Application of Satellite and GIS Technologies
for Land-Cover and Land-Use Mapping at the
Rural-Urban Fringe: A Case Study
Paul M. Treitz, Philip J. Howarth, and Peng Gong
Earth-Observations Laboratory, Institute for Space and Terrestrial Science, Department of Geography, University of Waterloo,
Waterloo, Ontario N2L 3G1, Canada
[AasTRACT: SPOT HRV multispectral and panchromatic data were recorded and co-registered for a portion of the rural-
trban fringe of Toronto, Canada. A tworstage digital analy
contextual classification of eight land-cover and land-use
ses resulted in an overall Kappa coetficent of,
‘algorithm incorporating 2 spectral-las fequency-based
2 percent
for training-ares data and a Kappa coefficient of 70.3 percent for test-area data. A matrivoverlay analysis was then
performed within the geographic information system (Cis) to combine the land-cover and land-use classes generated,
Rom the SPOT digital classification with zoning information for the area. The map that was produced has an estimated
Interpretation acmieacy of 78 percent. Global Positioning 8
tem (GPs) data provided a positional reference for new
road networks. These networks, in addition to the new land-cover and land-use map derived fom the SPOT HRV data,
provide an up-to-date synthesis of change conditions in the area.
INTRODUCTION
ECIONAL AND MUNICIPAL PLANNERS require up-todatein-
formation to effectively manage land development and plan
for change. In uiban areas, particulary af the rura-urban fnge,
this change is very vapid: Asa result is dificult to maintain
tupto-ute information on new housing and industal/com-
ircal development. This is particulary true fr regional mu
‘lipaltes whose juiadictions cover large areas.
‘The landvuse map, a8 source of thematic information, has
ben an important component of ban and regional planning
formany yeas. In areas where change s marginal or ver sow,
land-use maps that are considered relatively old (e., 10 to 26
years) may continue fo portay adequately current condi¥ons,
tnd thus provide useful information. However, this i not the
ase in areas of rapid change, such a8 the rural-urban fringe,
Sehere the entire landecape tan change overs short period of
time, Here, fields and open areas are converted to residential
subivisions and commercalindustral plazas. In sich areas,
the most recent map may be of litle value to the person re-
ailing uptodate infomation
Even with the availability of satelite imagery and computer
storage of information, the sage has not yet ben reached where
tptonate information ean be rapidly and easily provided.
However, steps are being taken to achieve this. In Ontario, for
txample, data obtsined ss part of the Ontario Basie Mapping
(Oth) program are being digitized and input to geographic ine
formation systems. This provides baseline information for an
area, but the information is only as up-to-date a6 the aeal
photographs trom which it was mapped. In practal terms,
there has been a great deal of improvement in recent years i
dato collection, storage, and presentation. However, mich in-
formation on land cover and land use is il five to ten years
Out of date and for many areas i to old for operational use
For example forthe Town of Markham atthe rural-urban finge
of Metropolitan Toronto, the latest O8M maps were published
in 1984, Thus, the land use shown for many areas is now in-
‘overt ora varely of planning purposes, it would be bene-
fia to have up-tocate Information. The queston i, how can
this be provided rapidly and economically?
Remote sensing has been recognized as a useful means of
supplying up-to-date information on activities within the urban
environment, including the rura-utban fringe Ehlers et, 190;
PROTOGRAMMETRIC ENGINEERING & REMOTE SENSING,
Vol. 8, No. 4, April 1992, pp. 439-48.
Forster, 1985; Jensen and Toll, 1982). However, itis felt that the
interfacing of Gis technology with remote sensing will provide
the maximum information content and analysis capabilities and
thus be of benefit to land-use planners (Nellis ef al., 1990). It
hhas been recognized (e-g., Quarmby and Cushnie, 1989; Fors-
ter, 1985; Welch, 1985) that there are many advantages to com=
bining remotely sensed data with existing spatial, image, and.
statistical data, thereby maximizing the information upon which
responsible decisions for land-use planning can be made. Geo-
‘graphic information systems (Gis) technology provides the me