coldwater refugia

By Ben Letcher and Jeff Walker

For aquatic species identified as priorities during NER workshops, occupancy models will be developed based on landscape, in-stream and weather variables within a hierarchical framework. The hierarchical framework will incorporate fixed effects and random effects for the variables across space that account for the spatial structure inherent in stream networks. Existing models for coldwater species will be adapted to specific NPS units; they will likely include watershed area, % forested, summer precipitation and mean July stream temperature (e.g., ecosheds.org/models/brook-trout-occupancy/latest). To estimate future probabilities of occupancy, we will use the future climate scenarios described in the previous section. Brook trout and mussel inventory data available to the coIs will help validate these mapped predictions.

summary

Check out Ben and Jeff’s research here.

Current Deliverable Status

We have completed the initial prototype of the Interactive Catchment Explorer (ICE) data visualization tool for exploring cold water refugia among National Park Service (NPS) park units in the northeast U.S (Figure 1). The application is currently available at https://ecosheds.org/dev/ice-nps-ner, and will be moved to a different URL when it is completed.

This tool integrates predictions from our stream temperature and brook trout occupancy models of the northeast U.S. along with the primary input variables to those models (e.g., elevation, land cover, climate). Both models generate predictions based on our high resolution catchment delineation, which includes nearly 400,000 catchments having an average area of about 2 km2. The tool aggregates these catchments to larger watershed units based on the Hydrologic Unit Code (HUC) with resolutions varying from 6-digit HUC to 12-digit HUC. Both models only generate predictions for catchments having cumulative drainage areas less than 200 km2. Therefore, predictions are not generated for larger rivers due to the greater complexity of the temperature dynamics.

The stream temperature model predictions are provided using three metrics. The first two metrics include the mean summer (June-August) stream temperature and the average number of days per year when temperatures exceed 22 °C, both of which were calculated over the historical period 1980-2019. The third metric indicates whether a catchment is considered coldwater refugia based on whether the average number of days per year when daily mean stream temperatures exceed 22 °C is less than one. The three metrics are provided based on both historical conditions (1980-2019) as well as three future climate change scenarios in which the air temperatures were increased by 2, 4 and 6 °C.

The tool also includes the results of our brook trout occupancy model, which generates predictions for the probability of brook trout occupancy in each catchment. Similar to the stream temperature model, the occupancy probabilities are provided under historical conditions and for the same three climate change scenarios (air temperature increases of 2, 4 and 6 °C). The occupancy model predictions also include a series of variables indicating the maximum air temperature increase for which each catchment could achieve 30, 50, and 70% occupancy. These metrics are a measure of how resilient each catchment is to future air temperature increases.

The ICE application includes a layer containing all NPS park units in the region. The user can click on a park unit to select it or they can choose a park by name using the associated drop down menu (Figure 2). When a park is selected, the user can zoom in to it (Figure 2a) and get a summary of all metrics based on the catchments that intersect that park unit (Figure 2b). For each variable, the value associated with the park unit is computed as the area-weighted mean of the associated catchments. It is important to remember that stream temperature and brook trout occupancy model predictions are not made for larger rivers, and therefore, these summary metrics are based only on catchments representing small to medium-sized river segments. The user can also use ICE to view the individual catchments within each HUC in and around the park unit (Figures 2c and 2d). This view can be used to determine which specific areas of the park may or may not provide coldwater refugia.

Next Steps

Our next steps in completing this deliverable include:

1.     Addressing minor bugs such as that identified during the Sept 25 meeting.

2.     Generating a spreadsheet containing the summary metrics for each park. These are the same values that can be viewed within ICE when a park is selected. Having this information in a separate data file may be useful for some users who wish to integrate the stream temperature and/or brook trout occupancy model results into other analyses.

3.     Creating a video tutorial that explains how the ICE application works and how to use it.

4.     Adding hyperlinks to the results of the wildlife and forest refugia analyses for each park pending further discussion with the project team. These links would be added to popup window showing the data summary of a selected park (Figure 2b).

5.     Deploying the final application to a permanent URL on ecosheds.org or usgs.gov (to be determined).

relevant rrc publications

Isaak DJ, Young MK, Nagel DE, et al. 2015. The cold-water climate shield: delineating refugia for preserving salmonid fishes through the 21st century. Global Change Biol 21. https://doi.org/10.1111/gcb.12879

Kurylyk BL, MacQuarrie KT, Linnansaari T, et al. 2015. Preserving, augmenting, and creating cold ‐water thermal refugia in rivers: concepts derived from research on the Miramichi River, New Brunswick (Canada). Ecohydrol 8: 1095-1108. https://doi.org/10.1002/eco.1566