Journal: Global Ecology and Biogeography Title: A Changing Climate is Snuffing Out Post-Fire Recovery in Montane Forests Authors: Kyle C. Rodman, Thomas T. Veblen, Mike A. Battaglia, Marin E. Chambers, Paula J. Fornwalt, Zachary A. Holden, Thomas E. Kolb, Monica T. Rother, Jessica R. Ouzts. Year: 2020 ----------------------------------------------------------------------------------------------------------------- Archive Size and Compression Files archived in .7z format using 7zip v. 16.04 on Windows v. 10.0.18362.657 Download size of final archived files = 13.1 gb ----------------------------------------------------------------------------------------------------------------- Data Archive Description and Structure Brief descriptions of each dataset included in the Dryad archive are described below. Numbers correspond to each compressed directory. Subfolders or individual files within each directory that are explained in greater detail are preceded by "-" characters. MD5 codes for each compressed directory are listed below individual names and descriptions. ----------------------------------------------------------------------------------------------------------------- Gridded Climate Data These data are separated into four individual compressed directories to allow for easier download and use (see numbers 1-4 below). They include all c. 250m climate grids used in this study, with the exception of 2006-2020 grids that were used to calculate running 30-year means for early years in the future period (Rodman et al. 2020a). These data are clipped to the bounds of the Southern Rocky Mountains Ecoregion (EPA Level III Ecoregion #21). All variables were initially calculated at c. 4km using the water balance equations of Dobrowski et al. (2013) and spatially downscaled using GIDS interpolation (Flint and Flint 2012). For a further description of processing, see Appendix S2 in Rodman et al. (2020a). AET is the total actual evapotranspiration (evaporation constrained by moisture availability) for the calendar year. CWD is total climatic water deficit (unmet evaporative demand of the atmosphere) for the calendar year. PPT is total growing season (April-September) precipitation during the calendar year. Though originally calculated in 32-bit floating point format, each file has now been cast to 16-bit integer to reduce file sizes. Therefore, units for each file are in mm * 10, to preserve one decimal unit of precision in conversion to integer values. For example 7602 in a grid cell from an annual layer of CWD corresponds to climatic water deficit of 760.2 mm during that year in a c. 250 m cell. If using these data, please cite Rodman et al. (2020a). Contact Kyle Rodman (kyle.rodman@colorado.edu) for further information about these data. 1) Directory "climateGridsFuture_CSIRO_2021_2099.7z" contains annual climate variables calculated from the MACA dataset of downscaled future climate projections for the southern Rocky Mountains derived from outputs of the CSIRO MK3.6.0 contribution to CMIP5 (under RCPs 4.5 and 8.5). Naming convention follows the format [Climate Variable]_CSIRO_[RCP]_[Year].tif. All files are in geotiff format and projected to NAD83 UTM Zone 13N (EPSG code 26913). File Size: 3.93 gb md5: cc6372cf706c2ed22dd400737fa46918 2) Directory "climateGridsFuture_GFDL_2021_2099.7z" contains annual climate variables calculated from the MACA dataset of downscaled future climate projections for the southern Rocky Mountains derived from outputs of the GFDL ESM2G contribution to CMIP5 (under RCPs 4.5 and 8.5). Naming convention follows the format [Climate Variable]_GFDL_[RCP]_[Year].tif. All files are in geotiff format and projected to NAD83 UTM Zone 13N (EPSG code 26913). File Size: 3.93 gb md5: c7cf47942f50670f75f7958774f37770 3) Directory "climateGridsFuture_IPSL_2021_2099.7z" contains annual climate variables calculated from the MACA dataset of downscaled future climate projections for the southern Rocky Mountains derived from outputs of the IPSL CM5A MR contribution to CMIP5 (under RCPs 4.5 and 8.5). Naming convention follows the format [Climate Variable]_IPSL_[RCP]_[Year].tif. All files are in geotiff format and projected to NAD83 UTM Zone 13N (EPSG code 26913). File Size: 3.93 gb md5: 547345aff030c45046d86e340662f49c 4) Directory "climateGridsRecent_1981_2015.7z" contains annual climate variables calculated from GridMET for the period of 1981-2015. Naming convention follows the following format. [apr_sep_ppt OR GridMET_AET_histor OR GridMET_CWD_histor]_[Year].tif. All files are in geotiff format and projected to NAD83 UTM Zone 13N (EPSG code 26913). File Size: 860.5 mb md5: d7b11bad36de6c87dc103344bec76d95 ----------------------------------------------------------------------------------------------------------------- Other Archived Datasets - Field Data, Statistical Model Outputs, and Other Spatial Data 5) Directory "ExampleSpatialModels.7z" contains subfolders "Abundance_nYrsGreaterHistDensities" and "EstablishmentSuitability." File Size: 190.8 mb md5: 97aad5e9e5fb19d6d0164bd82d46b53f - The subfolder "Abundance_nYrsGreaterHistDensities" includes summaries of generalized linear mixed model predictions for post-fire abundance of ponderosa pine and Douglas-fir at 60-m resolution throughout the montane zone of the Southern Rocky Mountains Ecoregion. We include only summarized values (used in Figure 6 in Rodman et al. 2020a) in the archive because including all annual projections results in an excessively large file, and all underlying data and statistical models used to make these predictions are included in this archive. Values in each image range 0-30, where cell-specific values give the number of years in a specified 30-year period in which post-fire regeneration could be expected to exceed minimum historical densities in the region of 50 ponderosa pine/ha or 15 Douglas-fir/ha. Naming convention follows the following format. [Species]_[GCM]_[RCP]_[Years].tif for future projections and [Species]GridMET1981_2010_25thpercent.tif for the recent (1981-2010) period. All files are in geotiff format and projected to NAD83 UTM Zone 13N (EPSG code 26913). - The subfolder "EstablishmentSuitability" includes a .csv file with predictions of the proportion of the montane zone of the Southern Rocky Mountains Ecoregion that is considered "above average" for post-fire seedling establishment in a given year, GCM, and RCP. These are derived from boosted regression tree analyses of annual establishment counts within each fire perimeter (Figures 3 and 5 in Rodman et al. 2020a). The included .csv file "projectedEstabSuitability.csv" is organized by year, with each GCM and RCP combination as a column. Column names follow the naming convention: [RCP]_[GCM] for future projections, and "obs_gridMET" gives predictions in the 1981-2015 period. "No data" values are -999. Values in BRT projections assume time since fire of 1 year. 6) Directory "FieldData.7z" includes two .xlsx files that summarize the underlying field data included in this study. Due to variations in data formats of the original studies, we include only the processed and cleaned data used in the analyses for this study. Raw data may be made available upon request. File Size: 2.1 mb md5: 50ecf0955065bb5337815c6a72722bdc - abundanceData.xlsx: A summary of post-fire seedling abundance, forest structure, ground cover, topography, and climate for each of 1301 field plots spanning 22 wildfires. - estabData.xlsx: A summary of counts of establishing seedlings dated to specific years within each fire. This is based on 717 destructive samples spanning 16 wildfires. Metadata is included within each file as a tab entitled "Metadata" that includes descriptions of data structure and definitions for each column. If using these field data for research purposes, we encourage the user to contact the study authors about opportunities to collaborate using these and related data. Also, we encourage users to cite the original studies that collected these data. Specifically, for abundance data, please refer to Ouzts et al. (2015; contributor "JO"), Chambers et al. (2016; contributor "MC"), Rother and Veblen (2016; contributor "MTR") and Rodman et al. (2020b; contributor, "KR"). For original studies of establishment data, please refer to Rother and Veblen (2017; for fires "bck", "eld", "hay", "him", "ovd") and Rodman et al. (2020; for other fires). See contributor contact information below: Contributor Name Contributor Code Affiliation Contact Information Marin E Chambers MC Colorado Forest Restoration Institute marin.chambers@colostate.edu Monica T Rother MTR University of North Carolina - Wilmington rotherm@uncw.edu Kyle C Rodman KR University of Colorado Boulder kyle.rodman@colorado.edu Jessica R Ouzts JO US Forest Service, Kaibab National Forest jessica.ouzts@usda.gov 7) Directory "R_ModelObjects.7z" gives .rds files from final statistical models included in Rodman et al. (2020a). These model objects were fitted with R packages "gbm" v. 2.1.5 and "dismo" v. 1.1-4 (for estabModel.rds), and "glmmTMB" v. 0.2.3 (for pipoAbundance.rds and psmeAbundance.rds). In combination with other data in this archive, these statistical models allow for reproduction of our results. File Size: 1.0 mb md5: c39da3d4dc71768d0deb1f3f1324402b 8) Directory "TerrainVariablesAndAWC.7z" includes other spatial datasets used in our statistical models and water balance calculations. File Size: 281.1 mb md5: 2260c4c64112e0a82cde4cda1b041be7 - Heatload_MK_Eq2_60m.tif: Heatload calculated following equation 2 in McCune and Keon (2002). Calculated using c. 30 m DEMs, and reprojected to 60m in NAD83 UTM Zone 13N (EPSG code 26913). - soil_AWC_4km_Polaris.tif: Soil available water capacity in the top 200 cm of soil derived from soil depth and fractional water capacity in the POLARIS database. Initially calculated at 30 m and aggregated to 4km for water balance model calculations. - tpi_15x15.tif: Topographic position index (TPI) calculated in a 15x15 cell neighborhood surrounding the focal cell. TPI values give the difference between cell elevation and the mean elevation of the surrounding area. We initially calculated TPI with a c. 30m DEM and later reprojected these data to 60m in NAD83 UTM Zone 13N (EPSG code 26913). 9) Directory "StudyFirePerimeters.7z" includes two shapefiles that give the perimeters of fires (i.e., fire_polys) and the Southern Rocky Mountains Ecoregion (i.e., SRME_bound) used in this study. File Size: 519.0 kb md5: 92baaf937b4274a075d72a2feae35dc3 ----------------------------------------------------------------------------------------------------------------- Relevant studies: Chambers, M. E., P. J. Fornwalt, S. L. Malone, and M. A. Battaglia. 2016. Patterns of Conifer Regeneration Following High Severity Wildfire in Ponderosa Pine – Dominated Forests of the Colorado Front Range. Forest Ecology and Management 378:57-67. Dobrowski, S. Z., J. Abatzoglou, and A. K. Swanson. 2013. The Climate Velocity of the Contiguous United States During the 20th Century. Global Change Biology 19:241-251. Flint, L. E., and A. L. Flint. 2012. Downscaling Future Climate Scenarios to Fine Scales for Hydrologic and Ecological Modeling and Analysis. Ecological Processes 1:1-15. McCune, B., and D. Keon. 2002. Equations for Potential Annual Direct Incident Radiation and Heat Load. Journal of Vegetation Science 13:603-606. Ouzts, J. R., Kolb, T. E., Huffman, D. W., and A. J. Sanchez Meador. 2015. Post-fire Ponderosa Pine Regeneration With and Without Planting in Arizona and New Mexico. Forest Ecology and Management 354:281-290. Rother, M. T., and T. T. Veblen. 2016. Limited Conifer Regeneration Following Wildfires in Dry Ponderosa Pine Forests of the Colorado Front Range. Ecosphere 7:e01594. Rother, M. T., and T. T. Veblen. 2017. Climate Drives Episodic Conifer Establishment after Fire in Dry Ponderosa Pine Forests of the Colorado. Forests 8:1-14. Rodman, K C., Battaglia, M. A., Chambers, M. E., Fornwalt, P. J., Holden, Z. A., Kolb, T. E., Rother, M. T., and J. R. Ouzts. 2020a (In Press). A Changing Climate is Snuffing Out Post-Fire Recovery in Montane Forests. Global Ecology and Biogeography. Rodman, K. C., Veblen, T. T., Chapman, T. B., Rother, M. T., Wion, A. P., and M. D. Redmond. 2020b. Limitations to Recovery Following Wildfire in Dry Forests of southern Colorado and northern New Mexico, USA. Ecological Applications 30:e02001.