ABSTRACT:

This dataset includes tipping bucket rain gauge and ephemeral stream stage data from Mohave and Yuma Wash

Locations of each measurement are included in locations.txt

Background: Details about the monitoring sites are available in Faulconer (2015). Thresholds for runoff generation in ephemeral streams with varying morphology in the Sonoran Desert in Arizona, USA. MS Thesis, Colorado State University https://mountainscholar.org/handle/10217/166929

Sensors: Rain gauges were TE525 and TB4 tipping buckets attached to Campbell Scientific loggers and RG3-M tipping buckets with Onset HOBO pendant event loggers
Stage sensors were In-Situ Inc. Rugged Troll 100 unvented pressure transducers.

Stage data included here were adjusted for barometric pressure and partially offset adjusted to bring no flow stage closer to zero at the beginning of the time series. Most values in each time series are
sensor noise, which is +/- 2 cm. Stream flow is only when stage rises >2cm above background stage. Background stage values can drift, so absolute stage should not be compared between sites or between events.

Note, sensors may have periods of missing data due to sensor failures.

ABSTRACT:

This resource includes two datasets for High Park Fire sediment yield and sediment load from June-October 2013.

The first file, HPF_hillslope_observed.csv contains hillslope sediment yield data that were originally reported in Schmeer et al. (2018).

The column values are:
"Site" = site ID
Area_ha = field-delineated hillslope area
SY_Mg_ha = measured hillslope sediment yield in Mg per ha
P_mm = total precipitation depth measured at the nearest rain gauge

The second file, HPF_watershed_simulated, contains simulated total watershed sediment loads for multiple models discretized at varying target hillslope resolutions.

The column values are:
"area_ha" = target area for the hillslopes
"watershed" = name of the watershed simulated, either Hill Gulch or Skin Gulch
"model" = name of the model used,
"Mg" = watershed total sediment load in Mg
"fraction" = fraction increase in watershed total sediment load relative to the sediment load simulated for 0.5 ha hillslope resolution

ABSTRACT:

This dataset includes geospatial data for three Colorado Front Range catchments: Mill Creek, Skin Gulch, and Gordon Gulch. In each catchment the extent of surface flow in the channel network was mapped twice in the field during summer 2016. These field maps of flow presence are included as polyline shapefiles in the dataset. Additional supporting data for each catchment are included for further analysis. These supporting data are watershed boundaries, flowlines from the National Hydrography Dataset High Resolution, digitized geologic maps, and 1 m resolution digital elevation models.

Flow presence maps: Polyline shapefiles named as Catchment_flow_mmddyy. Original source: Martin (2018)

Watershed boundaries: Polygon shapefiles named as Catchment_boundary. Original source: Martin (2018)

National Hydrography Dataset High Resolution: Polyline shapefiles named as Catchment_NHD. Original source USGS (2018)

Digitzed geologic maps: Polygon shapefiles of lithology named as Catchment_geology and polyline shapefiles of faults named as Catchment_structures. Original source: Abbott (1976) and Braddock et al. (1988) for Skin Gulch; Braddock et al. (1989) for Mill Creek; Gable (1980) for Gordon Gulch

Digital elevation models: .tif files named as Catchment_dem with units in meters. Original source: National Ecological Observatory Network (NEON) 2013 1 m LiDAR

ABSTRACT:

UPDATED August 2021.

This dataset contains hydrologic data from small research catchments organized along an elevation gradient in the Colorado Front Range. Catchments are located in either persistent or intermittent snow zones. Catchments are Michigan River (persistent), Andrews Creek (persistent), Dry Creek (intermittent), Bighorn Creek (intermittent), and Mill Creek (intermittent). These sites each have stream stage/discharge, rain, soil moisture, and temperature data. Data available for water years 2016-2019 at daily time steps. Some sensors have shorter time step data available on request. NA indicates missing data or data screened out for errors.

The following files are available:

-Geospatial: frontrange_locations.zip contains a shapefile (frontrange_locations.shp) and a kmz file (frontrange_locations.kmz) giving the sensor locations for each catchment. Catchments each have two points, one for stream gage locations and the other for rain gage, snow depth, temperature, and soil moisture monitoring locations.

-Metadata: description of sensors for each site

-Rating: stage-discharge data for rating curves

-Daily data by site: see Metadata for more information

ABSTRACT:

This resource provides the data and R code needed to apply the mean annual and mean monthly streamflow prediction models for Colorado presented in Eurich (2020). A web version of the shiny app for applying the models is available at https://cuahsi.shinyapps.io/CO_streamflow/.

The resource also includes attributes and streamflow values for watersheds throughout Colorado that were used to develop the models. All datasets are averages for water years 2001-2018.

Watershed attribute data to apply the models are contained in the Data folder within the CO_streamflow.zip file, with the exception of elevation, slope and aspect. Due to file size limitations, these are provided in separate compressed files. After download they should be uncompressed and moved into the Data folder. Attributes include the following spatial data files for the state of Colorado:

-Mean annual precipitation: File name p.tif. Units mm. Source PRISM Climate Group (2004).
-Mean annual potential evapotranspiration. File name pet.tif. Units mm. Source gridMET grass reference evapotranspiration (Abatzoglou 2013).
-Mean annual snow persistence. File name sp.tif. Units %. Source Hammond et al. (2017).
-Elevation: File name elevation.tif. Units m. Source 30-m digital elevation model from National Elevation Dataset.
-Aspect: Eight files for aspect bins N, NE, E, SE, S, SW, W, NW. File names aspect_bin.tif, each contained in subfolder Data/Aspect. Derived from elevation.tif.
-Slope: File name slope.tif. Units %. Derived from elevation.tif.
-Geology: File name Geology.shp. Source USDA-USGS Mineral Resources Program (2005).
-Hydrologic Region: File name HydroRegion.shp. Source Capesius and Stephens (2009).

Code to apply the models is supplied in the main folder, both as an R script (CO_Q.r) and as a Shiny app (CO_Qapp.r). Description of the code structure and source packages is available in Eurich et al. (submitted).

Streamflow data from the gauged streams are provided in the Data folder as streamflow_observed.csv. This file includes both watershed attributes derived from the same source files listed above as well as observed streamflow. The file has the following columns:

-gageID: this provides the Colorado Division of Water Resources (CDWR) gage abbreviation followed by the US Geological Survey (USGS) gage number. If no CDWR gage abbreviation is available, the value is only the USGS gage number.
-statname: this provides the full station name for the station.
-Type: this indicates whether the stream is natural or has within basin modifications. Natural streams have no documented reservoirs or flow diversions and less than 10% urban land cover. Within basin modification streams have <10% urban land cover and no transbasin diversions, but they do have reservoirs and/or diversions within the watershed.
-Area_sq.km: watershed drainage area in square kilometers
-Elev_m: watershed mean elevation in meters
-Slope_%: watershed mean slope in percent
-Dom_Aspect: aspect bin (N, NE, E, SE, S, SW, W, NW) covering the largest area in the watershed
-Geo_group: geologic group (permeable sedimentary, impermeable sedimentary, volcanic, intrusive, permeable metamorphic, impermeable metamorphic, modern alluvium/colluvium, glacial) covering the largest area in the watershed
-Hyd_reg: hydrologic region (Mountain, Northwest, Plains, Rio Grande, Southwest) of the watershed outlet
-SP: mean annual snow persistence for the watershed in %
-P_mm: mean annual precipitation for the watershed in mm
-PET_mm: mean annual potential evapotranspiration for the watershed in mm
-Qann_mm: mean annual observed streamflow in mm
-Qjan_mm: mean January observed streamflow in mm
-Qfeb_mm: mean February observed streamflow in mm
-Qmar_mm: mean March observed streamflow in mm
-Qapr_mm: mean April observed streamflow in mm
-Qmay_mm: mean May observed streamflow in mm
-Qjun_mm: mean June observed streamflow in mm
-Qjul_mm: mean July observed streamflow in mm
-Qaug_mm: mean August observed streamflow in mm
-Qsep_mm: mean September observed streamflow in mm
-Qoct_mm: mean October observed streamflow in mm
-Qnov_mm: mean November observed streamflow in mm
-Qdec_mm: mean December observed streamflow in mm
-Q1: 1st percentile non-exceedance flow in mm
-Q99: 99th percentile non-exceedance flow in mm
-CV_low: coefficient of variation for the annual low flow, unitless
-CV_high: coefficient of variation for the annual high flow, unitless

ABSTRACT:

This resource provides snow depth data from sensors and both snow depth and snow water equivalent data from snow surveys in 2021. Measurements were collected at burned and unburned locations within and near the Cameron Peak Fire in northern Colorado, USA.

ABSTRACT:

This resource contains water year 2021 data for rainfall, weather station, and streamflow monitoring stations within and near the 2020 Cameron Peak Fire in Northern Colorado.

Site names are in the data_dictionary.csv file
Site locations and sensor information are in the metadata.csv file

Daily summary data files are:

P_daily.csv: This contains columns for each site with a tipping bucket rain gauge, with values in mm. NA indicate gauge was not recording. These values assume 0.3 mm of precipitation per tip at all sites except Joe Wright, Tunnel, and Bighorn, where the value is 0.1 mm per tip. The values have not been adjusted for snow precipitation, so some tips may represent snowmelt into the gauge rather than rainfall.

Weather station files: these are separate files for each station, indicated by stationname_met_daily.csv. The data dictionary indicates the type of measurement and units for each column header. The only QA/QC conducted on these data is deletion of data representing sensor errors. NA indicates no measurement.

stream_daily.csv: contains columns for stage (cm) and discharge computed from a site-specific rating curve. Discharge values are normalized by catchment area and given in mm. All stream stage time series have been offset-adjusted to correct for differences in sensor placement between download dates. Dry and Washout also required offset adjustment for changes in bed elevation during post-fire storms. These offset adjustments are estimated because we have no means of knowing the exact bed elevation changes during storms.

discharge.csv: contains the field measurements of stream discharge and corresponding stream stage. Type indicates the discharge measurement method used, either salt dilution (salt), manual with a velocity meter, or manual with a bucket. These values were used to develop the rating curves that were used to compute discharge. These rating curves are a work in progress.

ABSTRACT:

This resource contains the data used in Peterson et al. "Predicting streamflow from crowd-sourced flow observations".

sensor_daily.csv contains the stage data from sensors presented as:
-stage = offset-adjusted and quality-checked sensor stage data
-norm = stage divided by the maximum stage recorded at the sensor
-gapfill = norm time series with short data gaps filled with linear interpolation

The Monthly Snow Persistence folder contains mean snow persistence layers for January and April. The average values are derived from MODIS satellite data from 2016-2020.

streamtracker.csv contains the Stream Tracker flow presence/absence data for April-October. Columns are Stream Tracker sites. Empty cells indicate no data; 0 indicates no flow or standing water, and 1 indicates flow.

attributes.csv contains the watershed attributes for each sensor and Stream Tracker site. Columns are:
-site = ID for sensor or Stream Tracker site
-type = observation type - sensor or Stream Tracker (st)
-elevation = elevation group; high for watersheds in the persistent snow zone and low for all others
-Nobs = number of observations at site; sensor sites set at 100 for graphing purposes
-fracFlow = calculated flow fraction from observations, (number of flow observations)/(total numer of observations)
-ma_p = mean annual precipitation (mm), PRISM
-mmp_may = mean may precipitation (mm), PRISM
-p_summer = mean summer (june, july, august) precipitation (mm)
-mmsp_jan = mean january snow persistence (Hammond et al. 2020)
-mmsp_apr =mean april snow persistence (Hammond et al. 2020)
-AI_summer = mean summer aridity index (p_summer / PET_summer) where PET is potential evapotranspiration from gridMET reference evapotranspiration
-area_km2 = drainage area in square km
-slope_local = slope of the stream at the monitoring point (m/m)
-aws = available water storage (mm), gSSURGO
-metamorphi = fraction area with metamorphic bedrock (Horton et al. 2017)
-ksat = saturated hydraulic conductivity (um/s), gSSURGO
-pct_clay = average % clay in soil, gSSURGO
-HSG_C = fraction area with hydrologic soil group C, gSSURGO
-LF_herb = fraction area with herbaceous land cover, LANDFIRE
-burn_10 = fraction area within a wildfire boundary from the last 10 years, MTBS

predicted_flow_fraction is a raster TIFF file with 30 m resolution displaying our best performing (≥10 sample) random forest model flow fraction predictions for the entire study area. Values are continuous from 0-1, where 1 represents 100% predicted mean annual flow. Cells draining the persistent snow zone (≥75% mean annual snow persistence) were excluded from our model andgiven values of -1.

Watershed Attributes Folder contains raster data for the attributes used as inputs in the random forest models. These rasters cover the entire study area, and all except slope_local were processed as weighted flow accumulations based on the NHD med-res flow accumulation. This means each cell represents the average value for that attribute in the drainage area contributing to that cell. This folder includes:
-AI_summer: mean summer (June-Sept) aridity index
-area_km2: drainage area in square kilometers
-aws: mean available water storage
-burn_10: percent burn area (composite of burn areas in 10 years prior to 2020)
-HSG_C: percent hydrologic soil group C
-ksat: mean hydraulic conductivity
-LF_herb: percent herbaceous land cover
-ma_p_2001-2020: mean annual precipitation (period 2001-2020)
-ma_sp_2001-2020: mean annual snow persistence (period 2001-2020)
-metamorphic: percent metamorphic bedrock
-mmp_may: mean May precipitation
-mmsp_apr: mean April snow persistence
-mmsp_jan: mean January snow persistence
-p_summer: mean summer (June-Sept) precipitation
-pct_clay: percent clay in soils
-slope_drainage: mean slope for the drainage area
-slope_local: local slope for each pixel

sensor_st_pts is a zip file containing a shapefile of sensor and Stream Tracker site locations. The column 'site' corresponds to the site IDs, and type indicates whether the point is a sensor or Stream Tracker (st) location

ABSTRACT:

This dataset contains 2001-2020 burned areas and climate variables for three regions with Mediterranean climates: South America from 31-46 degrees South, including Chile and the forested Andean region of Argentina; the western United States from 33-49 degrees North from the coast extending to the eastern extent of forest, and the Iberian Peninsula, including all of Spain and Portugal.

Burned areas are polygon shapefiles for all regions except Chile, for which the burn area is represented in a point shapefile. The data sources for the fire shapefiles are:
Chile: unpublished, originally from Corporación Nacional Forestal (CONAF) and compiled by Miranda
Argentina: unpublished, compiled by Diego Mohr-Bell and others at Centro de Investigación y Extensión Forestal Andino Patagónico (CIEFAP)
North America: NIFC 2023
Iberian Peninsula: EFFIS 2022

All of the fire shapefiles are contained within the zip folder fire_areas, and the individual regions are ch_fire (Chile), ar_fire (Argentina), na_fire (North America), ib_fire (Iberian Peninsula). The attributes of the shapefiles are the year and the fire area in square kilometers. For Chile, the fire start dates were documented. If the fire started in June-December, the year assigned is advanced by 1 from the original year. This is because the summer fire season straddles the calendar year boundary, and the fire year is assigned based on the year with most of the summer season. For Argentina, the end dates of the fire were available, so these end dates were used to assign the fire year.

Annual summaries of fire area and climate variables are provided in the fire_ann_all.csv file. The columns in this file are:
year
wetdryzone: dry if mean annual aridity index <1; wet if mean annual aridity index >1
cont: location, either Iberian Peninsula, North America, or South America
area_km2: total burned area in square km
AIann: annual aridity index calculated as total precipitation over total potential evapotranspiration
AIjas: summer aridity index for July-September in northern hemisphere; January-March, southern hemisphere
vpd: mean annual vapor pressure deficit (kPa)
vpd_jas: mean summer vapor pressure deficit (kPa)
def: mean annual climatic water deficit (mm)
def_jas: summer climatic water deficit (mm)

Climate data were obtained from TerraClimate (Abatzoglou et al. 2018).

Mean annual summaries of fire and climate data by aridity index zone are provided in the file AI_bins_meanannual.csv. AI zones/bins are in increments of 0.2. Columns are:
cont: location, either Iberian Peninsula, North America, or South America
AI_max: maximum AI for the AI zone. For example, if the value is 0.2, the zone is from AI=0 to AI=0.2
zone_area: total area of the climate zone in square kilometers
forest_area: total area forested in 2000 in square kilometers (Potapov et al. 2022)
fire_area: total area burned from 2001-2020 in square kilometers (same sources as fire shapefiles)
frac_forest: fraction of climate zone area that is forested
frac_fire: fraction of the climate zone area that was burned
mean_patch_area: mean size of forest patches identified within each AI zone in square kilometers
fwi: mean fire weather index from the European Center for Medium-range Weather Forecasts (Vitolo et al. 2020)

Original data sources:
Abatzoglou, J.T., Dobrowski, S.Z., Parks, S.A., Hegewisch, K.C. (2018), Terraclimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958-2015. Scientific Data, 170191.

European Forest Fire Information System EFFIS (2022). Burnt area mapped using Sentinel-2/MODIS images. Accessed September, 2022.

NIFC 2023. Interagency Fire Perimeter History https://data-nifc.opendata.arcgis.com/search?tags=Category%2Chistoric_wildlandfire_opendata, downloaded 3/25/23.

Potapov P., Hansen M.C., Pickens A., Hernandez-Serna A., Tyukavina A., Turubanova S., Zalles V., Li X., Khan A., Stolle F., Harris N., Song X.-P., Baggett A., Kommareddy I., Kommareddy A. (2022) The global 2000-2020 land cover and land use change dataset derived from the Landsat archive: first results. Frontiers in Remote Sensing, 3.

Vitolo, C., Di Giuseppe, F., Barnard, C., Coughlan, R., San-Miguel-Ayanz, J., Libertá, G., & Krzeminski, B. (2020). ERA5-based global meteorological wildfire danger maps. Scientific data, 7(1), 1-11.