The following is an excerpt from a paper submitted by RECON Enviromental,
Inc. Titled "Land Cover Data Assessment in Pima County"


Wildlife Habitat Inventory Project (WHIP)

The Pima County habitat inventory was conducted to collect habitat
information about the urban and suburban environment of the Greater
Tucson metropolitan area. The goal of this study was to develop a GIS
data base that includes delineations of systematically classified land
cover types which are keyed to quantitative vegetation characteristics
(Shaw et al. 1996).
 
Phase I consisted of a pilot study conducted to develop a typology of
urban and suburban land cover categories (Shaw et al. 1993). Phase II
mapped the study area using this typology and determined vegetative
attributes for each land cover type (Shaw et al. 1996). Land cover was
mapped by merging land use databases from Pima County and Tucson based
on 1990 orthophotos, then updating these using 1:12000 1995 aerial
photographs and ground truthing. Land cover categories were character-
ized by collecting quantitative and qualitative data for randomly
selected samples within each land cover category. Methods and variable
descriptions are detailed in the Phase II final report (see Shaw et al. 1996).

The WHIP GIS coverage, made available through DOT, provides higher
resolution mapping than GAP for this nearly one-million-acre study area.
Natural vegetation is classified according to BLP, although it is unclear
to what extent mapped polygons may represent disturbed vegetation. Riparian
areas in particular are delineated in the WHIP study where none appear in
GAP, and mesquite scrub vs. cottonwood-willow are differentiated, although
it is believed that the cottonwood-willow delineation and classification is
modest (Fonseca 1999a). An accuracy assessment of this data has not been
conducted, however, in comparing these data with other sources, and given
that they are based on recent aerial photographs, polygon boundary
delineations and series level land cover classifications are probably quite
accurate.

The urban and suburban land cover types, that are the focus of discussion
and analysis in the report, are missing from the current GIS coverage.
These appear as "no data" on fig 2.  Coverage polygons with no attribute
data appear as "unclassified." Except for a BLP code, there is no attribute
data for natural vegetation types, so it is assumed that attributes de-
tailing vegetation characteristics discussed in the WHIP report, are stored
in a separate database. These were certainly developed since they provided
the basis of GIS analyses discussed in the report. 

This coverage replaces GAP mapping in the composite land cover data layer.
Further research should be conducted to classify unclassified polygons, and
recover the missing urban/suburban land cover data as well as the land cover
attribute database upon which past and potential future habitat modeling
analyses could be conducted. WHIP land cover classification could also provide
the basis for a modified land cover classification system.