The Bureau of Land Management and the National Park Service Remote Sensing Investigation of Groundwater Dependent Ecosystems
in Surprise Canyon Creek Wild and Scenic River Corridor and Surprise Canyon Wilderness, Panamint Valley, California: Geospatial Database

Prepared by Blake Minor, Christine Albano, Guy Smith, and Justin Huntington from the Desert Research Institute, in coordination with the Bureau of Land Management and the National Park Service.

This database contains geospatial, hydrology, climate, and remote sensing data gathered/prepared in 2022 for the Surprise Canyon Creek Wild and Scenic River, Panamint Valley, CA. Federal Geographic Data Committee (FGDC) metadata is provided for each type of data in a separate "metadata" folder (a summary table describing the types of data collected at each site is provided in the metadata folder as well).

There are 8 folders containing various datasets and metadata:

et_output
gde_boundary
groundwater_level_data
ndvi_rasters
site_data
zonal_stats_aois
metadata
SurpriseCanyon_ArcPro

The "et_output" folder contains two subfolders titled "BLM_Mesquite_Bosque" and "BLM_NPS_WSRC". The former contains spatial summaries (BLM_Mesquite_Bosque_Region_ETg_output.xlsx) and rasters of annual and long-term average/median actual ET (ETa) and groundwater ET (ETg) for the mequite bosque and iodine bush polygons (SC_GDE_area_bounds_from_NCCAG_mesquite_iodine_bush_wgs84utmz11.shp) from the NCCAG GDE boundary in the valley floor of Surprise Canyon Wilderness. The rasters are provided in units of millimeters. In the BLM_Mesquite_Bosque_Region_ETg_output.xlsx file, the "Summary" tab highlights results of total ETa and ETg rates/volumes with units shown (feet per year and/or millimeters per year) and total area in acres and square meters. The "raw_output" tab contains spatial summaries for individual Landsat scenes with the following variables/units:

ZONE_NAME - Unique name for the polygon/region summarized
ZONE_FID - Unique feature ID
DATE - Date of Landsat Overpass
SCENE_ID - Landsat scene ID
PLATFORM - Landsat platform
PATH - Landsat path
ROW - Landsat row
YEAR - Year
MONTH - Month
DAY - Day of month
DOY - Day of year
PIXEL_COUNT - Number of pixels used to calculate zonal statistics
PIXEL_TOTAL - Number of pixels that could theoretically be used for zonal statistics based on intersection with polygon
FMASK_COUNT - Number of pixels masked
FMASK_TOTAL - Number of pixels that could theoretically be masked
ETSTAR_COUNT - Number of pixels meeting the minimum ET* threshold of the NDVI - ET* regression
CLOUD_SCORE - Simple cloud score as a percentage
QA - Quality assessment flag
NDVI_SUR - Average normalized difference vegetation index surface reflectance value (dimensionless)
NDVI_TOA - Average normalized difference vegetation index top-of-atmosphere value (dimensionless)
NDWI_TOA - Average normalized difference water index top-of-atmosphere value (dimensionless)
SAVI_SUR - Average soil-adjusted vegetation index surface reflectance value (dimensionless)
ALBEDO_SUR - Average albedo value (dimensionless)
TS - Average surface temperature value in kelvin
EVI_SUR - Average enhanced vegetation index value (dimensionless)
ETSTAR_MEAN - Average ET* value (dimensionless) predicted following Minor (2019)
ETG_MEAN - Average groundwater ET value in millimeters
ETG_LPI - Average groundwater ET lower 90% prediction interval value in millimeters
ETG_UPI - Average groundwater ET upper 90% prediction interval value in millimeters
ETG_LCI - Average groundwater ET lower 90% confidence interval value in millimeters
ETG_UCI - Average groundwater ET upper 90% confidence interval value in millimeters
ET_MEAN - Average actual ET value in millimeters
ET_LPI - Average actual ET lower 90% prediction interval value in millimeters
ET_UPI - Average actual ET upper 90% prediction interval value in millimeters
ET_LCI - Average actual ET lower 90% confidence interval value in millimeters
ET_UCI - Average actual ET upper 90% confidence interval value in millimeters
WY_ETO - Average water year total gridMET grass reference ET value in millimeters
WY_PPT - Average water year total gridMET precipitation value in millimeters

The latter file contains spatial summaries (BLM_NPS_WSRC_ETg_output.xlsx, no supplementary rasters) of annual ETa and ETg for all of the AOIs (BLM and NPS sites used in the zonal stats calculations described later on) within the Surprise Canyon Creek Wild and Scenic River Corridor from 1985-2021. The same variables described above are provided in the BLM_NPS_WSRC_ETg_output.xlsx file for the various AOIs analyzed (e.g., "Brewkiln_Reach" tab). ETa and ETg were estimated following the methodologies from Minor (2019).

The "gde_boundary" folder contains a shapefile delineating the area of potential groundwater discharge from phreatophyte vegetation (e.g. SC_GDE_area_bounds_from_NCCAG.shp), which are derived from the Natural Communities Commonly Associated with Groundwater (NCCAG) gde boundary with modifications to include the AOIs within the Surprise Canyon Creek Wild and Scenic River Corridor.

https://data.amerigeoss.org/dataset/i02-nccag-vegetation-0d90b

The "groundwater_level_data" folder contains the following files: 1) one shapefiles of well locations (e.g., NDWR_CDWR_and_USGS_Combined_wells.shp) 2) a symbology layer, which is used to illustrate the depth to groundwater (NDWR_CDWR_and_USGS_Combined_wells_WGS84UTM11N.lyr) in units of feet below ground surface, and 3) two excel files.

The first excel file (Combined_Site_Data.csv) contains site information about the wells. Site information includes the site name (well name), site ID, owner, well depth in feet, land surface elevation in feet above mean sea level (asl), basin number, agency, permit number, well log number, perforations, latitude in decimal degrees, and the longitude in decimal degrees.

The second excel file (Combined_WL_Data.csv) contains water level measurement information from each well. Columns include the site name (SITE_ID), measurement dates, and water level elevations in feet above mean sea level.

The "ndvi_rasters" folder contains 2 trend raster grids and 2 corresponding p-value grids. One of the trend rasters represents the long-term trends (1985-2021) in the annual July 15 - Sept 15 median Landsat Collection 2-derived normalized difference vegetation index (NDVI, dimensionless) based on the Theil-Sen slope estimator (ndvi_slope_desc_SurpriseCanyonHUC_85_21.tif). The other trend raster represents the long-term trends(1985-2021) in Landsat derived NDVI after accounting for potential water surplus/deficit using the Adjusted Kendall approach (ndviresid_slope_desc_SurpriseCanyonHUC_85_21.tif) (Alley 1988; see Albano et al. 2021 for details). Each of the NDVI trend rasters has a p-value raster associated with their trends (e.g. ndvi_pvalue_desc_SurpriseCanyonHUC_85_21.tif), which represents the Mann-Kendall p-value significance of their respective Theil-Sen slope. A Pearson's Correlation Coefficient raster (SurpriseCanyonHUC_pearsons_cor_NDVI-WS_85_21.tif) and corresponding p-value significance raster (SurpriseCanyonHUC_pearsons_pval_NDVI-WS_85_21.tif) image are also provided and are the results of linearly regressing NDVI with the water surplus (i.e., gridMET precipitation minus the gridMET potential evapotranspiration (ETo)). A raster of the 2012-2021 mean late-summer NDVI is also provided (mean_late_summer_NDVI_12-21.tif).

References for the statistical analysis of trends can be found here:

Alley, W. M. (1988). Using exogenous variables in testing for monotonic trends in hydrologic time series. Water Resources Research, 24(11), 1955–1961. https://doi.org/10.1029/WR024i011p01955

Kendall, M. G. (1975). Rank Correlation Methods. Griffin, London, UK.

Mann, H. B. (1945). Nonparametric Tests Against Trend. Econometrica, 13(3), 245–259. https://doi.org/10.2307/1907187

Sen, P. K. (1968). Estimates of the Regression Coefficient Based on Kendall’s Tau. Journal of the American Statistical Association, 63(324), 1379–1389. https://doi.org/10.1080/01621459.1968.10480934

https://docs.scipy.org/doc/scipy-0.15.1/reference/generated/scipy.stats.mstats.theilslopes.html

https://docs.scipy.org/doc/scipy-0.15.1/reference/generated/scipy.stats.kendalltau.html

For the NDVI trend maps, the red colors indicate areas where NDVI trends are decreasing, whereas blue colors indicate areas where NDVI trends are increasing. Yellow/tan colors indicate areas with little to no NDVI trends.

For the p-value maps, black indicates that the NDVI trend is significant at the 0.1 level, and white indicates that the NDVI trend is not significant at the 0.1 level.

The "site_data" folder contains 2 sub-folders:

field_survey_data
uas_data

Within the "field_survey_data" subfolder are ground_photograph_locations (e.g. SurpriseCanyon_photograph_locations.shp), corresponding ground_photographs (feature attributes in
locations shapefile match file names for photographs) taken at the field transects, and corresponding observation forms (scanned).

Within the "uas_data" subfolder are high-resolution (~3cm pixel resolution) true color orthomosaics and csv's containing information about the drone flights.

The "zonal_stats_aois" sub-folder contains aoi sub-folders, each containing three excel files and numerous figures summarizing zonal statistics computed for the AOIs. Along with the statistics folders there is a shapefile (e.g. ZonalStats_AOIs_WGS84UTM11N.shp) that was used for calculating zonal statistics.

The first excel file (site name followed by _gridmet_daily.csv) contains spatially averaged daily summaries of gridMET variables for AOIs: Precipitation (PPT) in millimeters, Penman-Monteith Grass Reference Evapotranspiration (ETO) in millimeters, Minimum Temperature (TMIN) in kelvin, Maximum Temperature (TMAX) in kelvin, and Mean Temperature (TMEAN) in kelvin.

The second excel file (site name followed by _gridmet_monthly.csv) contains spatially averaged monthly summaries of the gridMET variables for AOIs as above in the same units of either millimeters or kelvin.

The third excel file (site name followed by _landsat_daily.csv) contains spatially averaged Landsat derived variables. The columns in this file include the site name (ZONE_NAME), unique feature ID (ZONE_FID), image acquisition date (DATE), unique Landsat scene ID (SCENE_ID), satellite platform (PLATFORM), satellite path (PATH), satellite row (ROW), year (YEAR), month (MONTH), day of month (DAY), day of year (DOY), area of input feature (AREA) in units of acres, pixel size (PIXEL_SIZE) in meters, the total number of pixels that were used in the computation (PIXEL_COUNT), the total number of pixel that could theoretically be used in the computation (PIXEL_TOTAL), the amount of masked pixels (FMASK_COUNT), the total number of pixels that could theoretically be masked (FMASK_TOTAL), the percentage of masked pixels (FMASK_PCT), a simple cloud score value (CLOUD_SCORE, dimensionless), the QA number if applied (QA), surface albedo (ALBEDO_SUR, dimensionless), the enhanced vegetation index (EVI_SUR, dimensionless), the normalized difference vegetation index (NDVI_SUR, dimensionless), the land surface temperature (TS) in kelvin, the modified soil-adjusted vegetation index (MSAVI_SUR, dimensionless), the normalized difference water index using the green and near-infrared bands (NDWI_GREEN_NIR_SUR, dimensionless), the normalized difference water index using the green and shortwave-infrared bands (NDWI_GREEN_SWIR1_SUR, dimensionless), the normalized difference water index using the shortwave-infrared and green bands (NDWI_SWIR1_GREEN_SUR, dimensionless), the normalized difference water index using the near-infrared and shortwave-infrared bands (NDWI_NIR_SWIR1_SUR, unitless), the soil-adjusted vegetation index (SAVI_SUR, dimensionless), the blue wavelength band (BLUE_SUR, unitless), the green wavelength band(GREEN_SUR, dimensionless), the red wavelength band (RED_SUR, unitless), the near-infrared wavelength band (NIR_SUR, dimensionless), the 1st short-wave infrared wavelength band (SWIR1_SUR, dimensionless), and the 2nd short-wave infrared wavelength band (SWIR2_SUR, dimensionless), and an outlier score (OUTLIER_SCORE, dimensionless).

The "figures" folder within the "zonal_stats_aois" sub-folder contains summary plots of variables analyzed with zonal statistics. A list of variables are as follows:

EVI_SUR - Enhanced Vegetation Index surface reflectance (dimensionless)
ETO - ASCE Grass Reference Evapotranspiration (millimeters)
PPT - Precipitation (millimeters)
NDVI_SUR - Normalized Difference Vegetation Index surface reflectance (dimensionless)
NDVI_TOA - Normalized Difference Vegetation Index top-of-atmosphere (dimensionless)
Albedo - Surface Albedo (dimensionless)
NDWI - Normalized Difference Water Index (with wavelength bands used to compute the normalized difference: green = green, red = red, nir = near-infrared, swir1 = short-wave infrared) (dimensionless)
TS - Land Surface Temperature (kelvin)

Along with these summary figures are two comma separated files (site_name_gridmet_figures.csv, and site_name_landsat_figures.csv) used to generate the figures. All datasets are derived from either the gridded spatial data of GridMET or Landsat. Note: The units used for variables shown in the plots may differ than the corresponding CSV files provided (e.g., PPT, millimeters in tables vs. feet in figures).

The "metadata" sub-folder contains individual Federal Geographic Data Committee (FGDC) standard metadata files for the various datasets described above. Additionally, there is an excel file (summary_table.xlsx) that has more details about the AOI's visited and what type of sUAS data was collected at each of them.

The "SurpriseCanyon_ArcPro" sub-folder contains the ArcGIS Pro Project file (SurpriseCanyon_ArcPro.aprx) that can be opened to visualize the datasets described above. Additionally, an open-source QGIS Project (SurpriseCanyon_QGIS.qgz) is provided within the main folder in case the user is not licensed to use ESRI's ArcGIS Pro software.