Joshua Caster Temuulen T Sankey Joel B Sankey Matthew A Bowker Daniel Buscombe Michael C Duniway Nichole Barger Akasha Faist Taylor Joyal Jessica Mikenas 20210913 comma-separated values (.csv) tabular data Flagstaff, AZ U.S. Geological Survey https://doi.org/10.5066/P9CNLJ25 Joshua Caster Temuulen Sankey Joel B Sankey Matthew A Bowker Daniel Buscombe Michael C. Duniway Nichole Barger Akasha Faist Taylor Joyal 2021 journal manuscript ScienceDirect (online) Geoderma https://doi.org/10.1016/j.geoderma.2021.115369 This data release presents data used for analyzing spatial and temporal differences in soil surface roughness within selected biocrust communities. These records were collected by ground-based lidar for 121, 1m x 3m soil plots with biological soil crusts (biocrusts). Roughness was estimated from 5 mm resolution data (CloudCompare v. 2.10.2, 2019) for two Great Basin Desert sites (UTTR-1; UTTR-2) in December 2015 and one Chihuahuan Desert site (JER) in February 2016. Data were again collected in June 2018 for UTTR-1 and UTTR-2. Additional field and laboratory data were included within this study to understand differences in soil surface roughness between UTTR and JER as well as between the 2016 and 2018 surveys at UTTR. The associated manuscript (see Larger Work Citation) reports that soil surfaces at the UTTR are significantly rougher than JER for undisturbed biocrust but not necessarily for disturbed biocrust. Additionaly, discuss that once biocrust is disturbed, soil surface roughness can increase with biocrust recovery. These data can be further used to represent surface conditions at the time of collection at individual soil plots within both study areas, providing further information on spatial and temporal variability in roughness as well as other field and laboratory parameters included in this release. The purpose of these tabular data were to assess microtopographic differences between biological soil crust (biocrust) communities in the “cool” Great Basin Desert and the “hot” Chihuahuan Desert. Ground-based lidar was used to capture surface conditions of mature, undisturbed biocrust communities and recently disturbed biocrust communities within these deserts. Additionally, we captured a second set of data in the cool desert more than two years after the first surveys to evaluate how these communities change over time. These data were used to evaluate differences between surface conditions at seven spatial scales defined by focal radii from 4 cm to 50 cm to better refine our understanding of why surfaces are different between hot and cool desert climates and how these surfaces change over short time intervals. Here we provide soil surface roughness at all 121 soil plots and seven spatial scales as well as the evaluated field and laboratory results that can be used to further assess spatial and temporal variability among single soil plots or the collective study areas. These data are representative of the time at collection and can provide a baseline for assessing future conditions at these locations. These data provide multi-scalar roughness estimates for 121 plots within two desert climates along with field and laboratory characterization of selected soil properties. These data were collected to estimate surficial differences between biological soil crust (biocrust) communities and provide metrics useful for studying the influence of climate, soil development, disturbance and post-disturbance recovery on the soil surface. These data were collected and processed specifically for analysis of biocrust and therefore are not representative of other soils within the Great Basin or Chihuahuan deserts. Specifically, roughness estimates were calculated from the ground surface and therefore may not directly relate to wind frictional thresholds that incorporate vegetation. Additionally, these data are representative of field conditions at the time of collection and surficial properties at these locations are likely to change on a seasonal basis. Roughness estimates were calculated from the ground surface. Data users should read the entire metadata record and acquire the manuscript identified as the ‘Larger Work Citation’, and/or review the 'Cross References' to have a complete understanding of how these data were created and used. The data are specific to the uses identified in this metadata record and the 'Larger Work Citation', and any other use of these data would be inappropriate. See 'Distribution liability' statements for more information. 2016 2017 2018 ground condition Complete None planned -115.0000 -105.000 42.000 30.000 USGS information products data release USGS Thesaurus biological soil crusts field experiments field inventory and monitoring field methods habitat alteration and disturbance laboratory methods laboratory experiments lidar plot sampling remediation soil resources USGS Biocomplexity Thesaurus biological treatment ecosystem disturbance soil morphology ISO 19115 Topic Categories biota geoscientificInformation USGS Metadata Identifier USGS:606645a6d34edc0435be54fe None 5 mm resolution biocrust digital point cloud experimental restoration treatments field data ground-based lidar laboratory data soil plots soil surface roughness Geographic Names Information System (GNIS) Chihuahuan Desert Great Basin Hill Air Force Base Jornada Experimental Range Las Cruces New Mexico Salt Lake City Utah None Great Basin Desert southwestern United States none none Joshua J Caster U.S. Geological Survey Student Trainee (Geography) mailing and physical

2255 North Gemini Drive

Flagstaff AZ 86001 US 928-556-7129 jcaster@usgs.gov We would like to thank the environmental and logistics coordinators and other personnel at JER and Hill UTTR Air Force Base that helped with planning and access. We gratefully acknowledge the effort by Jessica Mikenas, who assisted with the lidar surveys and collection of aggregate stability measurements as well as the field and lab crews within the U.S. Geological Survey’s Southwest Biological Science Center and University of Colorado, Boulder. Jayne Belnap Susan L Phillips Dana L Witwicki Mark E Miller 2008 ScienceDirect (online) Journal of Arid Environments https://doi.org/10.1016/j.jaridenv.2008.02.019 J. E Herrick W.G Whitford A.G de Soyza J.W Van Zee K.M Havstad C.A Seybold M Walton 2001 ScienceDirect (online) CATENA https://doi.org/10.1016/S0341-8162(00)00173-9 Akasha M. Faist Anita J. Antoninka Jayne Belnap Matthew A. Bowker Michael C. Duniway Ferran Garcia‐Pichel Corey Nelson Sasha C. Reed Ana Giraldo‐Silva Sergio Velasco‐Ayuso Nichole N. Barger 2020 Restoration Ecology Wiley Online Library https://doi.org/10.1111/rec.13087 No formal attribute accuracy tests were conducted No formal logical accuracy tests were conducted Data set 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. No formal positional accuracy tests were conducted No formal positional accuracy tests were conducted Development of the Faist Data table: These data were originally reported by Faist and others (2020) and include 1) biocrust plot development, 2) aggregate stability estimates, 3) Soil Chlorophyl-A, 4) biocrust level of development (LOD), and 5) Biocrust cover estimates. Development and Processing: 1) Biocrust soil plots were set as 1 m x 3 m areas that were either left undisturbed (control) or were distubed by scraping the upper 2.5 cm of the soil surface and trampeling by foot the surface to remove surface structure. A subset of disturbed plots were treated with a temporary physical stablizer and biocrust innoculum over 10% of the plot area. (Completed 2015) 2) Each plot was divided into thirds along the 3 m plot edge and these three subplots were systematically sampled for aggregate stability. We estimated aggregate stability from six soil aggregates, two from each of the three subplots, that were collected within the upper 0.3 cm of each plot, using the field-based soil stability technique of Herrick et al. (2001). We report the mean of the six aggregate stability ranks (1 as the lowest and 6 as the highest aggregate stability) for each plot collected at the same time as the 2016 and 2018 surveys. (Completed 2016 and 2018) 3) Soil chlorophyl-A (CHLA) multiple samples were selected to include representative dark and light colored biocrust within each plot that were homogenized into a single, dried laboratory sample. Chla was extracted using an acetone solvent and classified using an Ocean Optics CHEMUSB4-VIS-NIR spectrophotometer. Chla results and laboratory methods have been previously published by Faist et al. (2020) (Completed 2016 and 2017) 4) Biocrust Level of Development was estimated using the procedure by Belnap et al. (2008) which uses a published visual index for evaluating biocrust cover, color, and visual structure with 1 being the lowest and 6 being the highest.(Completed 2016 and 2017) 5) Biocrust cover was estimated for cyanobacteria, lichens, and bryophytes in each plot using point intercept surveys along a medial transect. (Completed 2016 and 2017) 2017 Development of the Lidar Derived Data table: Reported products were developed from ground-based lidar surveys conducted over three field dates. Survey derived products include 1) Soil surface roughness estimates, 2) dessication crack coverage, 3.) percentage of vegetation points, and 4) percentage of topographic change area. Development and Processing: 1) Survey was conducted for each soil plot at a distance of 3 m from one of the short (1 m) plot edges with a Reigl VZ 1000 terrestrial laser scanner. Surveys produced a point cloud with an estimated minimum point spacing of measurements at 0.1 cm. We filtered vegetation out of the point cloud within CloudCompare (CloudCompare v. 2.10.2, 2019) using an iterative height filter above the minimum surface elevations within a focal window of 20 cm to 1 cm, and a low point density filter for areas with dense vegetation. After filtering, each plot was subsampled to a 5 mm resolution which reduced effects of remaining vegetation outliers.(Completed in 2019) 2) We calculated surface roughness using CloudCompare v. 2.10.2 (2019) from the 5 mm-resolution point cloud at a kernel radius (spatial scale) of 4 cm, 5 cm, 10 cm, 20 cm, 30 cm, 40 cm, and 50 cm. Within R version 3.6.1, roughness estimates were resampled with replacement and averaged 1000 times for each plot and spatial scale to remove potential outliers.(Completed in 2019) 3)Desiccation crack cover was quantified using oblique photographs collected with a Nikon D810 DSLR camera mounted on the TLS instrument during survey. We classified visible surface cracking within the photographs through manual interpretation into four ranks based on coverage in each plot; Rank 1: < 10% crack coverage, Rank 2: 10 – 40% crack coverage, Rank 3: 40 – 60% crack coverage, and Rank 4: > 60% crack coverage. Surface coverage was estimated visually for each photograph using a perspective-scaled grid with ten cells overlaid on the image and each photograph was evaluated three times in randomized order to get an average interpreted cover.(Completed in 2019) 4) To estimate vegetation canopy cover, point clouds with classified ground and vegetation points were converted to a 2-D raster. The vegetation cells were then divided by the total number of plot cells to approximate the percentage of vegetative cover. (Completed in 2019) 2019 Data Quality Assessment and Quality Control (QAQC): Data previously reported by Faist and others (2020) were not formally assessed for quality (i.e. Faist Data). Lidar-based roughness estimates were quality assessed in two ways. 1) We assessed the minimum measurable roughness and reproducibility limits of the Reigl VZ 1000 by following ground-based lidar scanning procedures on a 1 m x 2 m smooth, manufactured surface. We survey the manufactured surface in 6 different location then assessed spatial and temporal stationarity. Spatial stationarity was assessed using a t-test measuring differences in estimated roughness between two halves of the same survey. Temporal stationarity was assessed with an ANOVA to identify differences in roughness between all 6 scans of the same surface. We found no significant spatial or temporal differences (P > 0.1). This assessment showed that equipment did not significantly contribute to observed differences in roughness and provided an estimated minimum measurable roughness.(Completed in 2019) 2) A plot in the Utah training and Testing Range (UTTR) and a plot at the Jornada Experimental Range (JER) were surveyed from different positions and sides of the plot. Using a t-test we assessed differences in roughness estimated for these surveys of the same plot. We found no significant difference (P > 0.1) in mean plot roughness, suggesting that processing procedure did not significantly contribute to observed differences in roughness. (Completed in 2019). 2019 Finalize Data for Dissemination: Data sent to the Southwest Biological Science Center Data Steward for dissemination and preservation per USGS Data Management Policies SM 502.6, SM 502.7, SM 502.8 and SM 502.9 (1 October 2016). 2021 Faist Data These data represent experimental restoration treatments on biocrusts and associated field and laboratory data that were first reported by Faist and others (2020). These data were summarized to explain potential differences in soil surface roughness between biocrust communities based on biocrust development and cover. Producer defined Plot Plot unique identifier (letter###). Plot names after Faist and others (2020). Producer defined a textual and numeric description designated for data collection point, to assist in identifying its location Faist Treatment Experimental design soil plot treatment indentifier by Faist and others (2020). CON represents undisturbed biocrust plots (control). DIS, FC, LB, MI, and NO received mechanical disturbance. FC, LB, MI and NO were treated with a physical stablizer. FC, LB, and MI were treated with biocrust inoculum. Producer defined CON Control: Plot were undisturbed biocrust communities. Faist and others (2020) experimental design treatment. Producer defined DIS Disturbed: Plots were mechanically disturbed biocrust that received no restoration treatment. Faist and others (2020) experimental design treatment. Producer defined FC Field Collection: Plots were mechanically disturbed biocrust in which field collected biocrust was added back to the soil surface. Inoculation of biocrust was supplemented by a physical stabilizer used to reduce sediment movement after disturbance. Faist and others (2020) experimental design treatment. Producer defined LB Lab Biocrust: Plots were mechanically disturbed biocrust in which laboratory cultivated biocrust was added to the soil surface. Inoculation of biocrust was supplemented by a physical stabilizer used to reduce sediment movement after disturbance. Faist and others (2020) experimental design treatment. Producer defined MI Mixed Inoculum: Plots were mechanically disturbed biocrust in which greenhouse cultivated biocrust was added to the soil surface. Inoculation of biocrust was supplemented by a physical stabilizer used to reduce sediment movement after disturbance. Faist and others (2020) experimental design treatment. Producer defined NO No Biocrust: Plots were mechanically disturbed biocrust and soil was supplemented by a physical stabilizer used to reduce sediment movement after disturbance. Faist and others (2020) experimental design treatment. Producer defined Phy Stablizer Soil plot stabilization treatment indentifier by Faist and others (2020). A polyacrylamide solution (PM) or a straw border (ST) were applied to selected soil plots after mechanical disturbance to increase soil stability. Producer defined None No physical stabilizer was applied after mechanical disturbance in 2015 or to undisturbed biocrust communities. Producer defined PM A polyacrylamide solution was was applied to aproximately 10% of the biocrust plot after mechanical disturbance in 2015. Producer defined ST A verticle straw border approximately 15 cm high was placed around the exterior of the biocrust plot after mechanical disturbance in 2015. Producer defined AgStb_2016 Mean rank of three estimates aggregate stability (#.##) in 2016. Field kit stability measurements from December 2015. Producer defined 1 6 Index value (1 as the lowest and 6 as the highest aggregate stability) AgStb_2018 Mean rank of three estimates aggregate stability (#.##) in 2018. Field kit stability measurements from June 2018. Aggregate stability was not assessed at JER in 2018. ND = no reported data. Producer defined 1 6 Index value (1 as the lowest and 6 as the highest aggregate stability) CHLA_2016 µg/g of soil Chlorophyll-A in 2016. Data collected in December 2015 and first reported by Faist and others (2020). Producer defined 0.34 52.23 micrograms per gram (µg/g) CHLA_2017 µg/g of soil Chlorophyll-A in 2017. Data collected in 2017 and first reported by Faist and others (2020). Producer defined 0.26 31.26 micrograms per gram (µg/g) LOD_2016 Mean Level of Development index (#.##) in 2016. Data collected in December 2015 and first reported by Faist and others (2020). Producer defined 1 5 Index value (1 as the lowest and 6 as the level-of-development) LOD_2017 Mean Level of Development index (#.##) in 2017. Data collected in 2017 and first reported by Faist and others (2020). Producer defined 1 5.75 Index value (1 as the lowest and 6 as the level-of-development) BSCCov2016 Estimated percentage of plot cover with biocrust in 2016. Data collected in December 2015 and first reported by Faist and others (2020). Producer defined 0 100 Percent of area BSCCov2017 Estimated percentage of plot cover with biocrust in 2017. Data collected in 2017 and first reported by Faist and others (2020). Producer defined 0 100 Percent of area Lidar Derived Data table These data represent the summarized results of ground-based lidar surveys of 121, 1 m x 3 m biocrust soil plots. All numeric data were derived from 5 mm resolution digital point clouds. Producer defined Plot Plot unique identifier. Plot names after Faist and others (2020). Producer defined a textual and numeric description designated for data collection point, to assist in identifying its location Treatment Type of experimental treatment design. Represents mature, undisturbed biocrust and biocrust communities disturbed by scraping and trampling. Producer defined Disturbed Mechanically disturbed biocrust - a portion of disturbed bocrust was treated with a physical stablizer and a subset of those were treated with biocrust inoculum. Producer defined Undisturbed Undisturbed biocrust - representative sample of the local, mature biocrust community that was not treated during the study. Producer defined Inoculum Binary identifier for the addition of biocrust innoculum for experimental restoration. Producer defined 0 No biocrust inoculum was added to mechanically disturbed biocrust. Producer defined 1 Biocrust inoculum was added to mechanically disturbed biocrust. Producer defined Site Location of plot within either the Utah Training and Testing Range (UTTR) or the Jornada Experimental Range (JER). Producer defined JER A loamy aridisol at the Jornada Experimental Range. Site location is at lat 32.55, long -106.72 Producer defined UTTR-1 A silty loam aridisol at the Utah Training and Testing Range. Site location is at lat 41.11, long -113.00 Producer defined UTTR-2 A silty loam entisol at the Utah Training and Testing Range. Site location is at lat 41.12, long -112.95 Producer defined Climate Desert climate identifier; Climate designation is based on seasonal temperatures that often drop below 0 C (Cool) or rarely drop below 0 C (Hot). Producer defined Cool winter temperatures that regularly drop below 0° C - Great Basin Desert sites at the Utah Training and Testing Range Producer defined Hot winter temperatures that rarely drop below 0° C - Chihuahuan Desert site the Jornada Experimental Range Producer defined R4_2016 2016 survey mean roughness in centimeters estimated for a moving 4 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.00028 0.00215 centimeters R5_2016 2016 survey mean roughness in centimeters estimated for a moving 5 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.0003 0.00254 centimeters R10_2016 2016 survey mean roughness in centimeters estimated for a moving 10 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.00042 0.00338 centimeters R20_2016 2016 survey mean roughness in centimeters estimated for a moving 20 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.00079 0.00499 centimeters R30_2016 2016 survey mean roughness in centimeters estimated for a moving 30 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.0009 0.00528 centimeters R40_2016 2016 survey mean roughness in centimeters estimated for a moving 40 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.00084 0.00586 centimeters R50_2016 2016 survey mean roughness in centimeters estimated for a moving 50 cm moving kernel radius. The 2016 survey was in December 2015 for UTTR and February 2016 for JER. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Producer defined 0.00087 0.00869 centimeters R4_2018 2018 survey mean roughness in centimeters estimated for a moving 4 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0.00033 0.00224 centimeters R5_2018 2018 survey mean roughness in centimeters estimated for a moving 5 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0.00037 0.00255 centimeters R10_2018 2018 survey mean roughness in centimeters estimated for a moving 10 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0.00047 0.00366 centimeters R20_2018 2018 survey mean roughness in centimeters estimated for a moving 20 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0.00066 0.00445 centimeters R30_2018 2018 survey mean roughness in centimeters estimated for a moving 30 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = reported data. Producer defined 0.00091 0.00632 centimeters R40_2018 2018 survey mean roughness in centimeters estimated for a moving 40 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0.00115 0.00585 centimeters R50_2018 2018 survey mean roughness in centimeters estimated for a moving 50 cm moving kernel radius. The 2018 survey was only conducted at UTTR. Roughness was calculated in cloudcompare (v 2.1) and averaged by plot within R (v.3.6.1). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0.00159 0.0064 centimeters TopoChg Area of biocrust plot with significant change in elevation between the 2016 and 2018 surveys. The minimum elevation change threshold was 0.005 m (5 mm). Lidar survey was not conducted at JER in 2018. ND = no reported data. Producer defined 0 33.5 Percent of area Veg_2016 Estimated area of vegetatation cover within biocrust plot in December 2015. Vegetation area estimated by the number of raster cells classified as vegetation divided by the total number of raster cells within a plot. Vegetation area was not estimated for JER. ND = no reported data. Producer defined 0 72.46 Percent of area Veg_2018 Estimated area of vegetatation cover within biocrust plot in June 2018. Vegetation area estimated by the number of raster cells classified as vegetation divided by the total number of raster cells within a plot. Vegetation area was not estimated for JER. ND = no reported data. Producer defined 0 88.91 Percent of area Crack_2016 Mean rank of desication crack cover estimated from multiple photographic interpretations for December 2015; 1 represents < 10% plot coverage and 4 represents over 60% plot coverage. Desication crack cover was not estimated for JER or any undisturbed biocrust. ND = no reported data. Producer defined 1 4 Index value Crack_2018 Mean rank of desication crack cover estimated from multiple photographic interpretations for June 2018; 1 represents < 10% plot coverage and 4 represents over 60% plot coverage. Desication crack cover was not estimated for JER or any undisturbed biocrust. ND = no reported data. Producer defined 1 4 Index value U.S. Geological Survey - ScienceBase U.S. Geological Survey mailing and physical

Denver Federal Center, Building 810, Mail Stop 302

Denver CO 80225 United States 1-888-275-8747 sciencebase@usgs.gov The author(s) of these data request that data users contact them regarding intended use and to assist with understanding limitations and interpretation. 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. These data are in a comma-separated values (.csv) file format. The user must have software capable of reading or displaying machine-readable tabular data. 20210913 Joshua J Caster U.S. Geological Survey Student Trainee (Geography) mailing and physical

2255 North Gemini Drive

Flagstaff AZ 86001 US 928-556-7129 jcaster@usgs.gov Content Standard for Digital Geospatial Metadata FGDC-STD-001-1998