Using Spatial Datasets for Conservation Planning Under Climate Change in the Pacific Northwest

Chapter 9: Stream Temperature & Climate Velocity Predictions in the PNW

Dataset Overview

These datasets contain predictions of stream temperature climate velocity and stream-warming rates throughout 222,000 km of streams and rivers in the northwestern U.S (see chapter glossary for definitions of terms). Specifically, these are predictions of the rates at which mean August water temperature isotherms (units of the same temperature) shift through streams under six climate-change scenarios (figure 9.1). The predictions are based on modeled relationships between environmental covariates and water temperature patterns that vary among streams. Climate-change scenarios include a continuation of the historical rate of warming, and multiples of that historical rate (2 and 3 times), combined with stream sensitivity that either mirrors historical stream sensitivity (with cold streams less sensitive to air warming than warm streams) or else stream sensitivity that is just as high for cold streams as for warm streams. The velocity scenarios were derived from the NorWeST stream temperature model scenarios (Isaak et al., 2017), which are available to describe historical and future conditions for the months of June, July, August, and September (Isaak et al., 2016a). The stream temperature dataset encompassing 23,000 monitoring sites used to develop the NorWeST scenarios is also available as summaries of daily mean, maximum, and minimum temperatures (Chandler et al., 2016).

Data Access:

https://www.fs.fed.us/rm/boise/AWAE/projects/NorWeST/ModeledStreamTemperatureScenarioMaps.shtml

Figure 9.1

Conservation Applications

Potential conservation applications of this dataset could include the following:

These datasets are broadly applicable to conservation of many types of species inhabiting streams (e.g., fish, stream-dwelling amphibians, and invertebrates) but may be particularly important for cold-water fish populations. Climate model outputs that may be appropriate for evaluating terrestrial ecosystems (e.g., climate velocities or projected future air temperatures) are often inadequate to assess the degree of climate-change exposure that aquatic organisms will experience. These projections must be translated into water temperature changes, which these datasets accomplish.
Various management options may be available to help reduce stream temperatures—or slow the rate of warming—depending on the stream in question. These options may include ensuring minimum flows in summer through the regulation of water withdrawals, reconnecting streams to floodplains through stream restoration, and increasing stream shading by establishing riparian vegetation. These datasets provide information that could help inform prioritization of streams for these various approaches.

Applicable scales for detailed spatial assessments:

For conservation applications requiring detailed assessment of spatial variation of the dataset within a geographic boundary (such as a protected area), the following geographic scales may be most appropriate (see appendix 3 for more information).
The high density of stream temperature monitoring stations enabled scenarios and climate velocities to be estimated at 1-km resolution. Possible appropriate spatial scales include small headwater streams and large rivers at the scale of: a Bureau of Land Management (BLM) district, a river watershed (8-digit hydrologic unit code [HUC-8]), an individual county, a national forest, a level-3 ecoregion (e.g. the North Cascades), a single state (Washington, Oregon, or Idaho), a region (multiple states in the Pacific Northwest or in western North America).

Applicable scales for assessing general patterns:

Due to spatial resolution, the dataset may not show detailed spatial variation at the following geographic scales, however, the dataset may be useful to assess general patterns or for comparison to other locations (see appendix 3 for more information).
At spatial scales < 1 km within streams and rivers.

Use of the dataset in conservation applications may be limited by the following considerations:

The variables used in the NorWeST model generally do not represent human changes to stream networks and the surrounding landscape that could affect stream temperatures, such as channel realignments, stream diversions, and changes to riparian canopy cover. The 1-km resolution of the NorWeST model also means that finer-scale variations in stream temperature cannot be assessed, including the possible locations of cold microrefugia (1 – 10 m) that could be important for some fish during warm summer periods. In addition, climate change could have complex effects on stream temperatures that were not accounted for in the NorWeST model, such as changing vegetation dynamics along riparian corridors with implications for canopy shading. Changing disturbance patterns, such as from floods and fires that are altered due to climate change, may also have important effects on aquatic communities that are not accounted for by these datasets. Finally, it should be stressed that different aquatic species have different sensitivities to changing stream temperatures, such that these datasets need to be combined with species-specific distribution information in cases where management goals are targeted at individual species or groups of species.

Past or current conservation applications:

Chapter 9 Case Study

Dataset citations:

Chandler, G. L., S. Wollrab, D. Horan, D. Nagel, S. Parkes, D. J. Isaak, S. J. Wenger, and et al. 2016. NorWeST stream temperature data summaries for the western U.S. USDA Forest Service, Rocky Mountain Research Station Research Data Archive, Fort Collins, CO.

Isaak, D. J., S. J. Wenger, E. E. Peterson, J. M. Ver Hoef, S. W. Hostetler, C. H. Luce, J. B. Dunham, J. L. Kershner, B. B. Roper, D. E. Nagel, G. L. Chandler, S. P. Wollrab, S. L. Parkes, and D. L. Horan. 2016a. NorWeST modeled summer stream temperature scenarios for the western U.S., Fort Collins, CO: USDA Forest Service Research Data Archive.

Isaak, D. J., M. K. Young, C. H. Luce, S. W. Hostetler, S. J. Wenger, E. E. Peterson, J. M. Ver, M. C. Groce, D. L. Horan, and D. E. Nagel. 2016b. Slow climate velocities of mountain streams portend their role as refugia for cold-water biodiversity. PNAS 113:4374–4379.

Isaak, D., S. Wenger, E. Peterson, J. Ver Hoef, D. Nagel, C. Luce, and et al. 2017. The NorWeST summer stream temperature model and scenarios for the western US: A crowd‐sourced database and new geospatial tools foster a user community and predict broad climate warming of rivers and streams. Water Resources Research 53:9181–9205.

Dataset documentation links:

https://doi.org/10.2737/RDS-2016-0032 (open access)

https://doi.org/10.2737/RDS-2016-0033 (open access)

https://doi.org/10.1073/pnas.1522429113 (open access)

https://doi.org/10.1002/2017WR020969 (subscription or fee required)

Data access:

Stream temperature projections for various climate-change scenarios can be downloaded from: https://www.fs.fed.us/rm/boise/AWAE/projects/NorWeST/ModeledStreamTemperatureScenarioMaps.shtml

Climate-velocity projections may be obtained by contacting the corresponding author.

The dataset can be viewed online at: https://usfs.maps.arcgis.com/apps/webappviewer/index.html?id=bf3ff38068964700a1f278eb9a940dce

Metadata access:

Formal metadata can be downloaded from: https://www.fs.fed.us/rm/boise/AWAE/projects/NorWeST/downloads/ModeledStreamTemperatureMaps/metadata/NorWeST_PredictedStreamTempLines_GeneralMetadata.pdf

Additional metadata is available from: https://doi.org/10.2737/RDS-2016-0033 and https://doi.org/10.2737/RDS-2016-0032

Dataset corresponding author:

Daniel Isaak

USDA Forest Service, Rocky Mountain Research Station

[email protected]

Data type category (as defined in the Introduction to this guidebook): Climate, hydrology, stream and riparian, animal habitat

Species or ecosystems represented: This dataset represents stream and river ecosystems and the aquatic species that depend on them, including fish and aquatic invertebrates.

Units of mapped values:

Predicted stream temperatures: °C

Climate velocity: kilometers per decade (km / decade)

Range of mapped values: For the climate velocity dataset, ranges of mapped values depend on the future time frame and warming scenarios considered (see sections 6 and 7). For example, velocity values in a scenario using the historical warming rate with historical stream sensitivity ranged from 0 to greater than 16 kilometers per decade. For the stream temperature dataset, values range from 0 °C to 30 °C.

Spatial data type: vector and point data (points, lines, or polygons)

Data file format(s): Shapefile (.shp)

Geographic coordinate system: North American Datum of 1983

Projected coordinate system: North American Datum of 1983 Albers

Spatial extent: Regional

Dataset truncation: The datasets are truncated along the border between the United States and Canada.

Time period represented: Historical (1993 – 2015) and future (later than 2020)

Future time period(s) represented: Mid-century (2030-2059), end-of-century (2070-2099)

Baseline time period (against which future conditions were compared): 1993-2011.

Methods overview:

Mean August stream temperatures were modeled based on observations from the NorWeST database of water temperatures (Isaak et al., 2017). The NorWeST database consists of more than 220 million temperature observations from more than 23,000 stream sites in the western United States (Chandler et al. 2016). Several attributes of watersheds were used to model stream temperatures, including elevation, stream reach slope, drainage area, percent of watersheds containing lakes, glaciers, precipitation, streamflow, and air temperature; see table 1 in Isaak et al. (2017) for detailed information. Once models were constructed, they were used to make future predictions based on projected changes in air temperatures and streamflow under various climate-change scenarios. These scenarios are described in Isaak et al. (2017) and can be viewed at: https://www.fs.fed.us/rm/boise/AWAE/projects/NorWeST/downloads/NorWeST_HistoricalStreamTempScenarioDescriptions.pdf.

In general, scenarios are available for historical periods and future periods, 2030-2059 and for 2070-2099, using either a constant delta temperature increase for all streams within a region (NorWeST processing unit) or scenarios with differential warming and the smaller increases that occur in cold streams relative to warm streams. Scenarios are also presented using simple integer increments of warming applied to all streams (1°C, 2°C, and 3°C). Climate velocity was calculated using multiples (equal to, two times, and three times) of the historical warming rate (Isaak et al., 2016b). For more information, please consult the dataset citations listed in section 2 of this chapter.

This dataset relied on the following general types of models: Statistical models

This dataset employed the following specific models: The NorWeST stream temperature model (Isaak et al. 2017), which is a geostatistical moving average approach specific to stream networks (Ver Hoef et al. 2006; Ver Hoef and Peterson 2010).

Major input data sources for this dataset included: Historical climate observations or models, future climate projections, streamflow data or other hydrologic data, stream temperature observations, digital elevation models (DEMs), current land use

This dataset used the following general circulation models (GCMs): HadCM, CNRM-CM, CCSM3, ECHAM5, ECHO-G, CGCM-3.1_T47, PCM1, MIROC-3.2, IPSL-CM4, HadGEM1

An ensemble across GCMs is provided; individual files for individual GCMs are not available. More information about climate models is available in appendix 1. Detailed information about climate models, including model evaluation and comparison among models, is available from Randall et al. (2007), Hamlet et al. (2013), and Rupp et al. (2013).

This dataset used the following greenhouse-gas scenarios: SRES A1B

More information about greenhouse-gas scenarios is available in appendix 2 and from Knutti and Sedláček (2013).

Creation of this dataset involved the following methods to change the spatial resolution of climate models (e.g. to downscale or resample climate models): The hybrid delta method was used to downscale air temperature and streamflow information (Hamlet et al. 2013).

The mapped values of the dataset may be interpreted as follows:

Historical and future predictions of summer stream temperatures are in °C; larger values are warmer and smaller values are cooler. For stream climate velocity, predictions are in km/decade; larger values indicate faster stream isotherm shift rates, and smaller values represent more gradual shift rates.

Representations of key concepts in climate-change ecology:

These datasets represent a framework with which to examine climate-change vulnerability for streams and the aquatic species that depend on them. Streams that are projected to warm faster, and those that show greater sensitivity (greater stream temperature response to increasing air temperatures) may be more vulnerable to climate change than those that warm more slowly and are less sensitive to air temperature increases.

This dataset illustrates a larger concept in climate-change ecology, namely the importance of translating changes in regional climate conditions into changes in the environmental variables that directly regulate species’ habitat suitability (in this case, in-stream water temperatures). Warmer stream temperatures represent a potential threat to fish species requiring cold water for suitable habitat, including species that are important for subsistence, recreation, and commercial fisheries. Information on exposure to changing stream temperatures can be combined with information on species’ sensitivities to those changes, such as thermal tolerance limits, as well as species’ abilities to adapt to changing stream temperatures, such as through use of small cold-water refugia produced by springs.

This dataset involves the following assumptions, simplifications, and caveats:

In the study that produced the stream climate velocity dataset, stream temperature sensitivity to air temperature increases was considered only based on the current temperature conditions of the stream. The dataset authors considered other variables such as elevation, canopy cover, and stream size but were not able to predict stream temperature sensitivity using these variables. Differences in stream sensitivity could be attributable to processes such as shading from forest canopy, contribution of groundwater flow, or snowmelt, but these processes were not accounted for in the modeling.

Although the models that produced these datasets did account for changes in streamflow, there is considerable uncertainty concerning streamflow responses to climate change and streamflow data were available from fewer sites than stream temperature data. These issues are particularly acute for headwater streams.

Quantification of uncertainty:

Predicted August stream temperatures from the NorWeST model were compared to observed August stream temperatures for approximately 63,000 sites. The r2 value for this comparison was 0.91 (an r2 value of zero would indicate no ability to predict stream temperatures; an r2 of 1.0 would indicate perfect predictive ability). Prediction precision also varies spatially throughout the study domain networks and is dependent on the density of local temperature observations used to fit the model. One of the scenarios (S22) available at the NorWeST website contains the prediction standard errors at 1-km points along stream networks and can be used to map spatial uncertainty. Other metrics to assess the ability of the NorWeST model to predict August stream temperatures are described and presented in Isaak et al. (2017).

Field verification:

Cross validation was performed during model fits to compare predictions to field observations. The stream temperature scenarios have also been used with biological datasets for more than a dozen fish and amphibian species and thermal relationships match expectations based on the ecology of these species. Field verification of the future values predicted in this dataset was not possible because the dataset represents a future condition.

Prior to dataset publication, peer review was conducted by external review (at least two anonymous reviewers, each from a different institution).

Hamlet, A. F., M. M. Elsner, G. S. Mauger, S.-Y. Lee, I. Tohver, and R. A. Norheim. 2013. An overview of the Columbia Basin climate change scenarios project: approach, methods, and summary of key results. Atmosphere-Ocean 51:392–415.

Isaak, D. J., S. J. Wenger, E. E. Peterson, J. M. Ver Hoef, D. E. Nagel, C. H. Luce, and et al. 2017. The NorWeST summer stream temperature model and scenarios for the western US: A crowd‐sourced database and new geospatial tools foster a user community and predict broad climate warming of rivers and streams. Water Resources Research 53:9181–9205.

Isaak, D. J., M. K., Young, D. E., Nagel, D. L., Horan, and M. C. Groce. 2015. The cold-water climate shield: Delineating refugia for preserving salmonid fishes through the 21st century. Global Change Biology 21:2540–2553.

Knutti, R., and J. Sedláček. 2013. Robustness and uncertainties in the new CMIP5 climate model projections. Nature Climate Change 3:369–373.

Randall, D., R. Wood, S. Bony, R. Colman, T. Fichefet, J. Fyfe, V. Kattsov, and et al. 2007. Cilmate models and their evaluation. in S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. Averyt, M. Tignor, and H. Miller, editors. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, New York, NY.

Rupp, D. E., J. T. Abatzoglou, K. C. Hegewisch, and P. W. Mote. 2013. Evaluation of CMIP5 20th century climate simulations for the Pacific Northwest USA. Journal of Geophysical Research 118: 884–907.

Ver Hoef, J. M., and E. E. Peterson. 2010. A moving average approach for spatial statistical models of stream networks. Journal of the American Statistical Association 105: 6–18.

Ver Hoef, J. M., E. E. Peterson, and D.M. Theobald. 2006. Spatial statistical models that use flow and stream distance. Environmental and Ecological Statistics 13: 449–464.