f Spatial
on)
n sprawl
vided by
the UPC
'alidation
rtise and
Joaquim
ct.
ibul 2004
SPOT 5 Cadastral validation project in Izabal, Guatemala
M. Corlazzoli **, O.L. Fernandez à
7 UTJ-PROTIERRA, 21 Calle 10-58 Zona 13, colonia Aurora Il, Guatemala City, Guatemala,
meorlazzoli@utjprotierra.gob.gt and macacorlazzolic@hotmail.com
b oluciafm@ yahoo.com
Commission VIL, Working Group VII/4
KEY WORDS: Remote sensing, surveying, SPOT, identification, land, orthoimages, accuracy.
ABSTRACT:
Due to the resolution of the former generation of sensors, the use of satellite imagery in the surveying field ha been limited to the
construction of pre-cadastral solutions in rural areas. The project principal objective is to establish the potential use of SPOT 5
imagery - with its spatial resolution of 2.5m - as a main source for parcel identification in surveying activities. It was carried out by
the UTJ-Protierra, the Guatemalan project in charge cadastral im
plementation. The different areas of study were selected by their
distinct geographical features and cadastral interest. Two types of data were integrated to provide the database; namely SPOT 5
orthoimage and measurements of parcels with total sation and GPS equipments. The evaluation was made by comparing the
resulted digitised data from traditional measurements and the one from the image identification. The variation of surface between the
two datasets was considered very low for extended and medium estates. The variation of the distance between vertices was less than
two pixel resolution. Yet, properties in rugged terrain, small agricultural estates and peri-urban properties presented a considerable
variation of area and the difference between vertices was less than 10m. The use of the orthoimages as an input for indirect
methodology for cadastral projects is adequate for latifundium properties presenting low variation of topography, medium extensions
and for planning cadastral activities. In the case of urban centres, steep gradient areas, regions' presenting an elevated density of
parcel and dense vegetation fences, the use of SPOT orthoimage is limited.
1. INTRODUCTION
Digital analysis of remotely sensed data ha become an
important component of a wide range of land studies. However,
due to the resolution of the former generation of sensors, the use
of satellite images in the surveying field ha been limited.
A cadastre is a methodically arranged public inventory of data
concerning properties within a certain country or district, based
on a survey of their boundaries. Properties are identified by
direct or indirect methodologies such as total stations or aerial
photographies. The outlines of the properties and the parcel
identifiers are normally shown on large-scale maps which,
together with registers, may show for each separate property the
nature, size, value and legal right associated with the parcel. In
some developing countries, cadastral systems do not exist yet or
they are old and general, therefore new inexpensive operations
and approaches have to be established.
Lately, with the launch of new sensors with higher resolutions,
better capability to acquire derived DEMs, as well as a wider
availability of remote sensing imagery, from the internet for
instance, it is expected that satellite imagery will allow the
acquisition of primary data for surveyor's interests and
specifications. Nevertheless, few examples exist of applications
on high-resolution images such as SPOT 5.
Satellite images do not allow obtaining exhaustive cadastral
information such as legal status for instance, yet it is a global
and synthetic instrument that ha a significant impact on
mapping primary data.
This paper explores the potential use of SPOT 5 XS imagery for
parcel delimitation with cadastral specifications in Guatemala.
First, traditional parcel delimitation are exposed as well as
SPOT new characteristics are briefly explained. Second, a
concise explanation of the data and the study area are explored.
It is followed by different analysis of the potential use of SPOT
5 imagery and its cadastral delimitation accuracy. Finally the
results are summarised and discussed.
2. BACKGROUND
Several methodologies of cadastral parcel delimitation exist, in
general they can be divided in direct and indirect methods. The
principal objective of the direct methodologies is to locate the
relative position of points, so distance can be estimated later on.
Depending on the accuracy needed for the cadastral study
distinct techniques are employed, among the most used are
theodolite, total stations or GPS (Elayachi et al. 2001, Nabil
Nassif, 2002) and a combination of them. Yet, Mohd Said
(2003) reported that performance of GPS cannot be said on it's
capability under certain vegetation as tropical canopy.
On the other hand, in the indirect method, surveyors use aerial
photographies to delimitate parcel boundaries. The polygons
are digitized in a second step of the study. Gavish and Doytsher
(2002) mentioned that the use of indirect techniques is no less
accurate than direct terrestrial measurement. Nevertheless, as
noted by Juppenlatz (1991) conventional aerial
photogrammetric method, the production of air photos and the
resulting maps are so costly that certain cadastral budgets can
neither raise money nor justify such expenditures.
Different parameters defined the use of those methodologies,
for example in areas with elevated density of parcels like
villages, it is more recommended to use direct methodologies
with the use of theodolite or total stations. Where the density of
parcel is not elevated, surveyors usually use orthophotos.
Though, in rural areas, depending on the vegetation cover, the
parcel density and local factors, the two options can be selected.
Remote sensing with its utility for surveying large areas in a
time and cost effective manner, is considered as a possible