The USGS maintains a set of 1:24,000 scale 7.5-minute paper topographic maps for the State of Indiana with hand-drawn drainage divides for streams with drainage areas greater than 5 square miles. These maps were prepared for the report 'Drainage areas of Indiana streams' (Hoggatt, R.E., 1975, Drainage areas of Indiana streams: U.S. Geological Survey, 231 p.). The USGS digitized these drainage divides, using ARC/INFO Geographic Information System (GIS). The NRCS assigned each delineated drainage basin a 14-digit hydrologic-unit code (HUC). Drainage basins with the same 14-digit HUCs were merged to form the final data set of subwatersheds called in_hu14.
A TIC coverage was generated for the State of Indiana and immediate vicinity with a coordinate TIC at every 7.5 minutes of latitude and longitude. These TICS corresponded to the corner vertices of the 7.5-minute topographic maps that included any area in Indiana. The drainage divides on most of the maps (693 of 710) were digitized by hand on an ALTEK model AC-40 digitizing tablet. The drainage divides on the remaining 17 maps were scanned on a Scangraphics model CF1000 scanner.
Following are the procedures used to hand digitize the drainage divides on 693 maps:
Each map was taped to the digitizing tablet. The ARCEDIT module of ARC/INFO was used to determine the Root Mean Square Error (RMSE) of the 4 map corners as compared to the 4 corresponding TICs in the TIC coverage. All RMSE values for hand-digitized maps were 0.016 (the equivalent of 32 feet on the ground) or less. The RMSE values, along with other quality-assurance data, are recorded in the arc attribute table, in_hu14.aat. The drainage divides on the map were digitized in stream mode with a vertex every 0.020 inches (the equivalent of 40 feet on the ground).
The following ARCEDIT tolerances were set:
INTERSECTARCS ALL NODESNAP CLOSEST,32 ARCSNAP ON,32 SNAPPING CLOSEST,64 GRAIN 21.15783271851 WEEDTOLERANCE 16
The drainage divides on the following 17 topographic maps were scanned:
Bluffton, Gilman, Quincy, Cloverdale, Hazelrigg, Roll, Deerfield, Montpelier, Saline City, Domestic, Newport, Shannondale, Frankton, Omega, Wallace, Gaston and Petroleum
Following are the procedures used to scan the drainage divides on these 17 maps:
Each map was scanned at 200 dots per inch (true optics out). The Scangraphics Editor was used to remove extraneous lines from the resulting raster-image file. The ARC/INFO module ARCSCAN was used to vectorize the raster-image file. All RMSE values for scanned maps were 0.013 (the equivalent of 26 feet on the ground) or less.
The following Arc Environment Properties were used in ARCSCAN during vectorization:
Copy parallel: default Move parallel: default Rotate angle: default Align distance: default Vertex distance: 0.556 Node snap tol: Off Arc snap tol: Off Intersect arcs: Off Unsplit Item: Off
The following Tracing Environment Settings were used in ARCSCAN during vectorization:
Straighten Properties: Generalization: Tolerance: 0.01 Line Straightening: Distance: 10.000 Range: 30.000 Angle: 15 Corner: 30
Line Properties: Width: 16.000 Value: 1 Gap: 15.000 Dash: 20.000 Hole: 0.05 Fan angle: 45 2nd angle: 90 Variance: 200
The following general procedures were performed subsequent to digitizing or scanning in 6-10 maps:
After approximately 6-10 maps were digitized, the coverage was stripped of line overshoots and undershoots using the CLEAN command with DANGLE=50 and FUZZY_TOLERANCE=1. The newly digitized lines were then examined along the edges of the map where they should meet previously digitized lines from an adjoining map. Occasionally the ends of lines on adjoining maps did not join correctly. The lines on the original maps were examined, and in most cases, the drainage divides on the original paper maps did not precisely align across map edges. These lines were corrected on the original maps and the appropriate portion of the line was re-digitized.
All digitized lines were examined for node dangles, double lines, and sliver polygons. These errors were corrected and a checkplot of each digitized map was plotted at 1:24,000 scale with a line thickness of 0.025 inches. Checkplots were placed over the original maps on a light table, and any portion of the digitized linework that did not overlie the original line was re-digitized. For quality control, 36% of the digitized maps (256 maps) were later re-plotted and all lines were checked by either a USGS staff member (other than the individual who originally digitized the map) or an NRCS staff member. Dates for this quality-assurance activity are recorded in the arc attribute table in_hu14.aat.
Digitizing efficiency began to decrease after approximately 35-40 maps were digitized into the coverage. Undue amounts of time were being spent waiting for the image of digitized lines to be displayed on the computer screen. At this point, a second temporary TIC coverage was generated in the same manner as the original TIC coverage. Additional maps were then digitized into this temporary TIC coverage. Periodically, the temporary coverage was joined with the original coverage using the UNION command. All lines were then deleted from the temporary coverage. This process was repeated until all requisite maps were digitized.
Each polygon in the coverage represented a distinct drainage basin and was attributed with 1) a code indicating whether the polygon represented a contributing or non-contributing area, and 2) a number based on the discharge point for the basin. These discharge-point numbers were based on the first column of data in Table 3 in Hoggatt (1975). An electronic file of the discharge-point numbers was sent to NRCS staff. NRCS staff used copies of the 1:24,000 scale maps with delineated drainage basins to determine 14-digit HUs according to their guidelines in NRCS National Instruction No. 170-304 (U.S. Department of Agriculture, Natural Resources Conservation Service, 1992, Mapping and digitizing watershed and subwatershed hydrologic-unit boundaries: National Instruction no. 170-304, 26 p., accessed May 30, 1997 at URL http://www.nhq.nrcs.usda.gov/hu/ni70304.html). NRCS staff then updated the electronic file of discharge-point numbers with the appropriate 14-digit HUC's and returned the file to USGS staff. The USGS staff converted the file to an INFO table and placed the 14-digit HUC into the item called HUC14_C. The RELATE function in INFO was used, with the discharge-point number as the relate item, to attribute each polygon with the 14-digit HUC in an item called HUC_14. This procedure rendered several contiguous polygons with identical 14-digit HUC's (HUC_14). Lines dividing contiguous polygons with identical 14-digit HUC's were deleted using the DISSOLVE command to produce the final polygons in the subject data set.
The STATISTICS function in ARC was used to calculate the total non- contributing area for each 14-digit HUC. The RELATE function in INFO was then used to attribute each polygon in the final data set with the total non-contributing area, calculated in acres. Manual adjustments were made where individual 14-digit HU's were represented by two or more non-contiguous polygons.
To ensure that each 14-digit HU had the correct boundaries and the correct 14-digit HUC, checkplots of each map were plotted at 1:24,000 scale with each HU labeled with its HUC. These checkplots were then sent to NRCS. Any errors were noted and corrected, and new checkplots were generated and re-checked until every map was correct. Quality-assurance data for this activity is included in the arc-attribute table (AAT) of the final data set.
HU names, provided by the NRCS, were added to the polygon-attribute table (PAT) of in_hu14. These names are based on the primary waterway within each HU.
The arcs were assigned a 2-digit attribute, called HU_DIV, indicating the boundary type represented: 6 = Accounting Unit, 8 = Cataloging Unit, 11 = Watershed, 14 = Subwatershed, or the Indiana state line. Note that arcs representing the state line have a hu_div = 0 and all other arc attributes are blank or 0.
Three colleague reviews of the final data set of 14-digit hydrologic
units, the metadata and the Fact Sheet were performed in July 1999, two
reviews by USGS staff and one review by a staff member of one of the cooperating
agencies, the Indiana Department of Environmental Management.
Filename--IN_HU14.PAT:
| COLUMN | ITEM NAME | WIDTH | OUTPUT | TYPE | N.DEC
(number of decimal places) |
| 1 | AREA | 8 | 18 | F | 5 |
| 9 | PERIMETER | 8 | 18 | F | 5 |
| 17 | IN_HU14# | 4 | 5 | B | - |
| 21 | IN_HU14-ID | 4 | 5 | B | - |
| 25 | HUC_8 | 8 | 8 | C | - |
| 33 | HUC_11 | 11 | 11 | C | - |
| 44 | HUC_14 | 14 | 14 | C | - |
| 58 | HU_NAME | 55 | 55 | C | - |
| 113 | HU_ACRES | 4 | 12 | F | 1 |
| 117 | NONCONTRIB_ACRES | 4 | 12 | F | 1 |
Filename--IN_HU14.AAT:
| COLUMN | ITEM NAME | WIDTH | OUTPUT | TYPE | N.DEC
(number of decimal places) |
| 1 | FNODE# | 4 | 5 | B | - |
| 5 | TNODE# | 4 | 5 | B | - |
| 9 | LPOLY# | 4 | 5 | B | - |
| 13 | RPOLY# | 4 | 5 | B | - |
| 17 | LENGTH | 8 | 18 | F | 5 |
| 25 | IN_HU14# | 4 | 5 | B | - |
| 29 | IN_HU14-ID | 4 | 5 | B | - |
| 33 | QUADNAME | 25 | 25 | C | - |
| 58 | QUAD_PUB_DATE | 10 | 10 | C | - |
| 68 | DIGITIZE_DATE | 8 | 8 | I | - |
| 74 | RMSE | 4 | 12 | F | 3 |
| 78 | ARC_QA_DATE | 8 | 8 | I | - |
| 84 | ARC_REMARKS | 150 | 150 | C | - |
| 234 | HU_DIV | 2 | 2 | I | - |
Although these data have been used by the U.S. Geological Survey, U.S. Department of the Interior, no warranty expressed or implied is made by the U.S. Geological Survey as to the accuracy of the data.
The act of distribution shall not constitute any such warranty, and
no responsibility is assumed by the U.S. Geological Survey in the use of
this data, software, or related materials.
![[Go Home]](http://in.water.usgs.gov/pics/gohome.gif)