DATASET DOCUMENTATION

Major Wash Network (greater than 2000 CFS)

OVERVIEW

A major wash network was screen-digitized using 1996, 1998 and 2000 orthophoto imagery as a backdrop and using the existing Pima County washes.shp file as a guide to major washes. Only arcs with an attribute of greater than 2000 CFS in washes.shp were updated. Several attributes were added to this linework: Wash Name, Vector Source, Specific CFS Number, Bank Material, and General CFS Category. TerraSystems Southwest performed this work for Pima County Flood Control District (PCFCD) as a subcontractor to CMG Drainage, between September 2002 and February 2003. Following is a summary of the completed data set and documentation of the data sources and procedures used in its creation.

[Note: in addition to this line coverage, a polygon coverage of washes greater than 10,000 CFS was created. Though an initial version of this polygon cover has been delivered to Pima County, a new version synchronized with this line coverage will be forthcoming.]

WORK ACCOMPLISHED

A total of 1,393 linear miles of wash linework was digitized:

• 1,393 miles of wash linework attributed with bank material and general CFS
• 740 miles attributed with wash name
• 250 miles attributed with a general CFS Value of >10,000
• 1,142 miles attributed a general CFS Value of > 2,000 and < 10,000
• 145 miles assigned a specific downstream CFS value
• 286 miles of wash linework digitized based on 1996 1m digital Orthophotography
• 413 miles of wash linework digitized based on 1998 1-foot digital Orthophotography
• 694 miles of wash linework digitized based on 2000 1-foot digital Orthophotography

DOCUMENTATION

Vector Sources and Information

The linework in this coverage represents the low-flow channel, as best as could be determined from the available sources.

SOURCES

Several sources of vector data were used for this project

• Washes.shp – the original shapefile was used as a guide in determining which washes were greater than 2,000 CFS and 10,000 CFS
• Imagery

- 2000 Ortho (1' tiled, 1/2' ecw format)

- 1998 Ortho (1' ecw format)

- 1996 DOQQ (1m tiled, geo-tiff format)

• Topography

- 1998 and 2000 DTM, converted by TSSW to TIN by township

The new wash linework was digitized using the following procedures:

• Where a wash was clearly incised in the photo, it was digitized using only the aerial imagery as a guide.
• Where a wash was braided or indistinct in the photo, it was digitized using 1998 and 2000 TIN's as a guide. Where no TIN was available the heavy vegetation line was used as a guide for the wash line.
• Where water was visible in the photo, it was used as a guide for digitizing the low-flow channel. Where the water was visible but complexly braided, the line was digitized down the approximate center of the braids.

DIGITIZING SCALE

Washes were digitized at varying scales. A scale of 1:1500 to 2000 was used for the 1998 and 2000 1’ imagery. If the low-flow portions of the washes were difficult to see a TIN model was used to verify the channel and digitizing scale generally ranged from 1:300 - 1:1000. If a TIN model was not available for that particular area, the heavy vegetation line was used and digitizing scale varied widely, depending on amount of photo needed to determine this line. Washes based on the areas of 1996 DOQQ imagery were digitized at approx 1:3000.

GAPS IN COVERAGE

Where the source linework from washes.shp had gaps between major wash segments, these gaps were left unless an obvious path between them could be found in the photo or TIN model. In places where the wash went underground, a gap was left unless an obvious connection could be seen (e.g. under a bridge). This differs from washes.shp which would often show linework in these areas. In some cases, new arcs not in washes.shp were added to approximate the old topology where that old topology was radically changed by new development shown on the imagery. Any new arcs added are labeled "N" in PC_COVER field.

JUNCTIONS

Junctions between greater-than-10K washes and greater-than-2K washes were handled in two different ways. If a clear low-flow channel connecting the two was visible in the imagery, the 2K wash line was extended along that channel to connect to the 10K wash. If only a sandbar, or a complexly braided area was present between the end of the 2K wash and the 10K wash, then the two were connected with a straight segment.

Attribute Sources and Information

· ID – a unique feature id calculated from the machine id field of the coverage.

· Wash_Name - used washes_anno coverage provided by Pima County as the primary source and a PCFCD blueline set provided by CMG Drainage as a secondary source.

· CFS – this is a numeric item containing specific CFS values labeled for selected stream junctions on the PCFCD blueline set. This point value was assigned to the upstream arc. CMG Drainage considers this data inaccurate in places due to hydrologic changes upstream. It is included in this deliverable for historic value. It has not been comprehensively reviewed for accuracy.

· Source – this is the imagery data set that was used for screen-digitizing the vectors and attribute the bank_material field. The values are self-explanatory.

· PC_Cover – this field indicates whether the arc was present in the original washes.shp topology. A “N” value means that the arc was not present in the original cover and a “Y” value the opposite.

· Bank Material – this field contains a best estimate of the type of bank material present. A value with a “/” means that the opposite banks were different material. This field was attributed primarily from what could be seen from aerial imagery. These values were modified using information contained in bankprot.shp, downloaded from Pima County FTP site in February 2003.

· CFS_Gen – this is the general Cubic Feet per Second value for each wash segment as contained in washes.shp.

Quality Assurance Procedures

A number of quality assurance procedures were applied to this data set:

· Wash segments were randomly selected and verified against the highest resolution aerial imagery available.

· Questions as to flow path, gaps or other digitizing decisions were captured in a point file and reviewed with a Senior Hydrologist at CMG Drainage. A number of modifications to the washes were made as a result of this process.

· Each attribute field was summarized and non-standard values corrected.

· Each coded attribute field was symbolized and spot-checked against available source information. For example, the bank material field was compared against each segment contained in bankprot.shp.