Arid ecosystems are often vulnerable to transformation to invasive-dominated states following fire, but data on persistence of these states are sparse. The grass/fire cycle is a feedback process between invasive annual grasses and fire frequency that often leads to the formation of alternative vegetation states dominated by the invasive grasses. However, other components of fire regimes, such as burn severity, also have the potential to produce long-term vegetation transformations. Our goal was to evaluate the influence of both fire frequency and burn severity on the transformation of woody-dominated communities to communities dominated by invasive grasses in major elevation zones of the Mojave Desert of western North America. We used a chronosequence design to collect data on herbaceous and woody cover at 229 unburned reference plots and 578 plots that burned between 1972 and 2010. We stratified the plots by elevation zone (low, mid, high), fire frequency (1 to 3 times), and years postfire (YPF; 1 - 5, 6 - 10, 11 - 20, and 21 - 40 YPF). Burn severity for each plot was estimated by the difference normalized burn ratio (dNBR). A Geographic Positioning System (GPS) was used to match the plots as closely as possible with the corner of a dNBR pixel. Each plot was 0.1 ha (32 m x 32 m) and contained three randomly positioned 25-m transects. Point intercept sampling was conducted at 0.5 m intervals along each transect (N = 50 points per transect, 150 points per plot). All plants intercepted by a wooden rod (1mm diameter) were recorded to species at each point. Species cover in each plot was estimated as the sum of point intercepts for each species divided by 150.

This is a shapefile of the Mojave Desert, which was used as our study area boundary (MojaveEcoregion_TNS_UTM83.shp).

This is a shapefile of the Mojave Desert, which was used as our study area boundary (MojaveEcoregion_TNS_UTM83.shp).

This U.S. Geological Survey data release consists of 3 raster datasets representing estimates of probability of ignition (ProbIgnitPredict.tif), fire frequency (FrequencyPredictRF.tif), and burn severity (dNBRPredictRF.tif) in the Mojave Desert from 1984 to 2010. The data include: (1) A shapefile of the Mojave Desert that was used as our study area boundary (MojaveEcoregion_TNS_UTM83.shp). The original shapefile was obtained from NatureServe in 2009; (2) Three Tagged-Interchange Format (TIF) raster datasets representing probability of ignition, fire frequency, and burn severity. Resolution equals 30 meters, projection equals UTM Zone 11N.

This U.S. Geological Survey data release consists of 3 raster datasets representing estimates of probability of ignition (ProbIgnitPredict.tif), fire frequency (FrequencyPredictRF.tif), and burn severity (dNBRPredictRF.tif) in the Mojave Desert from 1984 to 2010. The data include: (1) A shapefile of the Mojave Desert that was used as our study area boundary (MojaveEcoregion_TNS_UTM83.shp). The original shapefile was obtained from NatureServe in 2009; (2) Three Tagged-Interchange Format (TIF) raster datasets representing probability of ignition, fire frequency, and burn severity. Resolution equals 30 meters, projection equals UTM Zone 11N.

We assessed the impacts of co-occurring invasive plant species on fire regimes and postfire native communities in the Mojave Desert, western USA by analyzing the distribution and co-occurrence patterns of three invasive annual grasses known to alter fuel conditions and community structure: Red Brome (Bromus rubens), Cheatgrass (Bromus tectorum), and Mediterranean grass (Schismus spp.: Schismus arabicus and Schismus barbatus), and an invasive forb, red stemmed filaree (Erodium cicutarium) which can dominate postfire sites. We developed species distribution models (SDMs) for each of the four taxa and analyzed field plot data to assess the relationship between invasives and fire frequency, years postfire, and the impacts on postfire native herbaceous diversity. The data include: (1) A shapefile of the Mojave Desert that was used as our study area boundary (MojaveEcoregion_TNS_UTM83.shp). The original shapefile was obtained from NatureServe in 2009. (2) Two Comma Separated Values (CSV) files with the geographic location (Survey Plot Information.csv) and abundance (percent cover) of the four invasive taxa (Cover Measurements.csv) throughout the Mojave Desert. (3) Four Tagged-Interchange Format (TIF) raster datasets representing the SDMs. These data support the following publication: Underwood, E.C., Klinger, R.C. and Brooks M.L., 2019, Effects of invasive plants on fire regimes and postfire vegetation diversity in an arid ecosystem, Ecology and Evolution, 00:1-15. https://doi.org/10.1002/ece3.5650

We assessed the impacts of co-occurring invasive plant species on fire regimes and postfire native communities in the Mojave Desert, western USA by analyzing the distribution and co-occurrence patterns of three invasive annual grasses known to alter fuel conditions and community structure: Red Brome (Bromus rubens), Cheatgrass (Bromus tectorum), and Mediterranean grass (Schismus spp.: Schismus arabicus and Schismus barbatus), and an invasive forb, red stemmed filaree (Erodium cicutarium) which can dominate postfire sites. We developed species distribution models (SDMs) for each of the four taxa and analyzed field plot data to assess the relationship between invasives and fire frequency, years postfire, and the impacts on postfire native herbaceous diversity. The data include: (1) A shapefile of the Mojave Desert that was used as our study area boundary (MojaveEcoregion_TNS_UTM83.shp). The original shapefile was obtained from NatureServe in 2009. (2) Two Comma Separated Values (CSV) files with the geographic location (Survey Plot Information.csv) and abundance (percent cover) of the four invasive taxa (Cover Measurements.csv) throughout the Mojave Desert. (3) Four Tagged-Interchange Format (TIF) raster datasets representing the SDMs. These data support the following publication: Underwood, E.C., Klinger, R.C. and Brooks M.L., 2019, Effects of invasive plants on fire regimes and postfire vegetation diversity in an arid ecosystem, Ecology and Evolution, 00:1-15. https://doi.org/10.1002/ece3.5650

This dataset provides the plot-level data of the relative cover of ten different plant associations derived from a structural topic model and resistance and resilience metrics for predicting their abundances. Data comes from four different fires across the Great Basin: the 2007 Murphy Fire, 2012 Rush Fire, 2012 Holloway Fire, and the 2015 Soda Fire. Additional data utilized in the cross referenced paper from the Orchard Combat Training Center was redacted due to military rules but can be requested through the Idaho National Guard Environmental Management Office. Bureau of Land Management data (Murphy, Holloway, and Rush) species cover data was collected using line-point intercept methods on plots with between one and three transects of 25m or 50m in length. An additional small set of data (<1%) was collected by the Idaho Fish and Game using Daubenmire cover grids. Soda fire species cover data was collected with overhead photos and a grid-point intercept technique using Samplepoint software (Booth et al. 2006). Structural topic modelling was run to get the relative cover of ten different plant associations at each plot (Applestein et al. 2024).

This dataset provides the plot-level data of the relative cover of ten different plant associations derived from a structural topic model and resistance and resilience metrics for predicting their abundances. Data comes from four different fires across the Great Basin: the 2007 Murphy Fire, 2012 Rush Fire, 2012 Holloway Fire, and the 2015 Soda Fire. Additional data utilized in the cross referenced paper from the Orchard Combat Training Center was redacted due to military rules but can be requested through the Idaho National Guard Environmental Management Office. Bureau of Land Management data (Murphy, Holloway, and Rush) species cover data was collected using line-point intercept methods on plots with between one and three transects of 25m or 50m in length. An additional small set of data (<1%) was collected by the Idaho Fish and Game using Daubenmire cover grids. Soda fire species cover data was collected with overhead photos and a grid-point intercept technique using Samplepoint software (Booth et al. 2006). Structural topic modelling was run to get the relative cover of ten different plant associations at each plot (Applestein et al. 2024).

This raster dataset represents spatially explicit predictions of fire frequency in the Mojave Desert based on models developed from data on perimeters of fires greater than 405 hectares that burned between 1972 through 2010. Raster resolution equals 30 meters, projection equals UTM Zone 11N.