AWRA 2015 Summer Specialty Conference
Climate Change Adaptation
Identifying Cisco Refuge Lakes in
Minnesota to Develop a
Landscape Approach for Climate
Change Adaptation
Xing Fang, Ph.D., P.E., D.WRE, F.EWRI, F.ASCE
Author H. Feagin Chair Professor of Civil Engineering
Department of Civil Engineering
Auburn University, Auburn, Alabama 36849-5337
E-mail: [email protected]
Homepage: http://www.eng.auburn.edu/users/xzf0001/
Peter C. Jacobson
Minnesota Department of Natural Resources, Park
Rapids, Minnesota, USA 56470, email:
[email protected]
Prof. Heinz G. Stefan
St. Anthony Falls Laboratory, Department of Civil
Engineering, University of Minnesota, Minneapolis,
Minnesota, USA 55414, email: [email protected]
Donald L. Pereira
Minnesota Department of Natural Resources, 500
Lafayette Road, St. Paul, Minnesota, USA 55155,
email: [email protected]
Outline
• Study history
• Year-round lake water quality model
• Fish habitat model using coupled T & DO
lethal-niche-boundary curve
• Fish habitat model using TDO3 in deep
lakes
• Identify refuge cisco lakes in Minnesota
• Landscape approach for climate change
adaptation
Studying Impacts of Climate Changes
on Fish Habitat in Lakes: Three Phases
• A methodology to estimate global climate change
impacts on lake and stream environmental
conditions and fishery resources in Minnesota.
• Study alterations of water availability, water
quality and fish habitats in small lakes (up to 10
km2) in the contiguous USA by climate change.
• Identify potential coldwater refuge lakes
important for sustaining cisco habitat under
climate warming scenarios in Minnesota.
Regional Water Temperature and Dissolved Oxygen Model for Lakes
Sample
Results
(Thrush
Lake in
MN)
Cisco Lakes in
Minnesota
• Cisco, the most common coldwater stenothermal fish in
Minnesota lakes,
physiologically require cold,
well-oxygenated water to
survive, grow, and reproduce
(Cahn 1927; Frey 1955).
• The Minnesota Department of
Natural Resources (MN DNR)
has sampled cisco from 648
lakes in netting assessments
since 1946 (Minnesota DNR
files). The lakes are scattered
throughout much of the central
and northern MN.
Cisco kill during the summer of 2006
(Jacobson et al. 2008)
Fish Habitat
Model
“lethal-nicheboundary
curve”
Good
Habitat
Hmax =24.4 m
As = 635 ha
Chla = 8.2 µg/L
Lethal
Hmax =12.2 m
As = 436 ha
Chla = 9.4 µg/L
DOlethal = 0.40 + 0.0000060 e 0.59Tlethal
The computed DOlethal is
the required minimum
DO concentration at a
given water temperature
Tlethal for cisco to survive.
Hmax =16.5 m
As = 1937 ha
Chla = 5.1 µg/L
Note: 1 stands for a Julian Day in 2006 and the number of continuous cisco lethal days from the lethal day predicted by the fish habitat model. 2
Julian Day followed by month and date in 2006 inside brackets, 3 the first Yes/No gives the agreement of cisco lethal prediction and reported
cisco mortality in 2006 and Yes/No inside brackets gives the agreement whether or not cisco lethal days from the model include reported the
date with cisco mortality.
Results
“Number of Cisco
Lethal Days”
Number of annual
cisco lethal days
(mean ± standard
deviation)
simulated for the
36 representative
lake types with 9
GR values under
past climate
conditions (19912008) and the
future climate
scenario.
Refuge Lakes for Cisco in Minnesota
• Cisco is a sensitive indicator
of ecological stresses such as
eutrophication and climate
warming.
• We studied 620 cisco lakes in
Minnesota and classified
/grouped them into Tier 1,
Tier 2 refuge lakes and
non-refuge lakes.
• Management strategies was
developed for refuge lakes.
Geographic distributions of 620 Cisco lakes grouped by the shortest distance
between three Class Ι weather stations (International Falls, Duluth and St.
Cloud), three weather stations (stars) and associated grid center points
(crosses) of CGCM 3.1 and MIROC 3.2 used for model simulations. .
Background shades identify ecoregions of Minnesota. Cisco lakes are
essentially in two ecoregions: (1) Northern Lakes and Forests, and (2) North
Central Hardwood Forests.
0
0
Dissolved Oxygen (mg/L)
10
15
20
5
25
0
Dissolved Oxygen (mg/L)
10
15
20
5
30
Temperature
DO
Temperature
DO
5
25
Depth (m)
10
TDO3
15
20
25
TDO3
(B) Future Climate
(MIROC 3.2)
(A) 07/17/1991
0
5
10
30
15
20
25
Temperature ( oC)
0
1984
MIROC 3.2
25
5
10
15
20
25
Temperature ( oC)
30
(C)
31-Day Benchmark Period
Fish Hook Lake
o
TDO3 ( C)
20
15
10
5
0
Jan
Feb
Mar
Apr
May
Oxythermal
Parameter
Habitat TDO3
Determinations
Jun
Jul
Aug
Sep
Oct
Nov
Dec
The TDO3 can be
determined by
interpolating the
temperature of water
at the DO
concentration of 3
mg/L from measured
or simulated vertical
temperature and DO
profiles.
Fixed benchmark
period from July 28
to August 27
Surface
Area
AS (km2)
1.2
Secchi Depth SD (m)
(Maximum lake depth Hmax = 24 m)
2.5
4.5
7.0
8.5
Geometry
Ratio
As0.25/Hmax
0.1
0.5
LakeC01
LakeC06
LakeC02
LakeC07
LakeC03
LakeC08
LakeC04
LakeC09
LakeC05
LakeC10
0.74
1.5
5.0
13.0
LakeC11
LakeC12
LakeC13
LakeC14
LakeC15
LakeC16
LakeC21
LakeC26
LakeC17
LakeC22
LakeC27
LakeC18
LakeC23
LakeC28
LakeC19
LakeC24
LakeC29
LakeC20
LakeC25
LakeC30
50.0
10.0
1.11
1.46
1.97
2.50
3.50
30 virtual cisco lakes
620 cisco lakes
21 cisco study lakes
9.0
Secchi Depth, SD, (m)
8.0
30 ‘virtual’ cisco
lake types were
used to represent
the entire set of
620 lakes.
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
It was not
feasible to run
MINLAKE2010
to simulate water
temperature and
DO profiles for
620 cisco lakes.
0.4
0.6
1.0
2.0
4.0
Geometry Ratio, GR, (m
6.0
)-0.5
10.0
20.0
Future Climate (MIROC 3.2)
International Falls
30 virtual cisco lakes
8
21
15
17
19
11
13
9
23
7
4
25
2
0
Duluth
The lake simulations
were under past
climate from 1961 to
2008 and under two
future climate
scenarios (CCC
CGCM 3.1 and
MIROC 3.2)
8
The AvgATDO3FB
obtained for each of
the 30 lakes was
plotted on a
coordinate system of
GR vs. SD and
contour lines
(isotherms) of
AvgATDO3FB ranging
from 7 to 29 oC were
interpolated.
23
21
19
15
17
25
9
7
4
11
13
23
6
29
25
Secchi Depth (m)
10
2
0
St. Cloud
10
8
27
19
21
27
23
17
9
29
7
4
11
13
15
6
25
Secchi Depth (m)
29 27
6
25
Secchi Depth (m)
10
2
0
0.3
0.5
0.8
1.0
2.0
Geometry Ratio (m -0.5
)
3.0
5.0
7.0
Tier 1 and 2 Refuge Lakes
• The lakes that are most suitable as refuge lakes
(AvgATD3 ≤ 11oC or Tier 1 lakes)
• The lakes are least suitable or non-refuge lakes
(AvgATD3 ＞17 oC or Tier 3 lakes).
• Tier 2 refuge lakes were in the range 11oC ≤ AvgATD3
≤ 17oC.
Secchi Depth (m)
10
Lakes near International Falls
11oC
17oC
8
23 Tier 1 refuge lakes
43 Tier 2 refuge lakes
103 Non-refuge lakes
Tier 2
6
Tier 1
4
Tier 3 or nonrefuge lakes
2
0
10
Lakes near Duluth
39 Tier 1 refuge lakes
50 Tier 2 refuge lakes
100 Non-refuge lakes
8
6
4
2
0
10
Lakes near St. Cloud
22 Tier 1 refuge lakes
34 Tier 2 refuge lakes
206 Non-refuge lakes
8
6
MIROC 3.2
future climate
scenario.
4
2
0
The figure
shows the
distributions of
cisco lakes
assigned to
International
Falls (169
lakes), to
Duluth (189
lakes), and to
St. Cloud (262
lakes) on plots
of SD vs. GR.
0.3
0.5
1.0
3.0
Geometry Ratio (m
-0.5
7.0
)
10.0
Refuge Lakes Identified
Closest
weather station
International
Falls
Duluth
St. Cloud
All three
stations
Climate
scenario
Tier 1 refuge
lakes
Tier 2 refuge
lakes
Total number of
refuge lakes
Non-refuge
lakes
Total number
of lakes
Past
CGCM 3.1
MIROC 3.2
Past
CGCM 3.1
MIROC 3.2
Past
CGCM 3.1
MIROC 3.2
Past
CGCM 3.1
MIROC 3.2
49 (8)
23 (4)
23 (4)
78 (13)
36 (6)
39 (6)
49 (8)
19 (3)
22 (4)
176 (28)
78 (13)
84 (14)
88 (14)
39 (6)
43 (7)
91 (15)
51 (8)
50 (8)
128 (21)
37 (6)
34 (5)
307 (50)
127 (20)
127 (20)
137 (22)
62 (10)
66 (11)
169 (27)
87 (14)
89 (14)
177 (29)
56 (9)
56 (9)
483 (78)
205 (33)
211 (34)
31 (5)
106 (17)
103 (17)
20 (3)
102 (16)
100 (16)
85 (14)
206 (33)
206 (33)
137 (22)
415 (67)
409 (66)
169 (27.2)
169 (27.2)
169 (27.2)
189 (30.5)
189 (30.5)
189 (30.5)
262 (42.3)
262 (42.3)
262 (42.3)
620 (100)
620 (100)
620 (100)
Note: for the fixed benchmark method
Geographic
distribution of Tier 1
and Tier 2 cisco refuge
lakes and Tier 3 nonrefuge cisco lakes
obtained from
simulations for the
future climate scenario
MIROC 3.2. The
boundary limits for
Tier 1 and Tier 2 refuge
lakes were contour
lines of AvgATDO3FB =
11 oC and 17 oC,
respectively. The fixed
benchmark method
and weather data from
principal weather
stations in
International Falls,
Duluth, and St. Cloud,
Minnesota, were used.
These selective lakes have a Secchi depth greater than 2.3
m (mesotrophic and oligotrophic lakes) and are seasonally
stratified (geometry ratio less than 2.7 m-0.5).
Location of
identified
cisco refuge
lakes
greater
than 40.5
ha in
Minnesota
against a
background
of land uses
Refuge Lakes using Landscape Approach
• 171 refuge lakes with having recent robust
cisco population and surface area greater than
40.5 ha
• It involve 1,021 catchments (~1.7 million ha)
• 565 catchments have protection exceeding
75%
• Many of sufficiently protected catchments are
in the north part of Minnesota (the Superior
National Forest)
Priority scores for
catchments of
refuge lakes
identified statewide
(4a) and for the
catchment of
Whitefish Lake in
Crow Wing County,
Minnesota (4b)
Summary and Conclusions
• To project its chances of survival under future warmer climate
conditions, using simulated daily T and DO profiles in 30 virtual lake
types, an oxythermal habitat variable, TDO3, i.e. water temperature at
DO = 3 mg/L, was calculated in each simulation day.
• About 208 (one third) and 160 (one fourth) of the 620 cisco lakes are
projected to maintain viable cisco habitat under the two projected
future climate scenarios using the fixed and variable benchmark
periods, respectively.
• These selective lakes have a Secchi depth greater than 2.3 m
(mesotrophic and oligotrophic lakes) and are seasonally stratified
(geometry ratio less than 2.7 m-0.5).
• A landscape approach was developed to identify important catchments
of refuge lakes. These catchments were prioritized based on two
components: (1) threat (changes in land use) and (2) investment
efficiency. Conservation strategies were implemented for some of the
prioritized catchments critical for water quality protection.
Thanks!
Questions?