Sensitivity of Summer
Stream Temperatures
to Climate Variability
in Pacific Northwest
Forests
Charles Luce , US Forest Service
Brian Staab, US Forest Service
Marc Kramer, U. Florida
Seth Wenger, U. Georgia
Dan Isaak, US Forest Service
Callie McConnell, US Forest Service
A Rapidly Growing Sensitivity Literature (>2012)…
Kanno et al. (2014). Paired stream–air temperature measurements reveal fine‐scale thermal
heterogeneity within headwater brook trout stream networks. River research and
applications, 30:745-755.
Kelleher et al. (2012) Investigating controls on the thermal sensitivity of Pennsylvania streams.
Hydrol. Process. 26:771–785.
Kurylyk et al. (2015) Shallow groundwater thermal sensitivity to climate change and land cover
disturbances: derivation of analytical expressions and implications for stream temperature
modeling. Hydrol. Earth Syst. Sci 19: 2469-2489.
Leach & Moore (2014) Winter stream temperature in the rain-on-snow zone of the Pacific
northwest: influences of hillslope runoff and transient snow cover. Hydrology and Earth
System Sciences 18: 819-838.
Luce et al. (2014). Sensitivity of summer stream temperatures to climate variability in the
Pacific Northwest. Water Resour. Res, 50, 2.
Mayer (2012). Controls of summer stream temperature in the Pacific Northwest. Journal of
Hydrology, 475:323-335.
Null et al. (2013). Stream temperature sensitivity to climate warming in California’s Sierra
Nevada: impacts to coldwater habitat. Climatic Change, 116:149-170.
Snyder et al. (2015) Accounting for groundwater in stream fish thermal habitat responses to
climate change. Ecological Applications 25:1397–1419.
Trumbo et al. (2014). Ranking site vulnerability to increasing temperatures in southern
Appalachian Brook Trout streams in Virginia: an exposure-sensitivity approach. Transactions
of the American Fisheries Society, 143:173-187.
Sensitivity defined as:
VarS*T
1) stream Δ˚C : air Δ˚C
2) stream Δ˚C : flow ΔSD(Q)
“Connectivity” Theory
Streams that are cold in summer
months are poorly “connected” to the
atmosphere via energy exchanges.
Prediction: cold streams should show
less variation relative to climate forcing
factors (air temperature and discharge).
Sensitivity Parameters from Regressions
Stream Temperature C
16.0
17.0
18.0
19.0
𝑇𝑠 = 𝑏0 + 𝑎𝑇𝑎 + 𝑏𝑄𝑠 + 𝜖
Slope = a = 0.74
𝑇𝑎 : Air temperature
anomaly
𝑄𝑠 : Standardized
streamflow
-2
-1
0
1
2
Air Temperature Anomaly C
Model Timestep & Functional Form
Temp (˚C)
Intra-annual vs. Inter-annual
25
Sun Δ = 0
15
5
Daily, weekly…
2/11/2012
5/21/2012 8/29/2012 12/7/2012 3/17/2013 6/25/2013 10/3/2013
-5
Predicted (°C)
Linear vs. Non-linear
25
20
>20ºC
15
< 0ºC
10
5
5
15
Observed (°C)
25
Mohseni et al. 1998.
Dataset (Region 6 USFS)
Stream Temp Sites
(n = 246; >7
summers of data)
USHCN Air Temp
Red (n = 25)
Streamflow
Blue (n = 15)
Small to medium size streams
Principal Component Reconstructions for:
𝑻𝒂 : Air temperature
anomaly
&
𝑸𝒔 : Standardized
streamflow
𝑋𝑆𝑌 = 𝐿𝑆𝑖 X 𝑃𝑖𝑌
P = Principal Component (time series)
i = which principal component (just using 1 & 2)
L= Loading (map)
S = Station
Y = Year
For a particular place, the time series is given by:
𝑋𝑡 = 𝐿1 X 𝑃1𝑡 + 𝐿2 X 𝑃2𝑡
49
0.3
0.24
48
𝑻𝒂 : Air temperature
anomaly
PC2 Loading
0.17
Air stations
46
0.05
-0.02
45
-0.08
44
-0.15
-0.21
43
-0.27
-0.34
-0.4
42
Latitude
2 PCs
1st – Regional
Mean
2nd – lat & lon
R2=0.81
-124
-122
-120
Longitude
-118
PC2 Air Temperature Anomaly
47
0.11
PCA Reconstruction
Example
0
-2
-4
Obs
Pred
NS R2=0.90
-6
Air Temp
Anomaly
TemperatureAnomaly
2
Halfway, OR
1990
1995
2000
2005
2010
49
𝑸𝒔 : Standardized
streamflow
0.43
48
0.36
Streamflow
Stations
47
0.22
46
0.08
45
0
44
-0.07
-0.14
43
-0.21
-0.28
-0.35
42
2 PCs
1st – Regional Mean
2nd – lat & lon
R2=0.90
Latitude
0.15
-124
-122
-120
Longitude
-118
PC2 Standardized Streamflow Anomaly
0.29
STEAMBOAT CREEK NEAR GLIDE,OREG.
0
1
NS R2=0.95
-1
Streamflow Anomaly
Standardized Departure of Annual Streamflow
2
Obs
Pred
1990
1995
2000
2005
2010
Stream Maximum Weekly Maximum Temperature (MWMT; °C)
Observed values, averaged across years
X-axis
> 25.75
24 - 25.75
22.25 - 24
20.5 - 22.25
18.75 - 20.5
17 - 18.75
15.25 - 17
13.5 - 15.25
11.75 - 13.5
10 - 11.75
< 10
Wide
range
Thermal Sensitivity (°C/°C)
Observed values, averaged across years
Y-axis
>0.9
0.8 - 0.9
0.7 - 0.8
0.6 - 0.7
0.5 - 0.6
0.4 - 0.5
0.3 - 0.4
0.2 - 0.3
0.1 - 0.2
< 0.1
Wide
range
2.0
1.5
1.0
0.5
0.0
mean = 0.5°C/°C
-0.5
Sensitivity to Air
C
air tempC (˚C:˚C)
toTemperature
Sensitivity
Sensitivity to Air Temperature
10
15
20
Stream MWMT (°C)
Average Weekly Maximum Temperature
25
C
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
Sensitivity to Streamflow
C SD Q
to flow (˚C:ΔSD
Sensitivity
Q)
Sensitivity to Annual Streamflow
10
15
20
Stream MWMT (°C)
Average Weekly Maximum Temperature
25
C
Bigger Data, Same Result…
(923 sites, 10-20 year monitoring records)
Sensitivity (˚C/˚C)
1.4
1
0.6
0.2
-0.2 5
10
15
20
August stream temperature (˚C)
25
Isaak et al. In Preparation
Why? What explains low sensitivity of cold streams
Elevation
BFI
Watershed size
Annual Precip
Canopy %
Reach slope
°C/°C
-0.07
0.02
-0.08
0.05
0.07
0.04
2.0
1.5
1.0
0.5
0.0
-0.5
Sensitivity to Air Temperature
C
C
Sensitivity to air temp
What about sensitivity of warmer streams?
10
15
20
Average Weekly Maximum Temperature
25
C
Stream MWMT (°C)
Two Possibilities: Local Climate Forcing &/or Buffering
Air microclimates in complex
valley morphologies
Groundwater influxes
downstream of headwaters
This information needed at
high resolution across
100,000s stream kilometers
Torgersen et al. 2012
Brute Force, Empirical Approach
SSN models to geostatistically krige sensitivity
n = 923 & growing
SSN
Website &
Freeware
• R2 = 0.91
• RMSE = 1.0ºC
Summary
• Streams exhibit a wide range of correlations
with air temperature & discharge (sensitivity)
• Colder streams are less responsive to climate
forcing across years
– Groundwater, late snowpack
– Indirect sensitivity to climate change could be
large (e.g., wildfires alter riparian vegetation &
limit site regrowth)
• Better understanding & prediction of
sensitivities (especially in warmer streams)
would enable better climate change forecasts