Ecoregional projections
Climate model projections suggest major North American biome shifts in response to anthropogenic climate change (Rehfeldt et al. 2012). Such shifts could have profound influences on native flora and fauna, many of which would have to move long distances to track their climatic niches. To evaluate potential ecosystem changes at a somewhat finer scale, the change in climate space was projected for level III ecoregions (Commission for Environmental Cooperation 1997) as surrogates for multiple associated species and ecological communities. First, a random forest model (Breiman 2001) was developed to predict ecoregion class from bioclimatic variables (see Table 1 in pdf), using 1-km interpolated climate data for the 1969-1990 normal period (Hamann et al. 2013), available at http://adaptwest.databasin.org/pages/adaptwest-climatena.

This model was then used to project ecoregions onto future mid-century (2041-2070) and end-of-century (2071-2100) climate conditions. Climate projections were based on 1-km downscaled climate anomalies (Wang et al. 2016) generated by an ensemble of 15 widely-used GCMs from the Coupled Model Intercomparison Project, Phase 5 (CMIP5, Taylor et al. 2012), available at http://adaptwest.databasin.org. Representative concentration pathway (RCP) 8.5 was used to represent the 21st century conditions that are to be expected without dramatic reductions in greenhouse gas emissions or technological fixes (Fuss et al. 2014). RCP 4.5 was also evaluated to represent a future in which significant emissions reductions are achieved. The change in area was also calculated for each Level III ecoregion (see Table 2 in pdf).

Results for RCP 4.5 and RCP 8.5 are shown below in Figures 1 and 2, respectively.

Figure 1. Model-predicted (a) baseline, (b) mid-century, and (c) end-of-century changes in North American ecoregions for RCP 4.5. Boreal, hemi-boreal, and western forested regions are shown in green and blue-green shades; arctic ecoregions are in blue shades; prairie/parkland ecoregions are in brown shades; and temperate forest ecoregions are in yellow and orange shades (see Table 1 for full list of ecoregions). Boreal ecoregions are also outlined in black.

Figure 2. Model-predicted (a) baseline, (b) mid-century, and (c) end-of-century changes in North American ecoregions for RCP 8.5. Boreal, hemi-boreal, and western forested regions are shown in green and blue-green shades; arctic ecoregions are in blue shades; prairie/parkland ecoregions are in brown shades; and temperate forest ecoregions are in yellow and orange shades (see Table 1 for full list of ecoregions). Boreal ecoregions are also outlined in black.

Ecoregional refugia
Climate-change refugia, or areas of species persistence under climate change, may vary in proximity to a species’ current distribution, with major implications for their conservation value. Thus, the concept of climate velocity—the speed at which an organisms must migrate to keep pace with climate change—is useful to compare and evaluate refugia. Using analog climate methods, both forward and backward velocity can be calculated, providing complementary information about spatio-temporal responses to climate change (Hamann et al. 2015). In particular, backward velocity calculations can be used to identify areas of high potential refugium value for a given time period and species or ecoregion (Stralberg et al. 2018). Refugia for a given ecoregion represent areas where the climates of that ecoregion may persist into the future.

Stralberg (2019) used random forest model projections of future ecoregions described above (Stralberg 2018)  to generate an index of climate-change refugia potential for individual ecoregions, using the methods outlined in Stralberg et al. (2018). The index ranges from 0 to 1, with values close to 1 indicating overlap or very close proximity to the current mapped ecoregion, across multiple climate models. Because the random forest algorithm is a classifier that assigns an ecoregion class to every future pixel, it does not account for novel climates that are not currently found in any North American ecoregion. Of course novelty is relative and can be measured in many different ways. A multivariate environmental similarity surface (MESS) was calculated following Elith et al. (2010) to generate an index of novelty for each future ecoregion (negative values indicate dissimilarity). For mapping purposes, novel climates for each ecoregion, RCP, and time period were identified as those with values lower than the 1st percentile of dissimiarity values for the baseline periods. 

Figure 3. Maximum ecoregional refugia values for (a) RCP 4.5, 2050s, (b) RCP 4.5, 2080s, (c) RCP 8.5, 2050s, and (d) RCP 8.5, 2080s. CEC Level III ecoregion boundaries are overlaid.

Data can be downloaded or viewed below, and are also found on Zenodo.org as linked below. A document with links to maps in pdf format is also available here.

Cite the ecoregional projection data as: Stralberg, D. 2018. Climate-projected distributional shifts and refugia for North American ecoregions [Data set]. http://doi.org/10.5281/zenodo.1407176. Available at https://adaptwest.databasin.org.

Cite the ecoregional refugia data as: Stralberg, D. 2019. Velocity-based macrorefugia for North American ecoregions [Data set]. Zenodo. http://doi.org/10.5281/zenodo.2579337. Available at https://adaptwest.databasin.org.

Also see: Stralberg, D., A. Camfield, M. Carlson, C. Lauzon, N. K. S. Barker, A. Westwood, and F. K. A. Schmiegelow. 2019. Strategies for identifying priority areas for passerine conservation in Canada’s boreal forest. Avian Conservation and Ecology. https://doi.org/10.5751/ACE-01363-140113

Ecoregional refugia:
Filename format:
_Zeco_curr.tif [Current mapped ecoregion]
_Zpred_curr.tif [Current predicted ecoregion]
ZZrefugia_XXXXX_YYYY.tif [Ecoregional refugia index]
ZZnovel_XXXX_YYYY [Dissimilarity index]
MaxRefugia_XXXX_YYYY [Maximum pixel-level refugia value]

where:
Z = Ecoregion ID
XXXXX = Representative Concentration Pathway (rcp45 or rcp85)
YYYYY = Time period (2050s or 2080s)

Supporting documents
--------------------
ecoregion_lookup.csv [Ecoregion definitions and projected change summaries]
EcoregionLabels.pdf [Ecoregion index map]

Projection information
----------------------
"+proj=lcc +lat_1=49 +lat_2=77 +lat_0=0 +lon_0=-95 +x_0=0 +y_0=0 +ellps=GRS80 +units=m +no_defs"
----------------------
Projection LAMBERT
Spheroid GRS80
Units METERS
Zunits NO
Xshift 0.0
Yshift 0.0
Parameters
49 0 0.0 /* 1st standard parallel
77 0 0.0 /* 2nd standard parallel
-95 0 0.0 /* central meridian
0 0 0.0 /* latitude of projection's origin
0.0 /* false easting (meters)
0.0 /* false northing (meters)

REFERENCES

Breiman, L. 2001. Random Forests. Machine Learning 45:5-32.

Commission for Environmental Cooperation. 1997. Ecological Regions of North America: Toward a Common Perspective, Montreal, Canada.

Elith, J., M. Kearney, and S. Phillips. 2010. The art of modelling range-shifting species. Methods in Ecology and Evolution 1:330-342.

Fuss, S., J. G. Canadell, G. P. Peters, M. Tavoni, R. M. Andrew, P. Ciais, R. B. Jackson, C. D. Jones, F. Kraxner, N. Nakicenovic, C. Le Quere, M. R. Raupach, A. Sharifi, P. Smith, and Y. Yamagata. 2014. Betting on negative emissions. Nature Climate Change 4:850-853.

Hamann, A., T. Wang, D. L. Spittlehouse, and T. Q. Murdock. 2013. A comprehensive, high-resolution database of historical and projected climate surfaces for western North America. Bulletin of the American Meteorological Society 94:1307-1309.

Hamann, A., D. Roberts, Q. Barber, C. Carroll, and S. Nielsen. 2015. Velocity of climate change algorithms for guiding conservation and management. Global Change Biology 21:997-1004.

Rehfeldt, G. E., N. L. Crookston, C. S enz-Romero, and E. M. Campbell. 2012. North American vegetation model for land-use planning in a changing climate: a solution to large classification problems. Ecological Applications 22:119-141.

Stralberg, D., C. Carroll, J. H. Pedlar, C. B. Wilsey, D. W. McKenney, and S. E. Nielsen. 2018. Macrorefugia for North American trees and songbirds: Climatic limiting factors and multi-scale topographic influences. Global Ecology and Biogeography 27:690-703. https://doi.org/10.1111/geb.12731

Taylor, K. E., R. J. Stouffer, and G. A. Meehl. 2012. An Overview of CMIP5 and the Experiment Design. Bulletin of the American Meteorological Society 93:485-498.

Wang, T., A. Hamann, D. Spittlehouse, and C. Carroll. 2016. Locally Downscaled and Spatially Customizable Climate Data for Historical and Future Periods for North America. PLoS ONE 11:e0156720.