but rather should be used to provide a
sense of how effective the management strategies are towards achieving the lake water quality goals. The
cost-benefit analysis, presented in Section 9.0, allows for a more direct comparison of the effectiveness of
the evaluated management strategies.
8.1.1 Changes in In-lake Phosphorus Concentrations
As described in Section 7.0, the P8 and calibrated in-lake models were used to predict the changes in total
phosphorus concentrations in each of the lakes throughout the summer months as a result of the various
watershed management activities. As an example, Figure 8-1 shows the predicted in-lake phosphorus
concentration changes in North Cornelia in 2015 when the various external load management strategies
were applied to the models. The black, dashed line represents North Cornelia’s phosphorus
concentrations under existing conditions and the red, dashed line represents the MPCA’s water quality
standard for shallow lakes in the north central hardwood forest ecoregion (60 µg/L). The goal of the
various management efforts is to bring the summer average phosphorus concentration below this
standard. As shown on Figure 8-1, in 2015 the external management efforts alone do not improve the in-
lake phosphorus concentrations in North Cornelia enough to meet the MPCA standard, however, the
external management efforts did result in notable changes to the in-lake phosphorus concentrations.
90
Figure 8-1 In-lake phosphorus concentrations that resulted from watershed management efforts in North Cornelia in 2015
8.1.2 Summer Average Total Phosphorus Concentrations
Figure 8-2, Figure 8-3, and Figure 8-4 summarize the current and predicted summer average total
phosphorus concentrations in North Lake Cornelia, South Lake Cornelia, and Lake Edina, respectively
based on the modeled watershed management strategies. The goal of the various management efforts is
to bring the summer average phosphorus concentration below the 60 µg/L standard. As shown in the
figures, while each of the four external management scenarios generally result in lower summer average
total phosphorus concentrations in the lakes, none of the strategies improve the lake enough to meet the
water quality standard.
91
Figure 8-2 North Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for Watershed Management Efforts
Figure 8-3 South Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for Watershed Management Efforts
92
Figure 8-4 Lake Edina In-Lake Summer Average Phosphorus Concentration Summary for Watershed Management Efforts
8.1.3 Phosphorus Loading Reductions
The in-lake total phosphorus concentrations of North and South Cornelia and Lake Edina are influenced
over time through loading reductions. Table 8-1 summarizes the total phosphorus loads and percent load
reductions from the four watershed management strategies, in comparison with current conditions.
93
Table 8-1 Total phosphorus (TP) load reductions resulting from watershed management efforts
Lake Model
Year
Current
Load (lbs TP)
Commercial Infiltration
BMPs Commercial Filtration BMPs Spent Lime/CC17 Treatment
Cell Weekly Street/Lot Sweeping
Model Load
(lbs TP)
Percent
Reduction (%)
Model Load
(lbs TP)
Percent
Reduction (%)
Model Load
(lbs TP)
Percent
Reduction (%)
Model Load
(lbs TP)
Percent
Reduction (%)
North
Cornelia
2015 456 329 28% 408 11% 438 4% 390 14%
2016 489 337 31% 432 12% 467 4% 411 16%
2017 673 514 24% 625 7% 652 3% 603 10%
South
Cornelia
2015 442 341 23% 416 6% 433 2% 397 10%
2016 444 347 22% 419 6% 433 2% 396 11%
2017 525 406 23% 507 3% 516 2% 485 8%
Lake Edina 2015 306 217 29% 288 6% 299 2% 237 23%
2016 430 321 25% 407 5% 424 1% 347 19%
2017 410 290 29% 395 4% 405 1% 340 17%
94
8.1.4 Watershed BMP-specific Results
8.1.4.1 Infiltration BMPs on Commercial Properties
Infiltration BMPs not only remove particulate and dissolved phosphorus loads and reduce peak flows, but
they also remove water loads to downstream water bodies, which can be an added benefit, reducing flood
bounce and downstream erosion potential. Runoff that is captured in infiltration basins recharges the
groundwater and can alter the water balance of a system. For this external management scenario, it was
assumed that all commercial properties within the Lake Cornelia and Lake Edina watersheds that have
soils conducive to infiltration would implement infiltration-based BMPs to capture and retain 1.1 inches of
runoff from the impervious surface areas of the parcels. Through capturing stormwater runoff in
infiltration basins, the volume of runoff reaching North Lake Cornelia was reduced by 38%, 39%, and 39%
for the complete model years of 2015, 2016, and 2017 respectively. Total phosphorus loading from
watershed runoff was reduced by 36%, 37%, and 37% for calibration periods 2015, 2016, and 2017
respectively due to enhanced infiltration practices.
As shown in Figure 8-2, implementation of this extensive BMP scenario resulted in only moderate
reductions in summer average total phosphorus concentrations in 2015 and 2016 and actually increased
the in-lake summer average phosphorus concentrations in 2017. Of the 3 years modeled for this UAA
effort, 2017 was the model year with the least amount of precipitation that occurred during the growing
season. While typically it would be expected that external phosphorus load reductions should improve
water quality conditions of the downstream lake, for a lake that is not only receiving high external loads,
but also significant internal loads, adjusting the water balance of a lake can have negative consequences
on water quality. This outcome was predicted by the in-lake model in 2017 for North Cornelia. In 2017,
lower-than-average precipitation occurred over the growing season. With lower-than-average runoff due
to dry climatic conditions and increased infiltration, internal phosphorus loading has a greater capacity to
affect water quality due to reduced predicted lake levels and reduced flushing.
The modeling results for this scenario indicate that if future adjustments in the watershed significantly
alter the water balance of North Cornelia without addressing the enhanced internal loading, negative
consequences to water quality could result (especially for years with low precipitation during the growing
season). Section 8.3 discusses how the in-lake concentrations of North Cornelia might change based on
combined external and internal management efforts. Additional information on model results for the
Infiltration BMPs on Commercial Properties scenario, including plots of the in-lake model results for 2016
and 2017 for North Cornelia and the in-lake model results for South Cornelia and Lake Edina for all
modeled years, is provided in Appendix D.
8.1.4.2 Filtration BMPs on Commercial Properties
Because model results indicated that the changes in hydrology caused by wide-spread implementation of
infiltration BMPs on commercial properties can result in increased internal loading and periodic increases
in North Cornelia’s in-lake phosphorus concentrations, a modified external management scenario was
considered using filtration BMPs. Under this scenario the commercial parcels were treated with filtration
basins rather than infiltration basins. Filtration BMPs have a lower pollutant removal efficiency as
95
compared with infiltration BMPs since water volume is not removed from the system. Filtration basins are
effective at treating particulate phosphorus, but have limited capabilities to remove dissolved phosphorus.
For this watershed management scenario, it was assumed that commercial properties within the Lake
Cornelia and Lake Edina watersheds would implement filtration-based BMPs to capture and filter
1.1 inches of runoff from the impervious surface areas of the parcels. Runoff volumes to North Lake
Cornelia were not affected by the filtration basins, although slight adjustments in the timing of the runoff
were noted. Total phosphorus loading from watershed runoff was reduced by 15%, 15%, and 14% for the
2015, 2016, and 2017 calibration periods respectively due to filtration practices.
As shown in Table 8-1, widespread implementation of filtration BMPs on commercial properties results in
lower load reductions than the infiltration BMP scenario. However, the predicted improvements in
summer average total phosphorus concentrations are similar or better under the filtration BMP scenario
for North and South Cornelia (see Figure 8-2 and Figure 8-3). While widespread implementation of
filtration basins improves the summer average total phosphorus concentrations in all three lakes, the
improvements do not bring the summer average phosphorus concentration below the 60 µg/L standard.
Additional information on model results for the Filtration BMPs on Commercial Properties scenario,
including plots of the in-lake model results for 2016 and 2017 for North Cornelia and the in-lake model
results for South Cornelia and Lake Edina for all modeled years, is provided in Appendix D.
8.1.4.3 Spent Lime/CC17 Treatment Chamber
A third external management scenario encompassed construction of a double-chamber spent lime/CC17
treatment chamber at the downstream discharge point of Swimming Pool Pond, which is directly
upstream of North Cornelia. The proposed spent lime/CC17 treatment chamber would serve as a
“polishing” step, diverting a portion of the discharge from Swimming Pool Pond through the spent lime
filtration chamber to remove dissolved phosphorus before discharge to Lake Cornelia.
For the modeling analysis, the spent lime/CC17 treatment chamber was sized to treat 2 cfs of flow
discharging from Swimming Pool Pond. This treatment capacity reduces the phosphorus load from
Swimming Pool Pond by 15% to 16% depending on the model year. Overall, the spent lime/CC17
treatment chamber reduces the total external phosphorus load to North Cornelia by approximately 7%. As
shown in Figure 8-2, Figure 8-3, and Figure 8-4, the spent lime/CC17 treatment chamber reduces in-lake
summer average phosphorus concentrations for each lake. However, the improvements do not bring the
summer average phosphorus concentration below the 60 µg/L standard. Additional information on model
results for the Spent Lime/CC17 Treatment Chamber scenario, including plots of the in-lake model results
for 2016 and 2017 for North Cornelia and the in-lake model results for South Cornelia and Lake Edina for
all modeled years, is provided in Appendix D.
8.1.4.4 Weekly Watershed-wide Street Sweeping
The fourth external management scenario assumed public streets and private commercial parking lots
were swept with a high-efficiency vacuum-assisted street cleaner on a weekly basis. It is recognized that
this would be an extensive change to the street sweeping program already in place (twice annually).
96
However, it sets a benchmark for the greatest potential of nutrient reductions from runoff through the
removal of sediment, leaves, and other organic detritus from urban streets and commercial lots. As shown
in Figure 8-2, Figure 8-3, and Figure 8-4, weekly street and parking lot sweeping resulted in the most
significant improvements in summer average total phosphorus concentrations in all three lakes. However,
the improvements do not bring the summer average phosphorus concentrations below the 60 µg/L
standard. Additional information on model results for the Weekly Watershed-wide Street Sweeping
scenario, including plots of the in-lake model results for 2016 and 2017 for North Cornelia and the in-lake
model results for South Cornelia and Lake Edina for all modeled years, is provided in Appendix D.
8.2 Lake Response to Internal Loading Management
Two in-lake management scenarios were explicitly modeled for North and South Cornelia: (1) curly-leaf
pondweed management, and (2) alum sediment treatments. No in-lake management processes were
modeled for Lake Edina since the calibration models indicated that internal loading is not a major source
of phosphorus loading to the lake. However, internal management efforts applied to North and South
Cornelia have downstream influences on water quality in Lake Edina.
8.2.1 Changes in In-lake Phosphorus Concentrations
As described in Section 7.0, the calibrated in-lake models were used to predict the changes in total
phosphorus concentration in each of the lakes throughout the summer months as a result of the various
in-lake management activities. As an example, Figure 8-5 shows the predicted in-lake phosphorus
concentration changes in North Cornelia, South Cornelia, and Lake Edina in 2017 when the two internal
loading management strategies, and combinations thereof, were applied to the models. The black, dashed
line represents North Cornelia’s phosphorus concentrations under existing conditions and the red, dashed
line represents the MPCA’s water quality standard for shallow lakes (60 µg/L).
The purple line represents the in-lake phosphorus concentrations that resulted when an alum treatment
was applied to the sediments for North Cornelia in 2017. During late summer and early fall is when the
alum treatment resulted in the largest percent changes in phosphorus concentrations due to the
significant internal loading that occurred during that period under existing conditions. The effect of the
alum treatment is not as notable in the spring as the benefits are masked by the extensive curly-leaf
pondweed die-off that occurs in late spring/early summer.
The green line represents the in-lake phosphorus concentrations that resulted from applying curly-leaf
pondweed management to the modeling. During late-spring and early-summer, when curly-leaf
pondweed typically dies and decays, is when the curly-leaf pondweed treatments resulted in the largest
percent changes in in-lake phosphorus concentrations. By the end of summer and into fall the in-lake
phosphorus concentrations tended towards existing conditions due to significant internal loading from
the sediments.
The greatest changes to the in-lake phosphorus concentrations resulted from combined internal
management efforts. The yellow line on Figure 8-5 represents the predicted total phosphorus
concentrations when an alum treatment is conducted and curly-leaf pondweed is managed.
97
Appendix E contains additional figures showing the effects of internal loading management on in-lake
phosphorus concentrations in North Cornelia, South Cornelia, and Lake Edina for each modeled year.
98
Figure 8-5 In-Lake Phosphorus Concentration Changes that resulted from internal management efforts in North Cornelia, South Cornelia, and Lake Edina in 2017
North Cornelia
South Cornelia
Lake Edina
99
8.2.2 Summer Average Total Phosphorus Concentrations
Figure 8-6, Figure 8-7, and Figure 8-8 summarize the current and predicted summer average (June
through September) total phosphorus concentrations in North and South Lake Cornelia and Lake Edina
based on the modeled internal load management scenarios and combinations thereof.
In-lake modeling shows that combined internal management of alum treatment and curly-leaf pondweed
management has the potential to reduce South Cornelia’s summer average phosphorus concentration to
meet MPCA’s water quality standard (model year 2015). Significant improvements were also predicted for
South Cornelia in 2016 and 2017, where the summer average phosphorus concentrations were reduced by
52% and 62% respectively. While notable decreases in summer average phosphorus concentrations were
also observed for North Cornelia (23% to 53%), internal management efforts alone were not sufficient to
reduce in-lake phosphorus concentrations within range of MPCA’s water quality standard. Improvements
in water quality may exceed model predictions if curly-leaf pondweed is controlled and the native aquatic
plant population thrives, as these species can uptake phosphorus from the water column and compete
with phytoplankton growth by increased shading (i.e., reduced light availability).
The internal load management efforts applied to North and South Cornelia also result in significant
improvements in Lake Edina’s water quality. Under combined internal management efforts (alum
treatment and curly-leaf pondweed management in Lake Cornelia), summer average phosphorus
concentrations ranged from 63-76 µg/L in Lake Edina for the three years modeled. Comparing these
values to existing conditions, summer average phosphorus concentrations were reduced by 22% to 41%
through upstream internal management.
100
Figure 8-6 North Lake Cornelia In-Lake Summer Average Phosphorus Concentration Summary for Internal Management Efforts
Figure 8-7 South Lake Cornelia In-Lake Summer Average Phosphorus Concentration
Summary for Internal Management Efforts
101
Figure 8-8 Lake Edina In-Lake Summer Average Phosphorus Concentration Summary for Internal Management Efforts
8.2.3 Phosphorus Loading Reductions
The in-lake total phosphorus concentrations of North and South Cornelia and Lake Edina are influenced
over time through loading reductions. Table 8-2 summarizes the total phosphorus loads and percent load
reductions from the two internal load management strategies, and combinations thereof, in comparison
with current conditions. Overall, under the combined scenario where both alum treatment and curly-leaf
pondweed management are conducted, a 19% to 44% reduction in total phosphorus loads to North
Cornelia is predicted based on the years analyzed in this study. For South Cornelia, the predicted
reduction in loading is approximately 48% to 59% based on the modeled years. For Lake Edina, the
predicted reduction in phosphorus loads due to internal management efforts in the upstream lakes is
approximately 22% to 33%.
The level of reduction observed in the total phosphorus load due to internal management is largely
dependent on the amount of precipitation and associated runoff that occurs in a given year. For example,
model year 2017 had the least amount of precipitation over the growing season. Therefore, a larger
percentage of the phosphorus loading that occurred over that model year was due to internal sediment
loading and curly-leaf die-off and decay. Thus, reductions in internal loading during that model year will
have a larger percent change in the total load than for other model years that had more precipitation.
While the in-lake phosphorus concentrations and loading reductions show positive responses to internal
loading management, it is important to note that internal loading improvements are only temporary if
external loading is not addressed. Internal loading has become an issue in North and South Cornelia due
102
in part to the significant external nutrient loads that the lakes have received over time. While internal
loading management can help address the nutrients already present in the sediment, the short-term and
long-term effectiveness of the treatments can be reduced if external load reductions aren’t also achieved.
103
Table 8-2 Total phosphorus (TP) load reductions resulting from internal loading management efforts
Lake Model
Year
Current
Load
(lbs TP)
Lake Cornelia Alum Treatment
Lake Cornelia
Curly-leaf Pondweed
Treatment
Lake Cornelia Alum + Curly-leaf
Pondweed Treatments
Model Load
(lbs TP)
Percent
Reduction (%)
Model Load
(lbs TP)
Percent
Reduction (%)
Model Load
(lbs TP)
Percent
Reduction (%)
North
Cornelia
2015 456 402 12% 384 16% 330 28%
2016 489 397 19% 4891 0% 397 19%
2017 673 451 33% 598 11% 376 44%
South
Cornelia
2015 442 356 19% 309 30% 223 50%
2016 444 335 25% 342 23% 233 48%
2017 525 356 32% 386 26% 216 59%
Lake Edina
2015 306 274 10% 267 13% 239 22%
2016 430 371 14% 364 15% 305 29%
2017 410 356 13% 330 20% 276 33%
1 A phosphorus loading specifically attributable to curly-leaf pondweed die-off was not observed in North Cornelia in 2016; therefore, treating for curly-leaf pondweed in North
Cornelia in 2016 has no effect on the total phosphorus load.
104
8.3 Lake Responses to Combined Internal and External
Management
The most effective approach to managing lakes with significant internal and external sources of
phosphorus is to implement a combination of internal and external management strategies. For lakes that
have been exposed to significant external nutrient loading for extended periods of time, appreciable
sediment and nutrients have accumulated in the lake bottom sediments. As nutrients continue to build-up
over time, internal loading potential is heightened exasperating water quality conditions in the lake. This
section looks at the effects of combined external and internal loading management on in-lake total
phosphorus concentrations. This section also discusses the effects of combined management on total
phosphorus summer averages and the percent changes to phosphorus loads.
The combined management scenarios analyzed for this UAA study are described below. Modeling of
combined management scenarios was complex and time intensive due to the chain of lakes and number
of modeled years. Accordingly, it was necessary to limit the number combination scenarios evaluated. Two
of the four watershed management BMP scenarios (infiltration BMPs on commercial properties and spent
lime/CC17 treatment chamber) were explicitly modeled to evaluate the effects of combined internal and
external load reductions. These two watershed management scenarios represent a wide range of external
phosphorus load reduction.
1) Infiltration BMPs on commercial properties with:
a. Alum treatments in North and South Cornelia
b. Curly-leaf pondweed treatments in North and South Cornelia
c. Alum and curly-leaf pondweed treatments in North and South Cornelia
2) Spent Lime/CC17 treatment chamber with:
a. Alum treatments in North and South Cornelia
b. Curly-leaf pondweed treatments in North and South Cornelia
c. Alum and curly-leaf pondweed treatments in North and South Cornelia
8.3.1 Changes in In-lake Phosphorus Concentrations
The P8 and calibrated in-lake models were used to predict the changes in total phosphorus concentration
in each of the lakes throughout the summer months as a result of the various combined watershed and
internal load management activities. Appendix F contains figures showing the effect of different combined
management efforts on in-lake phosphorus concentrations in North Cornelia, South Cornelia, and Lake
Edina for each modeled year.
8.3.2 Summer Average Total Phosphorus Concentrations
8.3.2.1 Infiltration BMPs on Commercial Property + Internal Load Management
Table 8-3 provides a comparison of the summer average (June through September) in-lake phosphorus
concentrations under existing conditions and with a variety of combinations of internal load management
105
and widespread infiltration BMPs. The combination of widespread implementation of infiltration BMPs
and internal load management (alum treatment and curly-leaf pondweed management) represents the
highest level of management modeled. Under this scenario, the reduction in summer average total
phosphorus concentration ranges from 32% to 56% for North Cornelia. Despite this level of reduction,
summer phosphorus concentration averages did not fall below 60 µg/L for any of the years modeled in
this study. In South Cornelia, the summer average total phosphorus concentration met the MPCA’s water
quality standard for all three modeled years under the highest level of management. The reduction in
summer average total phosphorus concentration under this scenario ranged from 61% to 69%. In Lake
Edina, the reduction in summer average total phosphorus concentration under this scenario ranged from
24% to 48%. In 2017, the predicted summer average total phosphorus concentration was below MPCA’s
water quality standard.
106
Table 8-3 Comparison of total phosphorus summer average concentrations under existing conditions to combined management (internal and commercial infiltration BMPs) conditions
Lake Model
Year
Total Phosphorus Summer Average Concentration (µg/L)
Current
Conditions
Internal Management External
Management Internal + External Management
Lake Cornelia
Alum Treatment
Lake Cornelia
Curly-leaf Treatment
Lake Cornelia
Alum + Curly-leaf
Treatments
Commercial
Infiltration
BMPs
Commercial
Infiltration BMPs +
Alum Treatment
Commercial
Infiltration BMPs +
Curly-leaf Treatment
Commercial
Infiltration BMPs +
Alum + Curly-leaf
Treatments
North
Cornelia
2015 163 142 132 111 156 132 121 97
2016 126 97 1261 97 119 86 1191 86
2017 185 142 160 87 193 110 164 81
South
Cornelia
2015 118 93 82 60 102 79 68 46
2016 152 120 105 73 152 118 93 59
2017 178 93 119 67 175 124 105 56
Lake Edina
2015 93 83 81 72 85 78 78 71
2016 113 98 91 76 102 90 80 68
2017 107 83 81 63 90 78 67 56
1 A curly-leaf pondweed phosphorus loading was not observed in North Cornelia in 2016; therefore, treating for curly-leaf pondweed in North Cornelia in 2016 has no effect on the total phosphorus summer average concentrations.
107
8.3.2.2 Spent Lime/CC17 Treatment Chamber + Internal Load Management
Table 8-4 provides a comparison of the summer average (June through September) in-lake phosphorus
concentrations under existing conditions and with a variety of combinations of internal load management
and installation of a spent lime treatment chamber. Through the combination of a spent lime/CC17
treatment chamber with internal management efforts, larger percent reductions in the in-lake phosphorus
concentrations were observed than modeling with an external management or internal management
scenario on its own. Under this scenario (spent lime/CC17 treatment chamber + curly-leaf pondweed
management + alum treatment) the reduction in summer average phosphorus concentrations ranged
from 27% to 55% for North Cornelia. Despite this level of reduction, summer phosphorus concentration
averages did not fall below 60 µg/L for any of the years modeled in this study. In South Cornelia, the
reduction in summer average phosphorus concentrations ranged from 51% to 64% reduction for this
scenario. For model year 2015, South Cornelia’s total phosphorus summer average fell under the MPCA’s
water quality standard. Under this combined scenario, the reduction in summer average total phosphorus
concentration in Lake Edina ranged from 23% to 43%. Despite this level of reduction, Lake Edina summer
phosphorus concentration averages did not fall below 60 µg/L for any of the years modeled under this
scenario.
108
Table 8-4 Comparison of total phosphorus summer average concentrations under existing conditions to combined management (internal and spent lime/CC17 treatment chamber) conditions
Lake Model
Year
Total Phosphorus Summer Average Concentration (µg/L)
Current
Conditions
Internal Management External
Management Internal + External Management
Lake Cornelia
Alum Treatment
Lake Cornelia
Curly-leaf
Treatment
Lake Cornelia
Alum + Curly-leaf
Treatments
Spent Lime/CC17
Treatment Cell
Spent Lime/CC17
Treatment Cell +
Alum Treatment
Spent Lime/CC17
Treatment Cell +
Curly-leaf Treatment
Spent Lime/CC17
Treatment Cell +
Alum + Curly-leaf
Treatments
North
Cornelia
2015 163 142 132 111 157 136 126 105
2016 126 97 1261 97 121 92 1211 92
2017 185 142 160 87 180 107 155 82
South
Cornelia
2015 118 93 82 60 112 90 79 58
2016 152 120 105 73 149 117 102 70
2017 178 93 119 67 175 123 116 64
Lake Edina
2015 93 83 81 72 91 82 80 71
2016 113 98 91 76 111 96 89 74
2017 107 83 81 63 106 88 80 62
1 A curly-leaf pondweed phosphorus loading was not observed in North Cornelia in 2016; therefore, treating for curly-leaf pondweed in 2016 has no effect on the total phosphorus load.
109
8.3.3 Phosphorus Loading Reductions
The in-lake total phosphorus concentrations of North and South Cornelia and Lake Edina are influenced
over time through loading reductions. Table 8-5 provides a summary of the changes in total phosphorus
loads to the lakes through the combined internal (alum treatments, curly-leaf pondweed treatments) and
external management (commercial infiltration BMPs, spent lime/CC17 treatment cell) efforts.
Figure 8-9 shows the total phosphorus loads (in pounds) contributed to North Cornelia in 2017 after the
various watershed and in-lake management practices are applied. In 2017, the existing conditions
phosphorus load to North Cornelia was approximately 673 pounds, whereas when applying a
management scenario that included commercial infiltration BMPs, alum treatment, and managing for
curly-leaf pondweed (brown bar) the remaining load to North Cornelia in 2017 is approximately
251 pounds (a 63% reduction). Figure 8-10 shows the total phosphorus loads (in pounds) contributed to
North Cornelia in 2017 after combinations of internal management and the spent lime/CC17 treatment
chamber are applied to the models. The combination of a spent lime/CC17 treatment chamber with
internal management efforts (alum treatment and curly-leaf pondweed management) results in a 48%
phosphorus load reduction (in 2017).
A comprehensive summary of all of the loading bar graphs for the three lakes and three model years can
be found in Appendix G.
Figure 8-9 Remaining Total Phosphorus Load to North Cornelia in 2017 with various combinations of internal and external commercial infiltration management
110
Figure 8-10 Remaining Total Phosphorus Load to North Cornelia in 2017 with various combinations of internal and external spent lime/CC17 management
111
Table 8-5 Total phosphorus load reductions summary for combined management (internal and external) conditions
Lake Model
Year
Current Load
(lbs TP)
Total Phosphorus Load Percent Reductions (%)
Internal Management External Management Internal + External Management
Lake
Cornelia
Alum
Treatment
Lake
Cornelia
Curly-leaf
Treatment
Lake
Cornelia
Alum +
Curly-leaf
Treatments
Commercial
Infiltration
BMPs
Spent
Lime/CC17
Treatment
Cell
Commercial
Infiltration
BMPs +
Alum
Treatment
Commercial
Infiltration
BMPs + Curly-
leaf Treatment
Commercial
Infiltration
BMPs +
Alum + Curly-
leaf
Treatments
Spent
Lime/CC17
Treatment
Cell + Alum
Treatment
Spent
Lime/CC17
Treatment
Cell + Curly-
leaf Treatment
Spent
Lime/CC17
Treatment
Cell + Alum +
Curly-leaf
Treatments
North
Cornelia
2015 456 12% 16% 28% 28% 4% 38% 42% 53% 16% 20% 32%
2016 489 19% 0%1 19% 31% 4% 48% 31%1 48%1 23% 4%1 23%1
2017 673 33% 11% 44% 24% 3% 53% 33% 63% 36% 14% 47%
South
Cornelia
2015 442 18% 30% 50% 23% 2% 41% 51% 69% 21% 32% 52%
2016 444 25% 23% 48% 22% 2% 44% 45% 67% 27% 25% 50%
2017 525 32% 26% 59% 23% 2% 50% 48% 74% 34% 28% 61%
Lake Edina
2015 306 10% 13% 22% 29% 2% 35% 36% 42% 12% 14% 23%
2016 430 14% 15% 29% 25% 1% 35% 38% 47% 15% 17% 30%
2017 410 13% 20% 33% 29% 1% 36% 43% 50% 14% 20% 34%
1 A curly-leaf pondweed phosphorus loading was not observed in North Cornelia in 2016; therefore, treating for curly-leaf pondweed in North Cornelia in 2016 has no effect on the total phosphorus load.
112
8.4 Management Alternatives Summary
Section 8.1, 8.2, and Section 8.3 discussed the various management alternatives that were modeled for
North Cornelia, South Cornelia, and Lake Edina. These management alternatives included:
• External (Watershed) Management
o Commercial Infiltration BMPs (treatment in North Cornelia and Lake Edina watersheds)
o Commercial Filtrations BMPs (treatment in North Cornelia and Lake Edina watersheds)
o Spent Lime/CC17 treatment chamber downstream of Swimming Pool Pond (North
Cornelia watershed)
o Weekly public street and private parking lot sweeping (North Cornelia, South Cornelia,
and Lake Edina watersheds)
• Internal Management in North and South Cornelia
o Alum Sediment Treatments
o Curly-leaf Pondweed Treatments
o Combined alum sediment and curly-leaf pondweed treatments
• Combined Internal and External Management
o Commercial Infiltration BMPs (treatment in North Cornelia and Lake Edina watersheds)
with:
Alum Sediment Treatments (North and South Cornelia)
Curly-leaf Pondweed Treatments (North and South Cornelia)
Combined alum sediment and curly-leaf pondweed treatments (North and South
Cornelia)
o Spent Lime/CC17 treatment cell downstream of Swimming Pool Pond (North Cornelia
watershed) with:
Alum Sediment Treatments (North and South Cornelia)
Curly-leaf Pondweed Treatments (North and South Cornelia)
Combined alum sediment and curly-leaf pondweed treatments (North and South
Cornelia)
Table 8-6 provides a summary of the summer average total phosphorus concentrations observed for each
modeled alternative. Table 8-7 provides a summary of the total phosphorus loads observed for each
modeled alternative. These loading values provided in the table represent the remaining load to the lakes
after the listed treatment is applied to the models. As discussed in previous sections, comparing the
phosphorus load reduction results to the summer average total phosphorus concentration results were
not always straightforward. When a proposed management alternative included water balance
adjustments (e.g., reduced runoff through enhanced infiltration in the watershed), reductions in
phosphorus loads did not always correspond to comparable reductions in summer average total
phosphorus concentrations. If the volume of water in the lake was reduced to a greater extent than the
113
total phosphorus load in the lake, the summer average concentration could increase or not decrease as
much as anticipated. For example, when viewing the management alternative that included commercial
infiltration BMPs with alum sediment treatments, the average total phosphorus load reduction to North
Cornelia was approximately 47% for the three modeled years (existing average load = 539 pounds;
management average load = 285 pounds). Of the management alternatives investigated in this study,
commercial infiltration BMPS with alum sediment treatment resulted in the second highest reduction in
total phosphorus loads to North Cornelia. The only management alternative that resulted in a greater
reduction in total phosphorus loads was the alternative that included commercial infiltration BMPS with
combined alum sediment and curly-leaf pondweed treatments. However, the percent reduction in the
mean summer average total phosphorus concentration of the three modeled years for the commercial
infiltration BMPS with alum sediment treatments scenario was approximately 31% (existing average
summer average concentration = 158 µg/L; management average summer average
concentration = 109 µg/L). Of the management alternatives investigated in this study, commercial
infiltration BMPS with alum sediment treatment resulted in the fourth highest reduction in total
phosphorus summer average concentrations in North Cornelia. Management alternatives, such as alum
sediment + curly-leaf pondweed treatments, which resulted in lower load reductions to North Cornelia,
resulted in higher percent reductions in summer average phosphorus concentrations.
Discrepancies in North Cornelia’s in-lake concentrations and total phosphorus load reductions occurred in
the models when lake volume reductions arose from enhanced infiltration and when all internal loading
was not managed. Continuing with the example above, in-lake phosphorus concentrations where not
reduced to the same extent as phosphorus loads because curly-leaf loading was permitted in the models
under this management scenario. Commercial infiltration BMPs reduced the volume of runoff reaching
North Cornelia and as a result, the volume of water in North Cornelia was less under proposed conditions.
When the lake volume was reduced and when the same loading (as calibrated for existing conditions)
from curly-leaf pondweed death/decay occurred, the model resulted in increased concentrations during
this time period. This suggests that if water balance adjustments are expected for the lakes, management
of internal loads will be imperative to control in-lake concentrations. Water balance adjustments would
not only include runoff reductions due to infiltration practices, but would also include water volume
changes due to climate impacts. Adjustments to the water balance from climate could include more
frequent, high-intensity precipitation events coupled with prolonged dry periods. Alterations to snowpack
accumulations is another example. This was observed in 2015 when lower-than-average snowpack
accumulated resulting in lower-than-average spring snowmelt and reduced lake levels.
Table 8-6 and Table 8-7 show the relative changes to in-lake phosphorus concentrations and phosphorus
loads that could occur from applying various management scenarios. The next section will look into the
costs and benefits of the different management alternatives.
114
Table 8-6 Comparison of total phosphorus summer average concentrations for all modeled management scenarios
Lake Model
Year
Total Phosphorus Summer Average Concentration (µg/L)
Current
Conditions
Internal Management External Management Internal + External Management
Lake
Cornelia
Alum
Treatment
Lake
Cornelia
Curly-leaf
Treatment
Lake
Cornelia
Alum +
Curly-leaf
Treatments
Commercial
Infiltration
BMPs
Spent
Lime/CC17
Treatment
Cell
Commercial
Filtration
BMPs
Weekly
Street/Lot
Sweeping
Commercial
Infiltration
BMPs + Alum
Treatment
Commercial
Infiltration
BMPs + Curly-
leaf Treatment
Commercial
Infiltration BMPs +
Alum + Curly-leaf
Treatments
Spent Lime/CC17
Treatment Cell +
Alum Treatment
Spent
Lime/CC17
Treatment
Cell + Curly-
leaf Treatment
Spent Lime/CC17
Treatment Cell +
Alum + Curly-leaf
Treatments
North
Cornelia
2015 163 142 132 111 156 157 147 141 132 121 97 136 126 105
2016 126 97 1261 97 119 121 113 107 86 1191 86 92 1211 92
2017 185 142 160 87 193 180 174 170 110 164 81 107 155 82
Average 158 127 139 98 156 153 145 139 109 134 88 112 134 93
South
Cornelia
2015 118 93 82 60 102 112 110 105 79 68 46 90 79 58
2016 152 120 105 73 152 149 144 137 118 93 59 117 102 70
2017 178 93 119 67 175 175 172 165 124 105 56 123 116 64
Average 149 102 102 67 143 145 142 136 107 89 53 110 99 64
Lake
Edina
2015 93 83 81 72 85 91 86 72 78 78 71 82 80 71
2016 113 98 91 76 102 111 107 93 90 80 68 96 89 74
2017 107 83 81 63 90 106 103 91 78 67 56 88 80 62
Average 104 88 84 70 92 103 99 85 82 75 65 89 83 69
1 A curly-leaf pondweed phosphorus loading was not observed in North Cornelia in 2016; therefore, treating for curly-leaf pondweed in North Cornelia in 2016 has no effect on the total phosphorus summer average concentrations.
115
Table 8-7 Comparison of total phosphorus loads for all modeled management scenarios
Lake Model
Year
Total Phosphorus Load (lbs)
Current
Conditions
Internal Management External Management Internal + External Management
Lake
Cornelia
Alum
Treatment
Lake
Cornelia
Curly-leaf
Treatment
Lake
Cornelia
Alum +
Curly-leaf
Treatments
Commercial
Infiltration
BMPs
Spent
Lime/CC17
Treatment
Cell
Commercial
Filtration
BMPs
Weekly
Street/Lot
Sweeping
Commercial
Infiltration
BMPs + Alum
Treatment
Commercial
Infiltration
BMPs + Curly-
leaf Treatment
Commercial
Infiltration BMPs +
Alum + Curly-leaf
Treatments
Spent Lime/CC17
Treatment Cell +
Alum Treatment
Spent
Lime/CC17
Treatment
Cell + Curly-
leaf Treatment
Spent Lime/CC17
Treatment Cell +
Alum + Curly-leaf
Treatments
North
Cornelia
2015 456 402 384 330 329 438 408 390 281 264 216 384 366 312
2016 489 397 4891 397 337 467 432 411 256 3371 256 375 4671 375
2017 673 451 598 376 514 652 625 603 317 448 251 430 577 355
Average 539 417 490 368 393 519 488 468 285 350 241 396 470 347
South
Cornelia
2015 442 356 309 223 341 433 416 397 262 218 139 347 299 213
2016 444 335 342 233 347 433 419 396 250 245 148 324 331 222
2017 525 356 386 216 406 516 507 485 265 275 136 347 377 207
Average 470 349 346 224 365 461 447 426 259 246 141 339 336 214
Lake
Edina
2015 306 274 267 239 217 299 288 237 199 195 176 270 263 235
2016 430 371 364 305 321 424 407 347 280 267 226 365 358 299
2017 410 356 330 276 290 405 395 340 261 233 205 352 326 272
Average 382 334 320 273 276 376 363 308 247 232 202 329 316 269
1 A curly-leaf pondweed phosphorus loading was not observed in North Cornelia in 2016; therefore, treating for curly-leaf pondweed in North Cornelia in 2016 has no effect on the total phosphorus load.
116
9.0 Cost-Benefit of Management Efforts
9.1 Opinions of Probable Cost for Modeled Scenarios
Planning-level opinions of probable cost were developed for each of the evaluated management
alternatives. These opinions of cost are intended to provide assistance in evaluating and comparing
alternatives and should not be assumed as absolute values. The estimated costs are summarized in
Table 9-1. Detailed opinions of probable cost are included in Appendix H.
The opinions of probable cost summarized in Table 9-1 generally correspond to standards established by
the Association for the Advancement of Cost Engineering (AACE). Class 5 feasibility-level opinions of costs
were used for most of the management practices based on the limited project definition, wide-scale use
of parametric models to calculate estimated costs (i.e., making extensive use of order-of-magnitude costs
from similar projects), and uncertainty, with an acceptable range of between -30% and +50% of the
estimated project cost. The opinions of probable cost for the alum treatment and curly-leaf pondweed
treatments are considered Class 2 level cost estimates, with an acceptable range of between -10% and
+20%.
Table 9-1 Planning-level cost estimates for modeled management alternatives
Description Planning-Level
Cost Estimate1 Planning-Level Cost Range
Estimated
Life of
Project
Lake Cornelia Alum Treatment $161,000 $145,000 - $194,000 5 years
Lake Cornelia Curly-leaf Pondweed
Management (annual) $12,000 $11,000 - $15,000 1 year
Commercial Infiltration BMPs $15,855,000 $11,100,000 - $23,780,000 30 years
Commercial Filtration BMPs $15,855,000 $11,100,000 - $23,780,000 30 years
Spent Lime/CC17 Treatment
Chamber $588,000 $412,000 - $882,000 30 years
Weekly Street/Lot Sweeping $1,060,0002 $742,000 - $1,590,000 10 years
1 Planning-level cost estimates do not include annual costs for operations and maintenance, with exception of the
weekly street sweeping which includes annual operational costs.
2 Cost estimate includes $727,000 capital costs plus $333,000 annual operation costs
117
9.1.1 Cost Details for Modeled Scenarios
9.1.1.1 Sediment Alum Treatments
The opinion of cost for the alum treatment of the North and South Lake Cornelia sediments is based on
correspondence with an alum application contractor and previous project applications. The primary
assumptions for the alum treatment opinion of costs are:
• Dose equivalent of 29,238 gallons of alum applied to North Cornelia
• Dose equivalent of 16,431 gallons of alum applied to South Cornelia
9.1.1.2 Curly-leaf Pondweed Management
The opinion of cost for the management of curly-leaf pondweed in North Lake Cornelia and South Lake
Cornelia is based on recent herbicide treatment efforts conducted by the City of Edina. The cost estimate
assumes that management will be coordinated by the City of Edina (as is conducted currently), and that a
dose of 5.0 ppm active ingredient will be used for 7.4 acres of treatment area. The opinion of cost does
not include costs related to permitting or monitoring and reporting that may be required as part of
permitting (e.g., temperature monitoring, herbicide residual monitoring, follow-up aquatic plant surveys,
and/or water quality monitoring).
9.1.1.3 Commercial Infiltration and Filtration BMPS
The opinion of costs for the commercial infiltration and filtration BMPs used a planning-level unit cost of
$15 per cubic foot of storage. This unit cost assumes that all infiltration and filtration BMPs are
constructed as subsurface treatment areas. A construction contingency of 30% and an engineering and
design percentage of 30% was applied to the construction cost. Maintenance costs were estimated to be
approximately 10% of the total project cost.
9.1.1.4 Spent Lime/CC17 Treatment Chamber
An itemized opinion of cost was developed for the Spent Lime/CC17 Treatment Chamber based on
correspondence with manufacturers and previous project implementation. A construction contingency of
30% and an engineering and design percentage of 30% was applied to the construction cost. An
estimated accuracy range of -30% to 50% was applied to the final project cost due to the limited design
work completed for the treatment chamber. Maintenance costs assumed periodic debris removal and that
the treatment material (spent lime and CC17 aggregate) would need full replacement every 5 years.
9.1.1.5 Weekly Street and Lot Sweeping
The weekly street and lot sweeping unit costs were based, in part, on the values summarized in the City of
Edina, MN Street Sweeping Management Plan (Emmons & Olivier Resources, Inc. (EOR), 2015). The
assumptions used to develop the opinion of cost are:
• The cost of a new Crosswind 4-Wheel Regenerative Air Sweeper is approximately $210,000 and
the buy-back cost after 10 years of use is approximately $20,000.
118
• Within the North Lake Cornelia, South Lake Cornelia, and Lake Edina watersheds there is
approximately 67.3 miles of sweepable public street curb and 80.4 miles of sweepable parking lot
area.
• Sweeper operation speed is estimated at 4.5 miles per hour and average fuel consumption is
5 miles per gallon.
• The sweeping path width of a high efficiency, two-sided broom sweeper with a pick-up head is
approximately 12 feet.
• 1.5 hours of labor is needed for every 4 hours of sweeping time.
• Total transit (brush off) is about 3 times the total amount of swept miles.
• The maximum number of hours worked in one week by a single worker is 40 hours.
• Sweeping occurs on a weekly basis May through November (approximately 30 weeks).
• Labor cost = $75/hour
• Fuel cost = $3.00/gallon
• Sweeper Maintenance cost = $4,800/year
• The City of Edina is responsible for the operations and maintenance of the street sweeping
program.
• The City of Edina already owns one high efficiency sweeper.
Using the outlined assumptions, to sweep all of the public streets and private parking lots in the Lake
Cornelia and Lake Edina watersheds on a weekly basis, four high efficiency regenerative air sweepers
would be needed. Since the City of Edina currently owns one high-efficiency regenerative air sweeper, the
opinion of cost assumes three additional high-efficiency regenerative sweepers would be purchased.
Vehicle maintenance, labor, and fuel costs as summarized as annual costs.
9.2 Cost-Benefit Analysis for Modeled Scenarios
The management strategies considered to help improve water quality in Lake Cornelia and Lake Edina are
wide ranging in type, scale, cost, and effectiveness. Some strategies include structural BMPs with large,
upfront capital costs, whereas others are more programmatic or may require periodic or annual repeat. To
account for these variations, a comparison of cost-benefit of the potential management strategies was
conducted. Results of the cost-benefit analysis help to understand the value derived and associated costs,
for each management practice and combinations thereof.
Estimated costs for the evaluated management activities were annualized to help compare the cost-
benefit ratio. The annualized cost for each management alternative is based on anticipated maintenance,
replacement costs, and anticipated useful life-span of the projects/treatments. A 3% inflation rate was
assumed. The annualized cost for each alternative is calculated as the value of ‘n’ equal, annual payments,
119
where ‘n’ is the anticipated useful life-span of the project or treatment. The annualized cost estimates for
each management alternative are summarized in Table 9-2.
For the cost-benefit analysis, two approaches were considered to quantify the benefits of each of the
evaluated management activities. The first approach quantifies the benefit in terms of phosphorus
removed (in pounds) during the time period of April through September (i.e., phosphorus that did not
enter the lake system as a result of the management practice). The second approach quantifies the benefit
in terms of reduced summer average total phosphorus concentration in the respective lakes (June
through September). Table 9-2 summarizes the results of the cost-benefit analysis for both of these cost-
benefit approaches, since total phosphorus load reductions to the lakes did not always result in the same
level of in-lake phosphorus concentration reductions.
Figure 9-1 compares the cost-benefit of each of the individual modeled management activities in terms of
lake water quality improvement (reductions in respective in-lake summer average total phosphorus
concentrations in µg/L). The internal loading management alternatives (curly-leaf pondweed treatment
and alum treatment in Lake Cornelia) result in the lowest annualized costs per unit reduction in summer
average in-lake phosphorus concentrations. Of the four watershed management alternatives evaluated in
this UAA study, the spent lime/CC17 treatment chamber is the most cost effective, with the lowest
annualized cost per unit benefit for all three lakes. While the street sweeping alternative has higher costs
per unit benefit in comparison with the spent lime/CC17 treatment chamber, it has a much lower cost per
unit benefit than the infiltration and filtration BMPs on commercial property scenarios. Estimated costs for
these scenarios assumed that all infiltration/filtration practices would be subsurface, which corresponds
with stormwater management trends observed in recent years as redevelopment has occurred throughout
the watershed. Promoting impervious area reduction (less parking lot) and installation of surface BMPs
could reduce costs and improve the cost-benefit.
As described in Section 8.3, the most effective approach to managing lakes with significant internal and
external sources of phosphorus is to implement a combination of internal and external management
strategies. Figure 9-2 compares the cost-benefit of the combined internal and external watershed
management activities that were modeled in terms of lake water quality improvement. As compared to
Figure 9-1, the watershed management activities have a significantly lower cost per unit benefit when
combined with internal loading activities. Of the combined internal and external load management
scenarios modeled, the spent lime/CC17 treatment chamber with curly-leaf pondweed management and
alum treatment provides the lowest annualized cost per unit benefit in lake water quality for all three
lakes.
120
Table 9-2 Cost-benefit summaries for North Cornelia, South Cornelia, and Lake Edina for modeled management alternatives
Lake Description Management
Type
Estimated
Annualized
Cost
Average Pounds
of TP Load
Removed
(April - Sept)
Annualized Cost
per Pound of TP
Removed
(April - Sept)
Summer Average
TP Concentration
Reduction in
µg/L
(June - Sept)
Annualized Cost
per µg/L
Reduction in
Summer Average
TP Concentration
(June - Sept)
North
Cornelia
Lake Cornelia Alum Treatment Internal $35,000 123 $300 31 $1,100
Lake Cornelia Curly-leaf Treatment Internal $12,000 49 $200 19 $600
Lake Cornelia Alum + Curly-leaf Treatments Combined $47,000 172 $300 59 $800
Commercial Infiltration BMPs External $2,394,000 146 $16,400 71 $342,000
Spent Lime/CC17 Treatment Cell External $31,000 20 $1,500 5 $5,900
Commercial Filtration BMPs External $2,394,000 51 $46,900 13 $180,100
Weekly Street/Lot Sweeping External $418,000 71 $5,900 19 $22,400
Commercial Infiltration BMPs + Alum
Treatment Combined $2,429,000 255 $9,500 49 $49,900
Commercial Infiltration BMPs + Curly-leaf
Treatment Combined $2,429,000 190 $12,800 23 $103,500
Commercial Infiltration BMPs + Alum +
Curly-leaf Treatments Combined $2,441,000 298 $8,200 70 $34,800
Spent Lime/CC17 Treatment Cell + Alum
Treatment Combined $66,000 143 $500 46 $1,400
Spent Lime/CC17 Treatment Cell + Curly-
leaf Treatment Combined $43,000 69 $600 24 $1,800
Spent Lime/CC17 Treatment Cell + Alum +
Curly-leaf Treatments Combined $78,000 192 $400 65 $1,200
South
Cornelia
Lake Cornelia Alum Treatment Internal $35,000 121 $300 47 $700
Lake Cornelia Curly-leaf Treatment Internal $12,000 125 $100 47 $300
Lake Cornelia Alum + Curly-leaf Treatments Combined $47,000 246 $200 83 $600
Commercial Infiltration BMPs External $2,394,000 106 $22,700 6 $382,900
Spent Lime/CC17 Treatment Cell External $31,000 10 $3,200 4 $7,900
Commercial Filtration BMPs External $2,394,000 23 $104,100 7 $330,100
Weekly Street/Lot Sweeping External $418,000 44 $9,400 14 $30,800
Commercial Infiltration BMPs + Alum
Treatment Combined $2,429,000 211 $11,500 42 $57,400
Commercial Infiltration BMPs + Curly-leaf
Treatment Combined $2,429,000 224 $10,800 61 $40,100
Commercial Infiltration BMPs + Alum +
Curly-leaf Treatments Combined $2,441,000 329 $7,400 96 $25,500
Spent Lime/CC17 Treatment Cell + Alum
Treatment Combined $66,000 131 $500 39 $1,700
Spent Lime/CC17 Treatment Cell + Curly-
leaf Treatment Combined $43,000 135 $300 50 $900
Spent Lime/CC17 Treatment Cell + Alum +
Curly-leaf Treatments Combined $78,000 256 $300 85 $900
Lake
Edina
Lake Cornelia Alum Treatment Internal $35,000 48 $700 16 $2,200
Lake Cornelia Curly-leaf Treatment Internal $12,000 62 $200 20 $600
Lake Cornelia Alum + Curly-leaf Treatments Combined $47,000 109 $400 34 $1,400
Commercial Infiltration BMPs External $2,394,000 106 $22,600 12 $198,600
Spent Lime/CC17 Treatment Cell External $31,000 6 $5,200 2 $18,000
Commercial Filtration BMPs External $2,394,000 19 $128,300 6 $418,200
Weekly Street/Lot Sweeping External $418,000 74 $5,600 19 $21,900
Commercial Infiltration BMPs + Alum
Treatment Combined $2,429,000 135 $17,900 22 $108,800
Commercial Infiltration BMPs + Curly-leaf
Treatment Combined $2,429,000 150 $16,200 29 $82,700
Commercial Infiltration BMPs + Alum +
Curly-leaf Treatments Combined $2,441,000 180 $13,600 40 $61,500
Spent Lime/CC17 Treatment Cell + Alum
Treatment Combined $66,000 53 $1,200 16 $4,200
Spent Lime/CC17 Treatment Cell + Curly-
leaf Treatment Combined $43,000 66 $600 22 $2,000
Spent Lime/CC17 Treatment Cell + Alum +
Curly-leaf Treatments Combined $78,000 113 $700 36 $2,200
1 Value reported represents the average of the 2015 and 2016 summer average TP concentration reductions (μg/L). Water volume changes were significant in model year 2017, which
negatively impacted in-lake TP concentrations due to existing internal loading. It is recommended that internal loading management occur concurrently with watershed infiltration
practices.
121
Figure 9-1 Annualized cost per unit reduction (µg/L) in summer average total phosphorus
concentration for individual management practices
122
Figure 9-2 Annualized cost per unit reduction (µg/L) in summer average total phosphorus
concentration for combined management practices
123
9.3 Opinions of Probable Cost for Other Evaluated Strategies
9.3.1 Winter aeration of Lake Cornelia using direct oxygen injection
One of the proposed management practices to promote a healthy and more diverse fishery is installation
of a winter aeration system using direct oxygen in North and South Cornelia. The purpose of winter
aeration using direct oxygen injection is to prevent winter kill of predator fish, therefore maintaining a
more balanced fishery. The planning-level opinion of probable cost to install a direct oxygen injection
system in Lake Cornelia is $122,000, with an anticipated range of $86,000 to $183,000 (-30% to +50%)
based on the limited project definition. A detailed opinion of probable cost for the direct oxygen injection
system is included in Appendix H.
9.3.2 Lake Edina Aquatic Plant Management
Invasive curly-leaf pondweed and Eurasian watermilfoil are both present in Lake Edina. Although curly-leaf
pondweed has remained at low levels in the lake since 2008, management of curly-leaf pondweed may be
warranted to maintain its low occurrence and prevent the accumulation of turions (i.e., similar to seeds).
Eurasian watermilfoil was first observed in Lake Edina during 2017, in which it was widespread.
Management of Eurasian watermilfoil in Lake Edina may also be warranted to control its expanding extent
and prevent it from further threatening the integrity of the lake’s aquatic community.
The planning-level opinion of probable cost for herbicide treatment of the curly-leaf pondweed and
Eurasian watermilfoil is approximately $30,000 per year of treatment, with a range of $27,000 to $36,000
(-10% to +20%). This estimate includes preparation of contract documents, permitting, and herbicide
application. The cost estimate also includes potential costs related to monitoring that may be required by
the DNR as part of permitting, including temperature measurements, herbicide residue monitoring, and
aquatic plant monitoring. A detailed opinion of probable cost for the curly-leaf pondweed and Eurasian
watermilfoil is included in Appendix H.
124
10.0 Conclusions and Recommendations
Water quality in Lake Cornelia is poor, with summer average total phosphorus and chlorophyll a
concentrations well above the state standard for shallow lakes. The poor water quality is a result, in part,
of excess phosphorus in the lake, which fuels algal production and decreases water clarity. Fish activity,
specifically the disruption caused by bottom-feeding species such as bullhead and goldfish, may also be
hampering water clarity. Poor water clarity (cloudy water) and the pervasive presence of invasive curly-leaf
pondweed are stressors on the aquatic plant community, which generally fails to meet the MDNR Lake
Plant Eutrophication Index of Biologic Integrity due to the limited number of species and quality of the
plant community.
Water quality in Lake Edina is also poor, with summer average total phosphorus and chlorophyll a
concentrations generally not meeting the state standard for shallow lakes. The aquatic plant community
also does not meet the MDNR Lake Plant Eutrophication Index of Biologic Integrity due to the limited
number of species and quality of the plant community. Invasive curly-leaf pondweed and Eurasian
watermilfoil are both present within the lake. In recent years, curly-leaf pondweed was observed at low
levels in two areas on the west side of the lake. Eurasian watermilfoil is widespread throughout the
shallow lake.
10.1 Phosphorus Sources
Watershed and in-lake modeling was conducted to quantify the sources of phosphorus for Lake Cornelia
and Lake Edina. External phosphorus loading from the watershed is the major contributor of phosphorus
to North Cornelia, ranging from 48% to 76% in modeled years. Internal sediment loading (14% to 40%)
and curly-leaf pondweed die-off/decay (0% to 16%) also contribute significant amounts of phosphorus to
North Cornelia.
The main contribution of phosphorus to South Cornelia comes from North Cornelia, ranging from
54% to 56% of the total phosphorus load in modeled years. The second major contribution of phosphorus
to South Cornelia is from the die-off and decay of curly-leaf pondweed (19% to 23%) and the third is
sediment internal loading (14% to 19%). For South Cornelia, direct watershed phosphorus loading does
contribute phosphorus, but to a much smaller extent than the other sources due to the relatively small
size of the direct watershed (13% of the size of the direct watershed to North Cornelia).
The two main sources of phosphorus loading to Lake Edina are the upstream lakes (North and South
Cornelia) and the direct watershed runoff. Internal phosphorus loading from sediments or curly-leaf
pondweed die-off/decay is minimal in Lake Edina.
10.2 Management Strategies
The watershed and in-lake models were used to predict changes in in-lake phosphorus concentrations in
North Lake Cornelia, South Lake Cornelia, and Lake Edina as a result of various external (watershed) and
internal management strategies. The four external management strategies modeled were (1) infiltration
BMPs on commercial properties, (2) filtration BMPs on commercial properties, (3) a spent lime/CC17
125
treatment chamber downstream of Swimming Pool Pond (North Cornelia watershed), and (4) weekly
watershed-wide street sweeping. The internal management strategies modeled were (1) alum sediment
treatment in Lake Cornelia and (2) curly-leaf pondweed management in Lake Cornelia.
The internal phosphorus load reduction strategies were limited to Lake Cornelia and did not include Lake
Edina. Model calibration and sediment cores retrieved from Lake Edina in 2018 indicate that internal
phosphorus loading from sediments is minimal. Furthermore, while small growths of curly-leaf pondweed
were discovered in Lake Edina in recent years, no curly-leaf pondweed phosphorus loading response was
found during model calibration for model years 2015, 2016, and 2017. Although, internal management
strategies were only applied to North and South Lake Cornelia, Lake Edina water quality is heavily
influenced by Lake Cornelia discharges. Therefore, any management efforts focused on North and South
Cornelia (whether internal or external) will have an impact on the water quality of Lake Edina.
10.2.1 In-lake Phosphorus Management
Model results indicate that management efforts directed at treating the sources of the internal
phosphorus loading will result in significant water quality improvements. Based on the 3 years modeled in
this study (2015, 2016, and 2017), the treatment of the lake’s sediments with alum and the management
of curly-leaf pondweed reduces the summer average total phosphorus concentrations by 23%, 34%, and
20% in North Lake Cornelia, South Lake Cornelia, and Lake Edina respectively (Figure 10-1). These percent
reductions are based on the average reductions modeled for 2015, 2016, and 2017. While large reductions
in the summer average total phosphorus concentrations are achieved with dual internal management of
the sediments and curly-leaf pondweed, these management efforts alone do not improve the lake water
quality enough to meet the MPCA’s water quality standard (<60 µg/L). This indicates that external
management efforts should also be considered within the watersheds.
10.2.2 External (Watershed) Phosphorus Management
Reducing external phosphorus loading is an important part of any lake management strategy. For lakes
like Lake Cornelia that have been exposed to significant external nutrient loadings for extended periods of
time, appreciable sediment and nutrients have accumulated in the lake bottom sediments. As
contributions from the watershed continue, phosphorus will continue to build-up over time in the lake
sediments; increasing the internal loading potential and worsening water quality conditions in the lake.
Several external management practices were evaluated to assess their effectiveness in reducing the
phosphorus concentrations in Lake Cornelia and Lake Edina. While each of the watershed management
practices resulted in predicted improvements in water quality in Lake Cornelia and Lake Edina, the spent
lime/CC17 treatment chamber located upstream of North Cornelia results in the greatest predicted
improvements per unit cost.
Model results indicate that implementation of external phosphorus load reductions results in improved
water quality; however, the predicted incremental improvements are not as significant as those achieved
by the internal load reductions. Figure 10-1 shows the predicted summer average total phosphorus
concentrations for the spent lime/CC17 treatment chamber in combination with alum treatment and
126
curly-leaf pondweed management in comparison with existing conditions and the dual internal
management scenario. While the incremental improvements in predicted lake water quality are not
sufficient to meet the state standard (<60 µg/L), the spent lime/CC17 treatment chamber does
substantially reduce the phosphorus load to the lakes. The phosphorus load from the 410-acre tributary
watershed is reduced by approximately 15% and the total external phosphorus load to North Cornelia is
reduced by approximately 7%.
Figure 10-1 Comparison of summer average total phosphorus concentrations (µg/L) for recommended management alternatives
10.2.3 Responses in Chlorophyll a Concentrations and Water Clarity
The in-lake model developed for this study predicts changes in phosphorus concentration as a result of
various management scenarios. Lake-specific regression relationships were used to estimate the
corresponding changes in chlorophyll a concentrations and Secchi disc transparency based on the
management efforts. Figure 10-2 shows the estimated improvements in summer average chlorophyll a
concentrations for the internal phosphorus management scenario (alum treatment and curly-leaf
pondweed management) and the spent lime/CC17 treatment chamber plus internal management
scenario. Figure 10-3 shows the estimated improvements in summer average water clarity for the internal
phosphorus management scenario (alum treatment and curly-leaf pondweed management) and the spent
lime/CC17 treatment chamber plus internal management scenario.
It should be noted that the response of chlorophyll a and Secchi disc depth to total phosphorus changes
is highly variable. Due to the high variability, the regression equations can be expected only to provide
general indication of the lake response to changing total phosphorus concentrations, and the predicted
chlorophyll a and Secchi disc depth values should not be construed as absolute.
127
Figure 10-2 Summer average chlorophyll a concentrations (µg/L) for recommended management alternatives
Figure 10-3 Summer average Secchi disc depths (m) for recommended management alternatives
128
10.2.4 Recommended Management Practices
The recommended management practices to improve water quality in Lake Cornelia include alum
treatment and curly-leaf pondweed management in Lake Cornelia (North and South) to target internal
phosphorus sources and watershed management practices to target external phosphorus sources.
Construction of a spent lime/CC17 treatment chamber, or equivalent treatment system, is recommended
as a watershed management practice. This BMP would target phosphorus removal, diverting a portion of
the discharge from Swimming Pool Pond through the spent lime filtration chamber to remove both
particulate and dissolved phosphorus before discharging to Lake Cornelia. The spent lime/CC17 system
would be located within Rosland Park on land owned by the City of Edina, which is an important factor
given the high cost and limited availability of land for stormwater management within the Lake Cornelia
watershed.
Management of the benthivorous (bottom-feeding) fish community is also recommended to reduce the
disturbance of bottom sediments and promote a healthy and more diverse fishery. While model results
indicate that alum treatment and curly-leaf pondweed management in Lake Cornelia, in combination with
implementation of watershed management practices, do not result in attainment of the water quality
goals for Lake Cornelia (see Figure 10-1), benthivorous fish management and enhancement of the native
aquatic plant population may result in significant additional gains in water quality improvement.
Water quality in Lake Edina is highly influenced by the water quality of Lake Cornelia. Accordingly, the
recommended management strategy for improving the water quality in Lake Edina is to implement the
management practices recommended for Lake Cornelia. In addition to upstream improvements,
opportunities to reduce the phosphorus loading from the direct watershed to Lake Edina should be
considered. Management of the invasive aquatic plants curly-leaf pondweed and Eurasian watermilfoil,
should also be considered. These management practices are discussed in further detail below.
Table 10-1 summarizes the planning-level opinions of probable cost for each of the recommended
management practices. These opinions of cost are intended to provide assistance in evaluating and
comparing alternatives and should not be assumed as absolute values. All estimated costs are presented
in 2019 dollars and include costs for engineering and project administration.
129
Table 10-1 Planning-level cost estimates for recommended management alternatives
Description Management
Type
Planning-
Level Cost
Estimate1
Planning-Level Cost
Range1
Lake Cornelia Alum Treatment Internal $161,000 $145,000 - $194,000
Lake Cornelia Curly-leaf Pondweed
Management (annual) Internal $12,0002 $11,000 - $15,000
Spent Lime/CC17 Treatment Chamber Watershed $588,000 $412,000 - $882,000
Direct Oxygen Injection System in Lake
Cornelia Internal $122,0003 $86,000 - $183,000
Lake Edina Curly-leaf Pondweed
Management (annual, as needed) Internal $20,000 $18,000 - $24,000
1 Planning-level cost estimates do not include annual costs for operations and maintenance, with exception of the
weekly street sweeping which includes annual operational costs.
2 Cost estimate reflects costs for recent annual curly-leaf pondweed treatments conducted by the City of Edina.
3 Costs for additional monitoring or implementation of management activities to reduce the benthivorous fish
population not included.
10.2.4.1 In-lake Phosphorus Management
Curly-leaf Pondweed Management in Lake Cornelia
The City of Edina has been conducting herbicide treatments in Lake Cornelia in recent years to reduce the
impact of curly-leaf pondweed die-back on water quality and promote a healthy native aquatic plant
population. Continued curly-leaf pondweed management is recommended, which would likely consist of
continued herbicide treatments at a treatment dose such that a lethal dose is attained and sustained for
the period of time sufficient to kill the curly-leaf pondweed. The planning-level cost estimate for annual
curly-leaf pondweed management is $12,000, based on recent City of Edina herbicide treatments (see
Table 10-1).
Alum Treatment of Lake Cornelia Sediments
A whole-lake alum treatment is recommended for both North and South Cornelia. Because Lake Cornelia
is shallow and there is potential for the pH to drop too low if only alum is applied, the aluminum should
be applied as a mixture of alum (4.4% aluminum by weight) and sodium aluminate (10.4% aluminum by
weight), as sodium aluminate acts as a buffer to keep pH at an acceptable range. The planning-level cost
estimate for an alum treatment of Lake Cornelia (North and South) is $161,000 (see Table 10-1).
An alum treatment can be conducted before the other lake management activities are completed.
However, given the extensive benthivorous fish community in Lake Cornelia, the longevity of the
treatment may be reduced as the aluminum floc that covers the lake bottom is mixed deeper into the
sediment over time. If treatment is conducted before other activities, it can be expected that a second
treatment will be needed in the relatively near future (5 to 10 years after treatment). Although a repeat
alum treatment may be necessary every 5 to 10 years, the annualized cost-benefit is still low in
comparison with the evaluated watershed management practices.
130
In recent years, the City of Edina has conducted algaecide treatments in Lake Cornelia and Lake Edina at
the request of residents and lake users, to reduce phytoplankton (algae) blooms. The algaecide
treatments, in the form of copper sulfate, kill phytoplankton, which settle to the lake bottom, along with
the phosphorus in the phytoplankton cells. While this can result in a temporary reduction in phosphorus
in the water column, the dead phytoplankton cells decay over time at the lake bottom and re-release
phosphorus. The decomposition process can also create anoxic conditions, which could promote
sediment phosphorus release. Upon completion of an alum treatment, it is recommended that the
algaecide treatments be suspended to more accurately measure the impacts of the alum treatment on
water quality and phytoplankton populations and assess whether future algaecide treatments are
warranted.
10.2.4.2 Watershed Management Practices
Spent Lime/CC17 Treatment Chamber
A significant portion of the watershed to North Cornelia (47%) flows through the Swimming Pool Pond
located just south of TH 62 and west of Valley View Road near the Edina Aquatic Center. Swimming Pool
Pond is effective in removing sediment and associated particulate phosphorus; however, little to no
dissolved phosphorus is removed as runoff flows through this waterbody into North Cornelia. The
proposed spent lime/CC17 treatment chamber would serve as a “polishing” step, diverting a portion of
the discharge from Swimming Pool Pond through the spent lime filtration chamber to remove dissolved
phosphorus before discharging to Lake Cornelia.
Using spent lime to treat stormwater is a relatively new and innovative approach to removing dissolved
phosphorus that several watershed management organizations throughout the Twin Cities metro area
have been experimenting with in recent years. The primary component of spent lime, a byproduct of the
water treatment process, is calcium carbonate. A spent lime treatment chamber uses chemical substitution
to exchange phosphate for carbonate. As runoff filters through spent lime, calcium will preferentially bind
to phosphate (over carbonate) and calcium phosphate will form. One drawback of spent lime is that the
material has limited capacity for prolonged inundation. Therefore, it is recommended that the bottom
layer of the spent lime treatment chamber be supplemented with CC17, a crushed limestone material, due
to the potential for periodic high water levels in North Cornelia. CC17 is more soluble that most limestone
aggregates and thus, provides calcium that can bind phosphate and create calcium phosphate.
The planning-level cost estimate for installing a spent lime/CC17 treatment chamber upstream of North
Cornelia is $588,000 (see Table 10-1). Although the capital cost of the spent lime/CC17 system is
significant, the unit cost per benefit achieved is relatively low compared to other evaluated watershed
management practices due to the regional nature of the treatment practice and the large amount of
water treated. The cost estimate is based on treating a flowrate of 2 cfs; however, the design could be
modified to reduce the capital cost, as necessary.
131
The spent lime/CC17 treatment chamber scenario aligns well with several of the target criteria that were
used when selecting and evaluating potential watershed management practices, including:
• Maximizes benefits to chain of lakes. The proposed spent lime/CC17 chamber is directly
upstream of North Cornelia, so all three lakes within the chain would benefit from the reduction in
phosphorus loading.
• Increases dissolved phosphorus removal. The spent lime/CC17 chamber would increase the
amount of dissolved phosphorus removed from the watershed runoff.
• Improves or “builds on” effectiveness of existing treatment systems. The proposed spent
lime/CC17 system offers a “treatment train” approach, providing additional phosphorus removal
from the water flowing from Swimming Pool Pond prior to discharge to Lake Cornelia.
• Includes mix of structural and non-structural BMPs. The spent lime/CC17 chamber would be
considered a structural BMP, however, since the proposed implementation location is within a
public park, signs could be posted near the BMP for educational benefit of the park users (non-
structural/programmatic).
• Provides reliable pollutant removal performance. The proposed spent lime/CC17 would
provide consistent, year-round treatment of low-flows from Swimming Pool Pond. However, high
flows would bypass the system and flow directly to Lake Cornelia without treatment. Pilot studies
conducted with spent lime and CC17 filters have demonstrated effective removal of phosphorus.
However, use of spent lime/CC17 to treat stormwater is still relatively new and should be
considered experimental.
• Reasonably cost effective. The spent lime/CC17 chamber uses materials considered “waste”
from previous applications helping to keep the construction costs reasonably cost effective.
• Land Availability: The proposed location for the spent lime/CC17 system is underneath an
existing parking lot in a public park. Installation of the proposed BMP would not require re-
purposing of the land area footprint.
Street Sweeping
The street sweeping scenario analyzed for this study assumed public streets and private commercial
parking lots were swept with high-efficiency vacuum-assisted street cleaners on a weekly basis from May
through November. While this scenario results in significant phosphorus load reduction (assumed 36%
during the modeled time period), it is recognized that this would be an extensive change to the street
sweeping program already in place (twice annually) and may not be feasible for the City of Edina in terms
of capital costs to purchase additional high-efficiency sweepers and annual labor, operations, and
maintenance costs.
Alternatively, an enhanced sweeping program, scaled back in frequency or in the total miles swept, could
be considered. While a scaled-back program would reduce the consistency of phosphorus removal
achieved, it could still be effective in periodically reducing phosphorus loading to Lake Cornelia and Lake
Edina. An enhanced sweeping program could be focused in the residential areas, where runoff currently
receives little or no treatment prior to discharge to Lake Cornelia or Lake Edina. Another option for
132
consideration is a pilot program targeting private parking lots to assess the phosphorus removal
potential. Large portions of the commercial parcels in the North Lake Cornelia watershed discharge to
downstream stormwater ponds. While stormwater ponds are effective at removing particulate
phosphorus, they have limited potential for removing dissolved phosphorus. A pilot program could be
used to assess the effectiveness of an enhanced street sweeping program in private lots to remove
detritus and leaf litter prior to it leaching dissolved nutrients.
Continued Implementation of NMCWD Stormwater Rules
Based on the results of this study, it is recommended that NMCWD continue its regulatory program to
protect local water resources through promotion of infiltration practices from tributary watersheds
through the permitting program. As additional infiltration BMPs are installed as redevelopment occurs, it
will be important to reassess internal loading potential of the downstream lakes to ensure that water
balance changes will not negatively impact water quality due to enhanced internal loading. If internal
loading is managed through alum sediment treatments and curly-leaf pondweed management negative
water quality impacts induced from infiltration practices are not anticipated.
10.2.4.3 Benthivorous (Bottom-feeding) Fish Management
Fish activity, specifically the disruption caused by benthivorous (bottom-feeding) species such as the
bullhead and goldfish found in Lake Cornelia, can influence phosphorus concentrations in a lake. These
fish feed on decaying plant and animal matter found at the sediment surface and transform sediment
phosphorus into phosphorus available for uptake by algae through digestion and excretion. Benthivorous
fish can also cause resuspension of sediments, causing reduced water clarity and poor aquatic plant
growth.
Winter Aeration of Lake Cornelia Using Direct Oxygen Injection
A recommended in-lake management activities to manage the abundant and unchecked population of
benthivorous fish is installation of a winter aeration system using direct oxygen to prevent winter kill of
predator fish, thereby reducing the number of benthivorous fish and maintaining a more balanced fishery.
Winter aeration would be conducted by injecting oxygen under the ice. The system would consist of: (1) a
unit that generates the oxygen and a structure that houses the generator, (2) a raft that holds the vertical
aeration tubes in place, and (3) a bubbler that directs the oxygen upwards through the aeration tubes.
The proposed direct oxygen system would be installed in North and South Cornelia at the deep holes of
the lake contingent upon the availability of power. The system would be sized to be operated only in the
winter when there is ice cover. The planning-level cost estimate for installing a direct-oxygen injection
aeration system is $122,000 (see Table 10-1).
Other Benthivorous Fish Management Options in Lake Cornelia
Other potential management activities to reduce the benthivorous fish population could include a
rotenone treatment of Lake Cornelia and the upstream water bodies and subsequent fish stocking and/or
installation of fish barriers. However, prior to considering these fishery management activities collection of
133
additional information is recommended regarding the migration and movement of carp and goldfish
throughout Lake Cornelia and the series of connected upstream shallow waterbodies.
10.2.4.4 Lake Edina Aquatic Plant Management
Invasive curly-leaf pondweed and Eurasian watermilfoil are both present in Lake Edina. In recent years,
curly-leaf pondweed was observed at low levels in two areas on the west side of the lake. Eurasian
watermilfoil is widespread throughout the shallow lake. The sections below discuss management
recommendations for these invasive plant species.
Curly-leaf Pondweed
The invasive curly-leaf pondweed has been observed at low levels in Lake Edina since 2008. In June of
2017, the species was observed at two locations, both in the western area of the lake. Although curly-leaf
pondweed has remained at low levels in the lake since 2008, management of curly-leaf pondweed may be
warranted to maintain its low occurrence and prevent the accumulation of turions (i.e., similar to seeds).
The goals of treatment would be to prevent curly-leaf pondweed from establishing dominance to avoid
the need for subsequent long-term annual treatments to reduce an established population that can
rebound once larger numbers of turions are present in the sediments. Management of the current curly-
leaf population would also minimize the potential for turions to be conveyed downstream to Normandale
Lake, causing a resurgence of curly-leaf pondweed after completion of the Normandale Lake water quality
improvement project.
The herbicide selected for a curly-leaf pondweed treatment would depend upon the extent; when greater
than 15 percent of the lake is covered, Endothall would be recommended to attain lake-wide control of
the invasive species. When the extent of curly-leaf pondweed is less than 15 percent of the lake, a cost
and benefit analysis would be recommended to determine whether Endothall or diquat would be most
appropriate. Based upon experience with other curly-leaf pondweed management projects, the
management of CLP would be expected to span several years. Management until neither curly-leaf
pondweed nor turions are observed in the lake would be most protective of the Lake Edina ecosystem as
well as downstream Normandale Lake.
Eurasian Watermilfoil
Eurasian watermilfoil was first observed in Lake Edina during 2017, in which it was widespread and
increased in extent between June and August. Continued periodic monitoring of the aquatic plant
community in Lake Edina is recommended to assess the need for future management of Eurasian
watermilfoil. Management of Eurasian watermilfoil in Lake Edina would control its rapidly expanding
extent and prevent it from further threatening the integrity of the lake’s aquatic community. Because
Eurasian watermilfoil fragments could be carried downstream to Normandale Lake, managing Eurasian
watermilfoil in Lake Edina would also protect the integrity of the Normandale Lake aquatic community.
134
11.0 References
Bannerman, R., Owens, D., Dodds, R., & Hornewer, N. (1993). Sources of pollutants in Wisconsin
stormwater. Water Science and Technology, 241-259.
Barr Engineering Co. (2007). Kohlman Lake Dredging Feasibility Study. Prepared for Ramsey-Washington
Metro Watershed District.
Barr Engineering Co. (2015). Memorandum: Lake Edina - Evaluation of Water Levels.
Barr Engineering Co. (2018). City of Edina 2018 Comprehensive Water Resources Management Plan.
Barr Engineering Co., & MPCA. (2018). Lower Minnesota River Watershed TMDLs - Draft.
Cayan, D. (2013). Future climate: projected average. Assessment of climate change in the southwest United
States: a report prepared for the National Climate Assessment., 101-125.
City of Edina. (2017). Bathymetric Survey. Edina, Minnesota, USA.
Cooke, G., Welch, E., Peterson, S., & Newroth, P. (1993). Restoration and Management of Lakes and
Reservoirs, Second Edition. Boca Raton, FL: Lewis Publishers.
Dettinger, M., Udall, M., & Georgakakos, A. (2015). Western water and climate change. Ecological
Applications, 2069-2093.
Dokulil, M. (2013). Old wine in new skins - eutrophication reloaded: Glogal perspectives of potential
amplification by climate warming, altered hydrological cycle and human interference.
Eutrophication: Causes, Economic Implications and Future Challenges, 95-125.
Dokulil, M. (2014). Impact of climate warming on European inland waters. Inland Waters, 4, 27-40.
Dokulil, M. (2016). Climate impacts on ecohydrological process in aquatic systems. Ecohydrology and
Hydrobiology, 16, 66-70.
Dokulil, M., & Teubner, K. (2011). Eutrophication and climate change: Present situation and future
scenarios. Eutrophication: Causes, Consequences, and Control, 1-16.
Emmons & Olivier Resources, Inc. (EOR). (2015). City of Edina, MN Street Sweeping Management Plan. City
of Edina, MN.
Giorgi, F., Im, E., Coppola, E., Diffenbaugh, N., Gao, X., Mariotti, L., & Shi, Y. (2011). Higher hydroclimatic
intensity with global warming. Journal of Climate, 24, 5309-5324.
Heiskary, S., & Wilson, C. (1990). Minnesota Lake Water Quality Assessment Report Second Edition. A
Practical Guide for Lake Managers. MPCA.
135
Hope, D., Naegeli, M., Chan, A., & Grimm, N. (2004). Nutrients on asphalt parking sufaces in an urban
environment. Water, Air, and Soil Pollution, 4, 371-390.
Huisman, J., Matthijs, H., & Visser, P. (2005). Harmful cyanobacteria. Springer.
Huser, B., & Pilgrim, K. (2014). A simple model for predicting aluminum bound phosphorus formation and
internal loading reduction in lakes after aluminum addition to lake sediment. Water Research, 53,
378-385.
I.E.P, Inc. (1990). p8 Urban Catchment Model. Version 3.5. Providence, RI: Prepared for the Narragansett
Bay Project. .
Indiana Department of Natural Resources. (2019, April 13). Aquatic Invasive Species (AIS) Eurasian Water
Milfoil. Retrieved from https://www.in.gov/dnr/files/EURASIAN_WATERMILFOIL.pdf
Jeppesen, E., Kronvang, B., Meerhoff, M., Søndergaard, M., Hansen, K., Andersen, T., . . . Olesen, J. (2009).
Climate change effects on runoff, catchment phosphorus loading and lake ecological state, and
potential adaptations. Journal of Environmental Quality, 38, 1930-1941.
Jeppesen, E., Meerhoff, M., Davidson, T., Trolle, D., Søndergaard, M., Lauridsen, T., . . . Nielsen, A. (2014).
Climate change impacts on lakes: An integrated ecological perspective based on a multi-faceted
approach, with special focus on shallow lakes. Journal of Limnology, 73, 88-111.
Kalinosky, P. (2015). Quantifying solids and nutrient recovered through street sweeping in a suburban
watershed. A Thesis submitted to the faculty of the University of Minnesota.
Kalinosky, P., Baker, L., Hobbie, S., Binter, R., & Buyarski, C. (2014). User Support Manual: Estimating
Nutrient Removal by Enhanced Street Sweeping. Minneapolis, MN.
Kharin, V., Zwiers, F., Zhang, X., & Wehner, M. (2013). Changes in temperature and precipitation extremes
in the CMIP5 ensemble. Climatic Change, 119, 345-357.
LaMarra, V. J. (1975). Digestive activities of carp as a major contributor to the nutrient loading of lakes.
Verh. Int. Verein. Limnol. 19, 2461-2468.
Maxwell (RPBCWD), J. (2018). Lake Cornelia System Fisheries Assessment. Riley Purgatory Bluff Creek
Watershed District.
Maxwell, J. (2018, October 9). Lake Cornelia System Fisheries Assessment.
MDNR. (2017). How the Climate of Minnesota is and is not Changing. (K. Blumenfeld, Performer) State
Climatologist Office, at the City Engineers Association of Minnesota (CEAM) Annual Meeting,
Minnesota.
Metropolitan Council. (2016). Metropolitan Council Land Use Coverage Dataset (GIS database).
136
MPCA. (2008). Minnesota Rules Chapter 7050: Standards for Protection of Water of the State.
MPCA. (2019, April 13). Eurasian Water Milfoil. Retrieved from https://www.pca.state.mn.us/eurasian-
water-milfoil
NMCWD. (2017, amended 2019). Nine Mile Creek Watershed District Water Management Plan.
NOAA, N. O. (2013). Atlas 14. Volume 8.
Passeport, E., & Hunt, W. (2009). Asphalt parking lot runoff nutrient characterization for eight sites in
North Carolina, USA. Jounral of Hydrologic Engineering, 14(4), 352-361.
Pilgrim, K., Huser, B., & Brezonik, P. (2007). A method for comparative evaluation of whole-lake and inflow
alum treatment. Water Research, 41, 1215-1224.
Sahoo, G., Forrest, A., Schladow, S., Reuter, J., Coats, R., & Dettinger, M. (2016). Climate change impacts on
lake thermal dynamics and ecosystem vulnerabilities. Limnology and Oceanography, 61, 496-507.
Schueler, T. (1987). Controlling Urban Runoff: A practical manual for planning and designing urban BMPs.
Washington D.C., USA: Prepared for Washington Metropolitan Water Resources Planning Board.
Metropolitan Washinton Council of Governments.
Selbig, W. (2016). Evaluation of leaf removal as a means to reduce nutrient concentrations and loads in
urban stormwater. Science of the Total Environment, 124-133.
Trenberth, K. (1999). Conceptual framework for changes of extremes of the hydrological cycle with climate
change. Climatic Change, 42, 327-339.
Trenberth, K. (2011). Changes in precipitation with climate change. Climate Research, 47, 123-138.
Trenberth, K., Smith, L., Qian, T., Dai, A., & Fasullo, J. (2003). The changing character of precipitation.
Bulletin of the American Meteorological Society, 84, 1205-1217.
Vighi, M., & Chiaudani, G. (1985). A Simple Method to Estimate Lake Phosphorus Concentrations Resulting
from Natural, Background, Loadings. Water Resources, 19(8), 987-991.
Walsh, J. (2014). Our changing climate. In Climate Change Impacts in the United States: the Third National
Climate Assessment. Washington D.C., USA: U.S. Global Change Research Climate Program.
Wei, Q., Zhu, G., Wu, P., Cui, L., Kaisong, Z., Zhou, J., & Zhang, W. (2010). Distributions of typical
contaminant species in urban short-term storm runoff and their fates during rain events: A case of
Xiamen City. Journal of Environemntal Sciences, 22(4), 533-539.
WI-DNR. (2004). Wisconsin Lake Modeling Suite (WILMS).
137
WI-DNR. (2012). Aquatic Plant Eurasian Watermilfoil. Retrieved from Wisconsin Department of Natural
Resources:
http://dnr.wi.gov/topic/Invasives/documents/classification/Myriophyllum%20spicatum.pdf
WI-DNR. (2012). Eurasian Watermilfoil - Beaver Dam Lake. Retrieved from Wisconsin Department of
Natural Resources: http://dnr.wi.gov/lakes/invasives/AISDetail.aspx?roiseq=1228
Wisconsin Wetlands Association. (2017). Invasive Plant Profile: Cattails. Retrieved from
https://wisconsinwetlands.org/updates/invasive-plant-profile-cattails/
World Health Organization. (2003). Guidelines for Safe Water Environments. In Coastal and Fresh Waters.
Volume 1 (p. 219). Geneva, Switzerland.
Appendices
(in Separate PDF)