Skip to main content
Dryad

Data from: Climatically robust multi-scale species distribution models to support pronghorn recovery in California

Cite this dataset

Bean, William T.; Butterfield, Scott; Howard, Jeanette; Batter, Thomas (2024). Data from: Climatically robust multi-scale species distribution models to support pronghorn recovery in California [Dataset]. Dryad. https://doi.org/10.5061/dryad.bcc2fqzmx

Abstract

We combined two climate-based distribution models with three finer-scale suitability models to identify habitat for pronghorn recovery in California now and into the future.

Location: California, United States 

Methods: We used a consensus approach to identify areas of suitable climate now (1980-2010) and future (2031-2060) for pronghorn in California. We compared the results of models from two separate hypotheses about their historical ecology in the state, specifically the migration hypothesis and the niche reduction hypothesis. We combined occurrences from GPS collars distributed across three populations of pronghorn in the state to create three distinct habitat models: (1) an ensemble model using Random Forests, Maxent, Classification and Regression Trees, and a Generalized Linear Model; (2) a step selection function; and (3) an expert-driven model. We evaluated consensus among both the climate models and the suitability models to prioritize areas for, and evaluate the prospects of, pronghorn recovery. 

Results: Climate suitability for pronghorn in the future depends heavily on model assumptions. Under the migration hypothesis, our model predicted that there will be on suitable climate in California in the future. Under the niche reduction hypothesis, by contrast, suitable climate will expand. Habitat also depended on the methods used, but areas of consensus among all three exist in large patches throughout the state.

Main Conclusions: Identifying habitat for a species which has undergone extreme range collapse, and which has very fine scale habitat needs, presents novel challenges for spatial ecologists. Our multi-method, multi-hypothesis approach can allow habitat modelers to identify areas of consensus and, perhaps more importantly, critical knowledge gaps that could resolve disagreements among the models. For pronghorn, a better understanding of their upper thermal tolerances and whether historical populations migrated will be crucial to their potential recovery in California and throughout the arid Southwest.

README: Climatically robust multi-scale species distribution models to support pronghorn recovery in California

https://doi.org/10.5061/dryad.bcc2fqzmx

Data include raw pronghorn locations and environmental predictors to generate distribution and habitat models, and the outputs of these models.

Description of the data and file structure

Data are of three kinds:

  1. Pronghorn locations ("all_gbif pronghorn.csv"), downloaded from the Global Biodiversity Information Facility (Supplemental information on Zenodo)
  2. Environmental predictors used to generate habitat or climate models (all distances in meters)
    1. "ag_agg.tif": proportion cover of agriculture land type, derived from the CalFire vegetation layer
    2. "all_road dens.tif": density of all roads within the raster cell, derived from TIGER lines
    3. "barren_agg.tif": proportion cover of barren land type, derived from the CalFire vegetation layer
    4. "bulk_dens agg.tif": bulk density of soil, derived from the Polaris soil data (Chaney et al. 2019)
    5. "desert_agg.tif": proportion cover of desert land type, derived from the CalFire vegetation layer
    6. "elev_agg.tif": elevation, derived from the National Elevation Dataset
    7. "grass_agg.tif": proportion cover of grassland land type, derived from the CalFire vegetation layer
    8. "maj_road dist.tif": distance to major roads (federal and state highways and county roads)
    9. "perc_clay agg.tif": percent clay of soil, derived from the Polaris soil data (Chaney et al. 2019)
    10. "ph_agg.tif": pH of soil, derived from the Polaris soil data (Chaney et al. 2019)
    11. "road_dist1.tif": distance to all roads, derived from TIGER lines
    12. "shrub_agg.tif": proportion cover of shrub land type, derived from the CalFire vegetation layer
    13. "slope_agg.tif": slope, derived from the National Elevation Dataset (in degrees)
    14. "tpi_fine-agg.tif": topographic position index, derived from the National Elevation Dataset
    15. "tri_fine-agg.tif": terrain ruggedness index, derived from the National Elevation Dataset
    16. "urban_agg.tif": proportion cover of urban land type, derived from the CalFire vegetation layer
  3. Outputs of distribution and habitat suitability models
    1. "combined-2041-thresh05-max.tif": estimated future suitability combining models of winter and summer distribution under migration hypothesis, using a threshold to include 95% of pronghorn occurrences in the contemporary model
    2. "expert-all.tif": output of the expert-driven suitability model
    3. "expert_future.tif": expert driven suitability model, clipped to areas of projected future climate suitability
    4. "expert-climate.tif": expert driven suitability model, clipped to areas of estimated contemporary climate suitability
    5. "expert-no_human.tif": expert driven suitability model, with all human-dominated (i.e. urban and agriculture) layers removed
    6. "gps_ensemble.tif": ensemble model of habitat suitability based on GPS data from three pronghorn populations in California
    7. "gps_low-p-thresh01-climthresh.tif": reduced predictor ensemble model of suitability based on GPS data, thresholded to include 99% of all GPS points and clipped to areas of contemporary suitability
    8. "gps_lowp-thresho01-future.tif": reduced predictor ensemble model of suitability based on GPS data, thresholded to include 99% of all GPS points and clipped to areas of future suitability
    9. "gps_lowp-thresh01.tif": reduced predictor ensemble model of suitability based on GPS data, thresholded to include 99% of all GPS points
    10. "gps_lowpred-ensemble.tif": reduced predictor ensemble model of suitability based on GPS data
    11. "hist_pronghorn-aea.tif" and "hist-pronghorn-wgs.tif": model of pronghorn climatic suitability in California based on niche reduction hypothesis using historical distribution of pronghorn (aea = Albers Equal Area projection; WGS = unprojected, dataum = WGS84)
    12. "hist_pronghorn-thresh05.tif": model of pronghorn climatic suitability in California based on niche reduction hypothesis using historical distribution of pronghorn, thresholded to include 95% of historical pronghorn locations
    13. "hist-2041-mean.tif": mean future climatic suitability projected for 2041-2060 based on niche reduction hypothesis
    14. "hist-2041-thresh05.tif": sum of all future climatic suitability projections based on niche reduction hypothesis, thresholded to include 95% of historical pronghorn locations
    15. "hsm_thresh-consensus-future.tif": sum of all future climatic suitability projections from ensemble suitability model, thresholded to include 99% of pronghorn locations
    16. "hsm_thresh-consensus.tif": mean of all future climatic suitability projections from ensemble suitability model
    17. "occ_ensemble-lowp.tif": ensemble model of habitat suitability using occurrences collected by California Department of Fish and Wildlife and a reduced set of habitat predictors
    18. "ssf_hsm.tif": contemporary habitat suitability model based on step-selection function using GPS data from three pronghorn populations in California
    19. "ssf_hsm-thresh01.tif": contemporary habitat suitability model based on step-selection function using GPS data from three pronghorn populations in California, thresholded to include 99% of pronghorn GPS locations
    20. "ssf_hsm-thresh01-climthresh.tif": contemporary habitat suitability model based on step-selection function using GPS data from three pronghorn populations in California, thresholded to include 99% of pronghorn GPS locations and clipped to contemporary climate suitability
    21. "ssf_thresh01-future.tif": contemporary habitat suitability model based on step-selection function using GPS data from three pronghorn populations in California, thresholded to include 99% of pronghorn GPS locations and clipped to future climate suitability
    22. "summer-only-suit.tif": model of contemporary climatic suitability in summer for pronghorn under migration hypothesis
    23. "summer_thresh-15th.tif": model of contemporary climatic suitability in summer for pronghorn under migration hypothesis clipped to include 85% of all pronghorn locations
    24. "summer-proj-2041.tif": model of future climatic suitability in summer for pronghorn under migration hypothesis, stack of all scenarios
    25. "summer-2041-mean.tif": model of future climatic suitability in summer for pronghorn under migration hypothesis, mean
    26. "summer-2041-thresh05.tif": model of future climatic suitability in summer for pronghorn under migration hypothesis, summed all thresholds
    27. "winter-only-suit.tif": model of contemporary climatic suitability in winter for pronghorn under migration hypothesis
    28. "winter_thresh-15th.tif": model of contemporary climatic suitability in winter for pronghorn under migration hypothesis clipped to include 85% of all pronghorn locations
    29. "winter-proj-2041.tif": model of future climatic suitability in winter for pronghorn under migration hypothesis, stack of all scenarios
    30. "winter-2041-mean.tif": model of future climatic suitability in winter for pronghorn under migration hypothesis, mean
    31. "winter-2041-thresh05.tif": model of future climatic suitability in winter for pronghorn under migration hypothesis, summed all thresholds

Funding

The Nature Conservancy