U.S. Geological Survey
Matthew Rigge
20201112
Projections of Rangeland Fractional Component Cover Across the Sagebrush Biome for Representative Concentration Pathways (RCP) 4.5 and 8.5 Scenarios for the 2020s, 2050s, and 2080s Time-Periods (ver. 1.1, April 2022)
Raster Digital Data Set
Earth Resources Observation and Science Center, Sioux Falls, SD
U.S. Geological Survey
Rigge, M., Shi, H., Postma, K. In Review. Projected change in rangeland component fractional cover across the sagebrush biome through 2085. Submitted to Ecological Applications.
Rigge, M., Homer, C., Shi, H., Meyer, D., Bunde, B., Granneman, B., Postma, K., Danielson, P., Case, A., and Xian, G. In Review. Trends in rangelands fractional components across the western U.S. from 1985-2018. Submitted to Rangeland Ecology and Management.
Rigge, M., H. Shi, C. Homer, P. Danielson, and B. Granneman. 2019. Long-term trajectories of fractional component change in the Northern Great Basin, USA. Ecosphere: e02762.
RuleQuest Research. Cubist, version 2.05; Rule-Quest Pty, St Ives: New South Wales, Australia, 2008.
Wylie, B. K., N. J. Pastick, J. J. Picotte, and C. A. Deering.2018. Geospatial data mining for digital raster mapping. GIScience & Remote Sensing 56: 406–429.
https://doi.org/10.5066/P9EC2094
Matthew Rigge
Hua Shi
Kory Postma
20201116
Projected change in rangeland component fractional cover across the sagebrush biome through 2085.
publication
Climate change over the past century has altered vegetation community composition and species distributions across rangelands in the western United States. The scale and magnitude of climatic influences are unknown. While a number of studies have projected the impacts of climate change using several modeling approaches, none has evaluated impacts to fractional component cover at a 30-m resolution across the full sagebrush (Artemisia spp.) biome. We used fractional component cover data for rangeland functional groups and weather data from the 1985 to 2018 reference period in conjunction with soils and topography data to develop empirical models describing the spatio-temporal variation in component cover. To investigate the ramifications of future change across the western US, we extended models based on historical relationships over the reference period to model landscape effects based on future weather conditions from two emissions scenarios (Representative Concentration Pathways [RCP] 4.5 and 8.5) and three time periods (2020s, 2050s, and 2080s). We tested both Generalized Additive Models (GAMs) and regression tree models, finding that the former led to superior spatial and statistical results. Our results suggest more xeric vegetation across most of the study area, with an increasing dominance of non-sagebrush shrubs, annual herbaceous, and bare ground over herbaceous and sagebrush cover in both RCP scenarios. In general, both scenarios yielded similar results, but RCP 8.5 tended to be more extreme, with greater change relative to the reference period. Results demonstrate that in cool sites some degree of warming to growing season maximum temperature or non-growing season minimum temperature could be beneficial to sagebrush and shrub growth. This is not the case, regardless of temperature, for non-growing season maximum temperature. This information can be used to inform management to prepare for future vegetation composition and cover through the prioritization of conservation and restoration and shed light on species’ range shifts.
*Please note that some suspect increases in sagebrush cover are projected in the Southern Great Basin and Southern Colorado Plateau, which we interpret as model error. We have developed a mask (see below) to apply to the future sagebrush cover layers. Masked areas meet three criteria: 1) they are within the sagebrush biome (Jeffries and Finn 2019), 2) within the following EPA level 3 ecoregions: Arizona/New Mexico Mountains, Arizona/New Mexico Plateau (excluding the San Luis Valley from the mask), Colorado Plateaus (excluding the Uinta Basin from the mask), Mojave Basin and Range, and Sonoran Basin and Range, and 3) have a maximum of 0% sagebrush cover observed over the 1985-2020 RCMAP time-series.
We use USGS “Back-in-Time” (BIT) fractional component cover data for rangeland functional groups (sagebrush, shrub, herbaceous, annual herbaceous, litter, and bare ground) and climate data from 1985 to 2018 (Rigge et al. 2019 ) in conjunction with soils and topography data to develop models describing the empirical spatio-temporal variation in cover across the sagebrush biome of the Western U.S (Rigge et al. In Review). Our primary objective is to apply the models developed over the 1985-2018 reference period to future weather conditions based on two emissions scenarios (Representative Concentration Pathways [RCP] 4.5 and 8.5) and three time periods (2020s, 2050s, and 2080s) to evaluate change in functional group abundance within the respective range of each across an unprecedented extent. The RCP climate projections represent the average of 15 models (CanESM2, ACCESS1.0, IPSL-CM5A-MR, MIROC5, MPI-ESM-LR, CCSM4, HadGEM2-ES, CNRM-CM5, CSIRO Mk 3.6, GFDL-CM3, INM-CM4, MRI-CGCM3, MIROC-ESM, CESM1-CAM5, GISS-E2R). We then model component cover under these scenarios and evaluate spatial and temporal patterns of change relative to the reference period. The projections can be used to inform management to prepare for future vegetation composition and cover through the prioritization of conservation and restoration and shed light on species range shifts.
Although this Federal Geographic Data Committee-compliant metadata file is intended to document the data set in nonproprietary form, as well as in ESRI format, this metadata file may include some ESRI-specific terminology.
20200101
20850101
projected
As needed
-124.7958
-101.2082
50.8481
32.5198
ISO 19115 Topic Category
biota
environment
geoscientificInformation
imageryBaseMapsEarthCover
USGS Thesaurus
shrubland ecosystems
terrestrial ecosystems
Alexandria Digital Library Feature Type Thesaurus
shrublands
trends
grassland change
shrubland change
vegetation change
climate change
rangeland management
None
shrub
sagebrush
herbaceous
annual herbaceous
litter
grass
vegetation
bare ground
rangeland
shrubland
Common geographic areas
United States
Washington
Oregon
Montana
North Dakota
South Dakota
Wyoming
Idaho
Nebraska
Nevada
Utah
California
Colorado
Arizona
New Mexico
None
NM
AZ
CO
CA
UT
NV
NE
ID
WY
SD
ND
MT
OR
WA
Great Basin
Arizona Plateau
Black Hills
Blue Mountains
Colorado Plateau
Columbia Plateau
Grand Canyon
Middle Rockies
Rocky Mountains
Gunnison
Sonoran Desert
Southwest Tablelands
Three Forks
Wasatch
Western US
Yellowstone
Northern Mountains
Snake River Plain
Any downloading and use of these data signifies a user's agreement to comprehension and compliance of the USGS Standard Disclaimer. Ensure all portions of metadata are read and clearly understood before using these data in order to protect both user and USGS interests.
There is no guarantee of warranty concerning the accuracy of the data. Users should be aware that these data were developed from models which can contain some local error. Users should not use these data for critical applications without a full awareness of their limitations. Acknowledgement of the originating agencies would be appreciated in products derived from these data. Any user who modifies these data is obligated to describe the types of modifications they perform. User specifically agrees not to misrepresent the data, nor to imply that changes made were approved or endorsed by the U.S. Geological Survey. Please refer to https://www.usgs.gov/privacy.html for the USGS disclaimer.
U.S. Geological Survey
Customer Services Representative
mailing and physical
47914 252nd Street
Sioux Falls
SD
57198-0001
U.S.
605-594-6151
605-594-6589
custserv@usgs.gov
U.S. Geological Survey (USGS), Bureau of Land Management (BLM)
Environment as of Metadata Creation: Microsoft Windows 7 Version 6.1 (Build 7601) Service Pack 1; Esri ArcGIS 10.5.1 (Build 7333) Service Pack N/A (Build N/A)
We inserted 1990, 2000, and 2010 climate data into the climate models to evaluate their degree of correspondence with 1990, 2000, and 2010 BIT predictions (as a cross-validation). We found similar mean component cover predictions between the GAM and BIT data. However, GAM standard deviation values tended to be smaller than BIT, indicating that GAM predictions were somewhat flatter, possibly due to a lack of imagery input or model generalization. Similarly, GAM predictions have a lower range of component cover predictions relative to BIT, though these portions of component histograms excluded by GAM tended to be quite uncommon in the reference period. Spatial correlations between GAM and BIT predictions are strong however, averaging an r of 0.66. Broken down by component, the spatial correlations averaged 0.62, 0.55, 0.80, 0.58, 0.61, and 0.78 for shrub, sagebrush, herbaceous, annual herbaceous, litter, and bare ground, respectively.
For the period of 1985-2018 we developed empirical spatio-temporal relationships between annual maps of fractional cover of rangeland components, weather data, topography, and soils variables. These relationships were obtained by testing two different models (Cubist Regression Tree [RT] [RuleQuest Research 2008] and Generalized Additive Models [GAM]). Both models were trained across the entire study area and time period in undisturbed rangelands. Next, we applied the models developed for range of conditions observed in the reference period to projected future climate data based on two emissions scenarios and three time periods. Since climate data are the only model inputs that change between reference period and future conditions, they are the sole driver of change in the future, though the soils and topography modulate the response.
This fractional estimation of six shrubland habitat variables for future conditions in the 2020s, 2050s, and 2080s across the sagebrush biome is considered complete for the information presented, as described in the abstract. Users are advised to read the rest of the metadata record carefully for additional details.
A formal accuracy assessment of the horizontal positional information in the data set has not been conducted.
A formal accuracy assessment of the vertical positional information in the data set has either not been conducted, or is not applicable.
For the period of 1985-2018 we developed empirical spatio-temporal relationships between annual maps of fractional cover of rangeland components, weather data, topography, and soils variables. These relationships were obtained by testing two different models (Cubist Regression Tree [RT] [RuleQuest Research 2008] and Generalized Additive Models [GAM]). Both models were trained across the entire study area and time period in undisturbed rangelands. Next, we applied the models developed for range of conditions observed in the reference period to projected future climate data based on two emissions scenarios and three time periods. Since climate data are the only model inputs that change between reference period and future conditions, they are the sole driver of change in the future, though the soils and topography modulate the response. Products were masked using the base land cover mask based on pixels identified as non-rangeland in the circa 2016 base year.
For the climate projections, we used monthly data downscaled using bilinear interpolation to 1000-m across the Western U.S. Projections were from an ensemble of 15 Coupled Model intercomparison project phase 5 atmosphere-ocean general circulation models (obtained from the AdaptWest Project). We obtained data for the low emissions scenario (RCP 4.5, with stabilized emissions with 4.5 W/m2 of radiative forcing and high emissions scenario (RCP 8.5, with increased emissions through early to mid-century) with 8.5 W/m2 of radiative forcing. Data were available for three time periods, 2020s, 2050s, and 2080s, representing the average of 2011-2040, 2041-2070, and 2071-2100, respectively. Using these data, we calculated the Growing Season (GS) (April-September) and Non-Growing Season (NGS) (October-March) average minimum and maximum temperature, and the GS and NGS precipitation totals for each year. This process yielded six annual climate variables GSPRCP, NGSPRCP, GSTMAX, NGSTMAX, GSTMIN, and NGSTMIN.
Training data were prepared by placing random points (n = 90,000 per year, total n = 2,880,000) within the rangeland portions of the sagebrush biome. We excluded areas identified by MTBS as burned, those with implemented vegetation treatment as identified by the Land Treatment Digital Library from 1985 to 2018. The goal of excluding known change is to derive a training dataset reflective of only climate/biophysical influence. Independent data variables (n = 20) were extracted to these points; 1) 1985-2018 climate data for six variables (GSPRCP, NGSPRCP, GSTMIN, NGSTMIN, GSTMAX, and NGSTMAX), 2) four topographic variables, and 3) five soils variables for two depth increments each. Additionally, we extracted the dependent variables, the 1985-2018 fractional component cover of 6 components, to the points. We transposed the data, so our training is spatio-temporal, which avoids some of the limitations of spatial and temporal models. Our training data include the influence of climate change and interannual climate variation in the 1985-2018 period.
We evaluated GAM models based on their ability to minimize residuals and have the most robust relationships with training data. Specifically, we generated GAM predictions for 1990, 2000, and 2010 and compared them to respective years of BIT predictions. For the GAM models we tested several degrees of freedom options; k=40, 6, and 3. Among the GAM models, 6 degrees of freedom produced the strongest models with the least amount of spurious relationships included (observed in the k=40 models), while not oversimplifying relationships (observed in the k=3 models). These were selected as the final models.
Based on the final selected GAM model developed for each component in the reference period, we replaced climate data with future projections for the 2020s, 2050s, and 2080s, independently. We then predicted future component cover spatially at 30-m resolution across the study area based on the empirical relationships observed in space-time of the reference period. We evaluated the occurrence and degree of extrapolation to novel climate conditions in future projections relative to the reference period. This is critical as spatial and temporal relationships between biomass production (or cover) and precipitation should not be extrapolated. Relative to the climate conditions observed through time and space in the reference period, we found very little novel climate conditions, even in the most extreme climate scenario of RCP 8.5 in the 2080s. This demonstrates that model extrapolation to novel climate conditions is limited.
Following the completion of all GAM predictions we implemented a post-processing procedure on each set of component predictions for each year/scenario. This process ensured the sum of primary components (bare ground, herbaceous, litter, and shrub cover) equaled ~100% by keeping the original modelled proportions intact. Next, we ensured that shrub cover was greater than sagebrush cover, and herbaceous cover was greater than annual herbaceous cover.
20200901
Raster
Grid Cell
59901
55334
1
Albers Conical Equal Area
29.5
45.5
-96.0
23.0
0.0
0.0
row and column
30.0
30.0
meters
WGS_1984
WGS 84
6378137.0
298.257223563
Attribute Table
Table containing attribute information associated with each component projection layer.
Producer Defined
Value
Column for value indicating per-pixel percent cover for bare ground, shrub, herbaceous, litter, sagebrush, and annual herbaceous, component range from 0 to 100, a value of 255 for the masked areas and for those outside our mapping extent.
Producer Defined
0.0
255.0
Percent
Count
All raster attribute tables include a column for count describing a nominal integer value that designates the number of pixels that have each value in the file; histogram column in ERDAS Imagine raster attributes table.
Producer Defined
0
382252689
Integer
MaskAreas that were masked outProducer Defined
1
value of 1 indicates areas of suspect sagebrush cover projected in the future to be masked (excluded)
Producer defined
The entity and attribute information provided here describes the tabular data associated with the data set. Please review the detailed descriptions that are provided (the individual attribute descriptions) for information on the values that appear as fields/table entries of the data set.
The entity and attribute information was generated by the individual and/or agency identified as the originator of the data set. Please review the rest of the metadata record for additional details and information.
U.S. Geological Survey
GS ScienceBase
mailing address
Denver Federal Center, Building 810, Mail Stop 302
Denver
CO
80225
United States
1-888-275-8747
sciencebase@usgs.gov
Unless otherwise stated, all data, metadata and related materials are considered to satisfy the quality standards relative to the purpose for which the data were collected. Although these data and associated metadata have been reviewed for accuracy and completeness and approved for release by the U.S. Geological Survey (USGS), no warranty expressed or implied is made regarding the display or utility of the data for other purposes, nor on all computer systems, nor shall the act of distribution constitute any such warranty.
Digital Data
https://doi.org/10.5066/P9EC2094
None
20220406
U.S. Geological Survey
Customer Services Representative
mailing and physical
47914 252nd Street
Sioux Falls
SD
57198-0001
U.S.
605-594-6151
605-594-6589
custserv@usgs.gov
FGDC-STD-001-1998
FGDC Content Standard for Digital Geospatial Metadata