|
PLEASE NOTE: This is a preliminary release of the North American 30 arc-second DEM. Constructive comments from users of this data set are greatly appreciated. Please contact Kris Verdin (kverdin@dg1.cr.usgs.gov) or Sue Jenson (jenson@dg1.cr.usgs.gov) with your comments. A Digital Elevation Model (DEM) consists of a sampled array of elevations for ground positions that are normally spaced at regular intervals. The EROS Data Center's (EDC) 30 arc-second DEM project includes generation of 30 arc-second DEM data for the entire world. These data are being made available to the public via anonymous ftp as they become available. As major geographic regions are completed, the data, along with various derivative data sets, will be published on CD-ROM. As of July 1995, North America, Africa, Japan, Madagascar and Haiti are complete and available for distribution. The South American data set is under development, as are Europe and Asia. The North American 30 arc-second DEM was compiled from various data sources. Elevation data at a 3 arc-second resolution exist for a large portion of the North American continent. The conterminous United States, Hawaii, Mexico and the Caribbean Islands have complete coverage at 3 arc-second resolution. Most of Alaska and parts of Central America and Canada are also covered by this higher resolution data. Where ever the higher resolution 3 arc-second data were available and distributable, they were generalized to the desired 30 arc- second resolution using a nearest neighbor resampling. (The 3 arc-second data covering the Caribbean, parts of Central America and Hawaii were received by EDC already resampled to 30 arc-second. The data for Alaska, Canada, the United States and Mexico were resampled at EDC). The 30 arc-second DEM for the parts of the North American continent lacking 3 arc-second data was derived from the 1:1,000,000 scale Digital Chart of the World (DCW) mapping. The point and contour hypsography coverages were processed into a 30 arc-second DEM using the ANUDEM gridding software. This software utilized the DCW hydrography information along with the hypsography to generate a hydrologically realistic DEM. This technique was used to produce the 30 arc-second DEM in all areas not covered by existing, distributable DEMs, including a large part of Canada and smaller portions of Alaska and Central America. The data generated in this fashion were merged with the existing resampled data using various mosaicing techniques to minimize the discontinuities between the different data sources. Data Distribution The North American 30 arc-second DEM covers the North American continent as well as Hawaii. The raster image contains representative elevation values for each pixel in meters above MSL. The DEM has a longitudinal extent of 180W to 50W and latitudinal extent of 6N to 90N. In order to facilitate distribution of the data, the DEM has been divided into four equal pieces as follows: EXTENT DEM Name Longitudinal Latitudinal na_30_dem1 180W - 115W 6N - 48N na_30_dem2 180W - 115W 48N - 90N na_30_dem3 115W - 50W 6N - 48N na_30_dem4 115W - 50W 48N - 90N Data are distributed as compressed BIL images, 16-bit straight raster image files in a latitude/longitude coordinate system. The image files were compressed using the GNU 'gzip' utility and can be identified by the .bil.gz extensions. These files must be uncompressed before use. They can be uncompressed with the gzip utilities or if you do not have access to gzip, the FTP server can uncompress the files as you retrieve them. To do this, simply leave off the ".gz" extension when retrieving the file (NOTE: This option is not available through MOSAIC). For example, to retrieve the file "na_30_dem1.bil.gz" without compression just use "get na_30_dem1.bil". Please note that the uncompressed files are typically many times larger than the compressed versions and therefore will take much longer to transmit. If you would like to obtain the gzip program, it is available via anonymous FTP at the following sites: prep.ai.mit.edu:/pub/gnu wuarchive.wustl.edu:/systems/gnu Each image file is accompanied by five ancillary files (header file, world file, statistics file, coordinate file, and data descriptor record). The DEM Name makes up the prefix of each file. The three digit suffix varies depending on the file type. The formats of the ancillary files are described below: Header File (.hdr) The header file can be used as input to the ARC/INFO Image Integration application. Information contained within this file can also be used with other image display packages such as IMDISP. The following list identifies the keywords used in the header file (ARC/INFO User's Guide, Image Integration, 5-4 to 5-10): byteorder: byte order in which image pixel values are stored M = Motorola byte order (high order byte first--Sun, HP, etc.) Example Header file: BYTEORDER M World File (.blw) The world file can be used for image-to-world transformation when displaying an image using the ARC/INFO Geographic Information System (GIS) Image Integration routine. The following is an example of a world file with a description of each record (ARC/INFO User's Guide, Image Integration, 2-3 to 2-5): Example World file: Description 0.0083333333333333 x dimension of a pixel (decimal degrees) 0.00000000000000 rotation term (will always be zero) 0.00000000000000 rotation term (will always be zero) -0.0083333333333333 negative y-dimension of a pixel (in decimal degrees) -179.99583333333333 longitude of the center of the upper-left pixel 47.995833333333333 latitude of the center of the upper-left pixel
The statistics file contains the band number, the minimum pixel value (elevation expressed in meters above mean sea level), the maximum pixel value, the mean, and the standard deviation. Example Statistics File: 1 -79.0 4314.0 1081.0 679.4 The Coordinate File (.tik) The coordinate file is named after the image with a .tik extension. The following list identifies the keywords used in the coordinate file. ulxmap: longitude of the center of the upper-left pixel ulymap: latitude of the center of the upper-left pixel urxmap: longitude of the center of the upper-right-pixel urymap: latitude of the center of the upper-right-pixel llxmap: longitude of the center of the lower-left-pixel llymap: latitude of the center of the lower-left-pixel lrxmap: longitude of the center of the lower-right-pixel lrymap: latitude of the center of the lower-right-pixel xdim: x dimension of a pixel in geographic units (decimal degrees) ydim: y dimension of a pixel in geographic units (decimal degrees) Example Coordinate File: ulxmap -179.9958333333333 ulymap 47.9958333333333 urxmap -115.0041666666666 urymap 47.9958333333333 llxmap -179.9958333333333 llymap 6.0041666666666 lrxmap -115.0041666666666 lrymap 6.0041666666666 xdim 0.008333333333333 ydim 0.008333333333333 The Data Descriptor Record (.ddr) This file has been included to allow the image to be read directory by EDC's LAS software. To take advantage of this, the image file extension will have to be renamed from .bil to .img. This file contains all of the information contained in the other data files. Example Data Descriptor Record: IMAGE NAME:na_30_dem1.bil NL:5040 NS:7800 NB:1 DTYPE:INTEGER*2 LAST MODIFIED: DATE: TIME: SYSTEM:ieee-std PROJ. CODE:(0)GEOGRAPHIC Valid:VALID ZONE CODE:0 Valid:INVALID DATUM CODE:0 Valid:INVALID PROJ. PARM: Valid:VALID A: 0.00000000000000E+00 0.00000000000000E+00 0.00000000000000E+00 B: 0.00000000000000E+00 0.00000000000000E+00 0.00000000000000E+00 C: 0.00000000000000E+00 0.00000000000000E+00 0.00000000000000E+00 D: 0.00000000000000E+00 0.00000000000000E+00 0.00000000000000E+00 E: 0.00000000000000E+00 0.00000000000000E+00 0.00000000000000E+00 CORNER COOR: Valid:VALID ULcorner:4.799583333333333E+01 -1.79995833333333E+02 URcorner:4.799583333333333E+01 -1.15004166666666E+02 LLcorner:6.004166666666666E+00 -1.79995833333333E+02 LRcorner:6.004166666666666E+00 -1.15004166666666E+02 PROJ. DIST:8.33333333333333E-03 8.33333333333333E-03 Valid:VALID PROJ. UNITS: degrees Valid:VALID INCREMENT:0.00000000000000E+00 0.00000000000000E+00 Valid:INVALID MASTER COOR:0 0 BAND NO:1 MINIMUM:-7.9000000000000E+01 Valid:VALID MAXIMUM:4.31400000000000E+03 Valid:VALID DATA SOURCE: SENSOR TYPE: CAPT. DIRECTION: DATE: TIME: Tape Organization This README.txt file will be the first record on the tape, then followed by the five ancellary files and image. Each ancellary file is one record, and the for the image, each row is a record. North America will be on two tapes. Each will have the README.txt. Tape 1 will have na_30_dem1 and na_30_dem2 Tape 2 will have na_30_dem3 and na_30_dem4 The file list for Tape 1 is: README.txt File: 1 Record: 1 Record Size in Bytes: 19891 na_30_dem1.hdr 2 2 308 .blw 3 3 230 .stx 4 4 29 .tik 5 5 324 .ddr 6 6 1803 na_30_dem1.bil 7 7 - 5046 15600 na_30_dem2.hdr 8 5047 362 .blw 9 5048 230 .stx 10 5049 25 .tik 11 5050 376 .ddr 12 5051 1803 na_30_dem2.bil 13 5052 - 10091 15600 The file list for Tape 2 is: README.txt File: 1 Record: 1 Record Size in Bytes: 19891 na_30_dem3.hdr 2 2 364 .blw 3 3 230 .stx 4 4 25 .tik 5 5 390 .ddr 6 6 1803 na_30_dem3.bil 7 7 - 5046 15600 na_30_dem4.hdr 8 5047 360 .blw 9 5048 230 .stx 10 5049 25 .tik 11 5050 386 .ddr 12 5051 1803 na_30_dem4.bil 13 5052 - 10091 15600 DCW Gridding Techniques For the areas in Canada, Alaska and Central America lacking 3 arc-scond data, the ANUDEM software was used to grid a DEM from the 1:1,000,000 scale DCW data set. The map source for the DCW database is the U.S. Defense Mapping Agency DMA)'s Operational Navigation Chart (ONC) series. This is the largest- scale map series that provides consistent, continuous global coverage of essential basemap features. A brief description of the techniques used to produce a DEM from the DCW data set is included here. Drainage Orientation: Initially, DCW files are stored as vector contours (hypsography) and hydrology. Using automated stream orientation procedures, contour line and point data are compared to the hydrology network to determine drainage direction. Drainage lines are oriented in the direction of flow, so that they are ordered from the highest point to the lowest. After the stream orientation process is complete, the resulting output is verified and edited to ensure that all streams are represented as flowing downhill. DEM Generation: Once data are pre-processed, they are used as input into the Australian National University Digital Elevation Model (ANUDEM) generation program developed by Michael Hutchinson, Centre for Resources and Environmental Studies, Australian National University. This program first reads input elevations, windows the data to the specified map limits, then generates a grid at 30-arc-second intervals. Elevation data are generalized by accepting a maximum of 4 data points per grid cell and discarding any remaining points. Contour and hydrography line data are generalized by accepting a maximum of one line per grid cell. The program then employs a multi-grid method that calculates grids at successively finer resolutions until the specified grid resolution is achieved. During this process, drainage conditions are imposed to remove sinks where possible. Values at grid points not occupied by data points are calculated by Gauss-Seidel iteration with over relaxation (SOR method) subject to an appropriate roughness penalty and ordered chain constraints. The ordered chain constraints are obtained from the stream line, sink point, and contour line data and through automatic drainage enforcement as calculated by the program. Starting values for the first coarse grid resolution are calculated from a least squares plane fit to the data points. Values for each succeeding grid are linearly interpolated from the preceding grid (M.F. Hutchinson, unpub. data, 1991). Data Characteristics Spatial Resolution Spacing of the elevations along and between each sample is 30-arc-seconds (approximately 1 km). The horizontal datum is WGS84. Elevation values are expressed in meters above mean sea level. Accuracy The absolute accuracy of the DCW vector information is 2000 meters circular error (horizontal) and + or - 650 meters linear error (vertical) at 90-percent confidence as defined by the Defense Mapping Agency (DMA). The grid generated from these data will be no more accurate than this source. The accuracy for the grid has not been measured or calculated. Data Availability Procedures for Obtaining Data 30 arc-second data sets for North America, Africa, Japan, Madagascar and Haiti are available through an Internet anonymous File Transfer Protocol (FTP) account at the EDC (at no cost). To access this account: 1. FTP to 152.61.128.6 (edcftp.cr.usgs.gov) 2. Enter "anonymous" at the Name prompt. 3. Enter your email address at the Password prompt. 4. Change (cd) to the "pub/data/30ASDCWDEM" subdirectory. 5. The files are located under the subdirectories: NORTH_AMERICA AFRICA HAITI JAPAN MADAGASCAR 6. Files are named after the image with the following extensions: *.bil.gz = compressed image file *.blw = world file *.hdr = header file *.stx = statistics file *.tik = coordinate file *.ddr = data descriptor record 7. Enter "binary" to set the transfer type 8. Use get or mget to retrieve the desired files. Orders for tape copies can be placed through the GLIS Inventory SUMMARY RESULTS screen or by contacting the EROS Data Center, Customer Services section, at the following address: Customer Services EROS Data Center Mundt Federal Building Sioux Falls, SD 57198 TEL: (605) 594-6151 FAX: (605) 594-6589 Customer Services can also be contacted for information about DCW DEM data that are in progress. As additional geographic areas become available they will be announced in the GLIS News. Products and Services 30 arc-second DEM data are available from the EROS Data Center on unlabeled magnetic tape. North America, Africa, Madagascar, Japan and Haiti datasets can be obtained through anonymous FTP. Instructions for accessing the anonymous FTP account can be found under Procedures for Obtaining Data. Applications and Related Data Sets Moderate resolution (100 meter to 1 kilometer) topographic data have applications in many diverse land science fields such as geology and geophysics, ecology, soil science, botany, and glaciology. Topographic data are also critical to procedures used for correcting and/or presenting remotely sensed satellite and other global data. The 30 arc-second DEM data, produced for use in conducting large-area studies, have been generated at a resolution which is compatible with the Advanced Very High Resolution Radiometer (AVHRR) sensor.
Environmental Systems Research Institute, 1991, Image Integration ARC/INFO User's Guide: Redlands, California, Environmental Systems Research Institute, Inc. Hutchinson, M.F., 1988, Calculation of hydrologically sound digital elevation models: Proc. Third Inter. Symp. Spatial Data Handling, Columbus, Ohio, August 17-19. Hutchinson, M.F., 1989, A new method for gridding elevation and stream line data with automatic removal of pits: J. Hydrol, 106, 211-232 p. Hutchinson, M.F., 1991, A continental hydrological assessment of a new grid- based digital elevation model of Australia: Hydrological Processes 5, 45-58 p. |