Date: January 22, 2020Presented to: MN Climate Change Adaptation ConferencePresenter: Lucinda Johnson

Or: What can long-term data tell us that we wouldn’t otherwise know?

Benefits of long-term data for adaptation planning

Climate Stress

Con

ditio

n

Stress

Con

ditio

n

Quantile Regression

Many Possible Response Patterns to Climate Stress(& Potential Analytical Approaches)

Climate StressC

ondi

tion

T

B

Climate Stress

Con

ditio

n

T

B

Stress

Con

ditio

n

Piecewise QuantileRegression

T

B

StressC

ondi

tion

Piecewise Quantile Regression

T T

B

B = Biological Condition BreakpointT = Environmental Threshold

From: J. Ciborowski

Logistic RegressionSimple Linear Regression Piecewise Linear Regression

Lessons Learned

• 2007: Millions of data records (over 4000 lakes in the database), but overall, little overlap among data sets, e.g., only 27 lakes with both long term fish and water chemistry data. Heinz Stefan (UMN)

• 2013: “Long-term, consistent data sets are sorely needed to detect robust trends.” Lucinda Johnson (NRRI) Talk 3/27/13 to DNR Fisheries Managers

Linear & Nonlinear Trends… historic and future

- Climate - Climate Change Responses: examples

Ø Ice OutØ Water TemperatureØ Water ChemistryØ Fish Species traitsØ Fish Species Distributions

4NRRI • Innovative Research • Minnesota Value • Global Relevance • www.nrri.umn.edu

Northern MN is warming more and faster

than southern MN

Summer Temperatures: Lake Superior

From: Austin and Colman, 2006.

Water temperatures are rising faster than air

temperatures due to reduced ice cover.

Trend = 1.4 days / decadeSince 1950’s (n = 71)

Staples, et al. in prep.

Ice Out 1948 -2008

http://minnesota.publicradio.org/display/web/2013/01/25/regional/minnesota-cold-winter-photos#3

Ice out dates are earlier and

the rate of ice loss is

accelerating.

Historic Climate Regime Contrasts(summer mean water quality)

Variable n “Cold” “Hot” pEC25 (μs/cm)

39 310.8 307.9 n.s.

Thermocline depth (m)

66 5.6 m 4.8 m <0.001

Secchi depth (m)

524 2.63 2.57 0.005

Trophic State Index

537 50.34 51.3 <0.001

Chlorophyll (ug/L)

32 29.6 28.4 n.s.

Phosphorus (mg/L)

39 0.10 0.16 n.s.

(“Hot” = > 2 S.D. mean annual temp; “Cold” = < 2 S.D. mean annual temp)

Blumenfeld et al., unpublished data.

Some water quality variables are more

responsive than others to changing temperature

regimes.

Range ShiftFish Species Occurrence

1940 – 2005

K. Schneider, 2010

Some fish are more robust & adaptive than others

to changing temperature regimes.

Models

ØPredictive Models:Ø Lake temperatureØ Stream temperatureØ Brook Trout presence / absenceØ Cold water fish habitat

ØCan be used for hind casting as well as forecasting

ØCo-located data are powerful

101/28/20

Temperature: 427 sites 1996 - 2009

Brook Trout: 371 sites 1997 – 1999, 2008 - 2011

Co-located fish + temperature

n = 79within 1km distance and no tributary between sites.

From: Johnson, et al. 2013

Lessons Learned

• 2007: Millions of data records (over 4000 lakes in the database), but overall, little overlap among data sets, e.g., only 27 lakes with both long term fish and water chemistry data. Heinz Stefan (UMN)

• 2013: “Long-term, consistent data sets are sorely needed to detect robust trends.” Lucinda Johnson (NRRI)

• 2019: “Over 450 Minnesota lakes now have fish and water quality data; many of these also have plant surveys as well.” Jacquelyn Bacigalupi, (DNR)

Talk 3/27/13 to DNR Fisheries Managers

Despite Minnesota Being a Data-Rich State, Data Gaps in Aquatic Response

Studies Still Persist:

1. Groundwater maps throughout state2. Full coverage of detailed soil maps3. Detailed surficial geology maps4. Depth to bedrock5. Bedrock fracture patterns6. Hydrologic models of groundwater flow

131/28/20

Thank YouSpeaker Name(s)TitleContact Information

14

NRRI Duluth(218) 788-26945013 Miller Trunk Hwy, Duluth, MN 55811

NRRI Coleraine(218) 667-4201One Gayley Avenue, Coleraine, MN 55722

nrriinfo@d.umn.edu // www.nrri.umn.edu