Overview:
The Copernicus DEM is a Digital Surface Model (DSM) which represents the surface of the Earth including buildings, infrastructure and vegetation. The original GLO-30 provides worldwide coverage at 30 meters (refers to 10 arc seconds). Note that ocean areas do not have tiles, there one can assume height values equal to zero. Data is provided as Cloud Optimized GeoTIFFs. Note that the vertical unit for measurement of elevation height is meters.

The Copernicus DEM for Europe at 3 arcsec (0:00:03 = 0.00083333333 ~ 90 meter) in COG format has been derived from the Copernicus DEM GLO-30, mirrored on Open Data on AWS, dataset managed by Sinergise (https://registry.opendata.aws/copernicus-dem/).

Processing steps:
The original Copernicus GLO-30 DEM contains a relevant percentage of tiles with non-square pixels. We created a mosaic map in VRT format and defined within the VRT file the rule to apply cubic resampling while reading the data, i.e. importing them into GRASS GIS for further processing. We chose cubic instead of bilinear resampling since the height-width ratio of non-square pixels is up to 1:5. Hence, artefacts between adjacent tiles in rugged terrain could be minimized:

gdalbuildvrt -input_file_list list_geotiffs_MOOD.csv -r cubic -tr 0.000277777777777778 0.000277777777777778 Copernicus_DSM_30m_MOOD.vrt

In order to reduce the spatial resolution to 3 arc seconds, weighted resampling was performed in GRASS GIS (using r.resamp.stats -w and the pixel values were scaled with 1000 (storing the pixels as integer values) for data volume reduction. In addition, a hillshade raster map was derived from the resampled elevation map (using r.relief, GRASS GIS). Eventually, we exported the elevation and hillshade raster maps in Cloud Optimized GeoTIFF (COG) format, along with SLD and QML style files.

Projection + EPSG code:
Latitude-Longitude/WGS84 (EPSG: 4326)

Spatial extent:
north: 82:00:30N
south: 18N
west: 32:00:30W
east: 70E

Spatial resolution:
3 arc seconds (approx. 90 m)

Pixel values:
meters * 1000 (scaled to Integer; example: value 23220 = 23.220 m a.s.l.)

Software used:
GDAL 3.2.2 and GRASS GIS 8.0.0 (r.resamp.stats -w; r.relief)

Original dataset license:
https://spacedata.copernicus.eu/documents/20126/0/CSCDA_ESA_Mission-specific+Annex.pdf

Processed by:
mundialis GmbH & Co. KG, Germany (https://www.mundialis.de/)

Here we provide a mosaic of the Copernicus DEM 30m for Europe and the corresponding hillshade derived from the GLO-30 public instance of the Copernicus DEM. The CRS is the same as the original Copernicus DEM CRS: EPSG:4326. Note that GLO-30 Public provides limited coverage at 30 meters because a small subset of tiles covering specific countries are not yet released to the public by the Copernicus Programme. Note that ocean areas do not have tiles, there one can assume height values equal to zero. Data is provided as Cloud Optimized GeoTIFFs. The Copernicus DEM is a Digital Surface Model (DSM) which represents the surface of the Earth including buildings, infrastructure and vegetation. The original GLO-30 provides worldwide coverage at 30 meters (refers to 10 arc seconds). Note that ocean areas do not have tiles, there one can assume height values equal to zero. Data is provided as Cloud Optimized GeoTIFFs. Note that the vertical unit for measurement of elevation height is meters. The Copernicus DEM for Europe at 30 m in COG format has been derived from the Copernicus DEM GLO-30, mirrored on Open Data on AWS, dataset managed by Sinergise (https://registry.opendata.aws/copernicus-dem/). Processing steps: The original Copernicus GLO-30 DEM contains a relevant percentage of tiles with non-square pixels. We created a mosaic map in https://gdal.org/drivers/raster/vrt.html format and defined within the VRT file the rule to apply cubic resampling while reading the data, i.e. importing them into GRASS GIS for further processing. We chose cubic instead of bilinear resampling since the height-width ratio of non-square pixels is up to 1:5. Hence, artefacts between adjacent tiles in rugged terrain could be minimized: gdalbuildvrt -input_file_list list_geotiffs_MOOD.csv -r cubic -tr 0.000277777777777778 0.000277777777777778 Copernicus_DSM_30m_MOOD.vrt The pixel values were scaled with 1000 (storing the pixels as integer values) for data volume reduction. In addition, a hillshade raster map was derived from the resampled elevation map (using r.relief, GRASS GIS). Eventually, we exported the elevation and hillshade raster maps in Cloud Optimized GeoTIFF (COG) format, along with SLD and QML style files.

Digital Elevation Models (DEMs) for the Murray-Darling Basin at 1 arc second, 25 metre and 5 metre resolution. Elevation for the whole MDB sourced from LIDAR where available at June 2021 and backfilled with hydrologically enforced 1 second SRTM. Developed as part of the Murray-Darling Water and Environment Research Program and Murray-Darling Ecosystems Function Project. Lineage:

LIDAR Lidar data was source from existing CSIRO past project LIDAR holdings as at June 2021. These holdings comprised of datasets of varying projections resolutions (mostly 5m).

LIDAR that was not already at 5m resolution were transformed to 5m. 1m LIDAR was aggregated to 5m (ArcGIS->SpatialAnalyst->Generalization->Aggregate, 5x5cell, MEAN) and 2m LIDAR was resampled to 5m (ArcGIS->DataManagement->RasterProcessing->Resample, bilinear). Once all input layers existed at 5m resolution they were merged (ArcGIS->DataManagement->RasterDataset->Mosiac) into a single target 5m raster in the GA Albers GCS (EPSG 3577).

Overlapping values of input layers were resolved by taking the MINIMUM of the overlapping cell values. Minimum was used to ensure that LIDAR images with actual in-channel elevations (due to being water free at the time flown) or those which had bathymetry enforced, were not overridden by hydro-flatttened in-channel values from an overlapping image. Visual inspection of the result found there to be little edge effect, with any visible “seams” between LIDAR edges deemed to be insignificant given the subsequent aggregation of the data to 25m that would occur for blending with the SRTM. The available LIDAR provided at continuous high resolution coverage along the length of the basin's major rivers save for a gap along the Namoi River between Wee Waa and Narrabri. This was augented with 5m photogrammetry elevation data.

Further details are given in: Teng J, Penton D, Marvanek S, Mateo C, Khanam F, Ticehurst C and Vaze J (2022) Description and metadata for a composite dataset used for development and validation of a predictive flood inundation and volume model. CSIRO, Australia.

(Link to Metadata) This dataset is derived from the multi-resolution National Elevation Dataset (NED), at resolutions of both 1/3 arc-second (approx. 10 meters) and in limited areas, 1/9 arc-second (approx. 3 meters). Contours derived from this data may appear to have ?shifted? when compared to the 7.5 minute USGS Quad Maps, i.e., Digital Raster Graphics, for a variety of reasons: 1) The NED is a multi-resolution dataset, e.g., areas with LiDAR source data have superseded the original "quad" contours; 2) A result of the original contour vectors undergoing a NAD27 to NAD83 conversion, then the contour vector-to-raster resampling that produced the initial grid, followed by a resampling of that initial 10m grid to the 1/3 arc-second NED (~10.29m) and finally the raster-to-vector conversion yielding the current contours. VCGI extracted the Vermont portion of the NED and re-projected into Vermont State Plane Meters NAD83 (vertical units in feet). Production artifacts were filtered out of this source data prior to acquisition resulting in a much-improved base of elevation data for calculating contours, slope and hydrologic derivatives. The NED is the primary elevation data product produced and distributed by the USGS. The NED provides the best available public domain raster elevation data of the conterminous United States, Alaska, Hawaii, and territorial islands in a seamless format. The NED is derived from diverse source data, processed to a common coordinate system and unit of vertical measure. The source data are distributed in geographic coordinates in units of decimal degrees, and in conformance with the North American Datum of 1983 (NAD 83). The source elevation values are provided in units of meters, and are referenced to the North American Vertical Datum of 1988 (NAVD 88) over the conterminous United States. The NED is updated on a nominal two month cycle to integrate newly available, improved elevation source data.

(Link to Metadata) This dataset is derived from the multi-resolution National Elevation Dataset (NED), at resolutions of both 1/3 arc-second (approx. 10 meters) and in limited areas, 1/9 arc-second (approx. 3 meters). Contours derived from this data may appear to have ?shifted? when compared to the 7.5 minute USGS Quad Maps, i.e., Digital Raster Graphics, for a variety of reasons: 1) The NED is a multi-resolution dataset, e.g., areas with LiDAR source data have superseded the original "quad" contours; 2) A result of the original contour vectors undergoing a NAD27 to NAD83 conversion, then the contour vector-to-raster resampling that produced the initial grid, followed by a resampling of that initial 10m grid to the 1/3 arc-second NED (~10.29m) and finally the raster-to-vector conversion yielding the current contours. VCGI extracted the Vermont portion of the NED and re-projected into Vermont State Plane Meters NAD83 (vertical units in feet). Production artifacts were filtered out of this source data prior to acquisition resulting in a much-improved base of elevation data for calculating contours, slope and hydrologic derivatives. The NED is the primary elevation data product produced and distributed by the USGS. The NED provides the best available public domain raster elevation data of the conterminous United States, Alaska, Hawaii, and territorial islands in a seamless format. The NED is derived from diverse source data, processed to a common coordinate system and unit of vertical measure. The source data are distributed in geographic coordinates in units of decimal degrees, and in conformance with the North American Datum of 1983 (NAD 83). The source elevation values are provided in units of meters, and are referenced to the North American Vertical Datum of 1988 (NAVD 88) over the conterminous United States. The NED is updated on a nominal two month cycle to integrate newly available, improved elevation source data.

This dataset includes expected Net Value Change (eNVC) values summed by Fireshed Project Areas that intersect National Forest administrative boundaries in the Intermountain Region. Project Areas are approxiamately 25,000 acre accounting units nested within Firesheds. Values are classified in quartiles and given adjective ratings: Very High (>75th Percentile Loss)High (50th-75th Percentile Loss)Moderate (25th-50th Percentile Loss)Low (0-25th Percentile Loss).Housing Density and Critical Infrastructure (power lines and communication sites only) HVRA datasets were obtained from the Risk Management Assistance National Wildfire Risk Assessment (RMA-NWRA). The Surface Water HVRA uses the Forest to Faucets 2.0 (F2F) dataset with response functions from the RMA-NWRA. The analysis area used for modeling these three HVRAs includes the Firesheds that intersect the National Forest administrative boundaries and Wildfire Crisis Strategy Landscapes in the Intermountain Region.HVRA datasets used in the 2023 Assessment.HVRADESCRIPTIONHousing Density (HD)Housing Unit Density from Wildfire Risk to Communities (RMA-NWRA).Critical Infrastructure (CI)Combined infrastructure datasets, transmission lines and communication sites only, from national HIFLD open data (RMA-NWRA).Surface Water (SW)Forests to Faucets 2.0 data with Existing Vegetation Type (EVT) and slope covariates and response functions used in the RMA-NWRA. Only used IMP levels 6-10 (or top 50%).All HVRAs used annual burn probability (BP) data from 2020. The BP data are from the large-fire simulator (FSim) modeling, using a LANDFIRE 2020 fuelscape at 270m resolution. The flame length probability grids use WildEST modeling with a 2023 fuelscape incorporating disturbances through 2022. For risk calculations, the BP data were ‘up-sampled’ using the Pyrologix methodology (running two low-pass (3x3 pixels) focal-mean filters and a bilinear resample to 30m) on the BP raster to give pixels that are burnable at 30m resolution, but non-burnable at 270m resolution, a BP value.For details on the quantitative wildfire risk assessment process, refer to:Scott, Joe H.; Thompson, Matthew P.; Calkin, David E. 2013. A wildfire risk assessment framework for land and resource management. Gen. Tech. Rep. RMRS-GTR-315. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 83 p.