In: Wagner W„ Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B
580
REMOTELY SENSED DATA & GIS IN LAND RESOURCES MANAGEMENT FOR
REGIONAL PLANNING OVER SEMI-ARID PARTS OF NE BRAZIL
Harendra S.Teoria, Klaus A Ulbricht, Guttemberg da Silva Silvino
Laboratory of Remote Sensing and GTS. DSER/CCA/UFPB. Areia-PB. Brazil - teotia@terra.com.br
KEY WORDS: Semi-Arid, Land Resources Management, SPOT Imagery, ERDAS , Image Processing, Regional Planning,
Remote Sensing, Geographical Information system
ABSTRACT:
The present study was carried out by digital interpretation (Supervised Classification-MAXCLAS), which is based on the
methodology utilized by Kennard et all (The research work of our project was carried out through the 1988). The Satellite Data were
received from the Aero Space Research Establishment of Germany (DLR-Oberpfaffenhofen). The LANDSAT and SPOT data
were processed with ERDAS Software, operating on a high performance micro-computer. The digital interpretation was applied to
one million pixels in each area of each State (Paraiba, Cearâ and Piaui) to derive land use/land cover and major soil associations
information. In the digital image classification, based on the field observations and using interactive capabilities of ERDAS, total 120
observations (40 observation in each area), were selected for land use/ land cover classes. The digital interpretation was modified and
corrected in accordance with the conditions of the area. By using RECODE program of ERDAS Software for land use/land cover
classes, two maps, such as, land use/land cover map at the Level-II and major soil associations map for each area were generated.
The comparison of digital interpretation with reference information indicated that the digital interpretation is closely resembled field
observation and the overall classification accuracy was observed more than 85% in all the three areas of three States. The results of
our study were found very beneficial for land development, natural resources management, land evaluation, soil conservation and
land reforms programs of the Federal Government in the semi-arid region of northeastern Brazil.
1. INTRODUCTION
2. BACKGROUND DISCUSSIONS:
The northeast part of Brazil comprises of nine states and covered
by various physiological relief features and morphological forms.
The area generally has leveled to slightly undulating and
mountainous topography. At some locations the area is also
covered by natural pastures and waste lands, forest patches, and a
network of of rivers, many of which run only during the rainy
seasons. The approach for natural resources development and
management varies from one type of environment to another,
The soil and land management in semi-arid region is totally
different from the soil and land management in humid areas.
Nowadays, there is a tremendous pressure on the availability of
soil and land resources due to increasing population and grow
consumption of land for constructing houses and buildings in
major cities. In such a situation we must make an optimum
planning and management of our land through remote sensing
and GIS technology. These two technologies are the major tools
for solving of our complex natural resources planning problems.
The purpose of our research using remote sensing and GIS
technologies and ERDAS Software is to apply image processing
and pattern recognition techniques to Satellite Data, such as
LANDSAT-TM and SPOT HRV to drive various earth resources
information for the land development and management of the
region. Also, provide information about the land use/land cover
classification, Soil classification, land capability classification,
soil slope and relief features, which are used for the resources
management and regional planning of semi-arid regions of
northeastern Brazil. The aim of this research was to know the
realistic assessment of the extent to which SPOT and LANDSAT
data can be used for the survey of the study area, and to provide
basic information on the nature of the digital remote sensing data
and how they can be used for different applications.
Various types of remote sensing data, such as, SAR, MSS, TM,
ETM, MOMS, AVHRR and SPOT etc. lave been used for earth
resources management and development of semi-arid regions by
various governmental agencies, institutions, and universities of the
world. For example: Kennard et al. (1988) have worked on a GIS
system for land use planning and management of semi-arid
regions of northeastern Brazil, using digital image processing on
Landsat-TM and SPOT data. LaBash et al (1989) conducted a
digital image analysis of Landat-TM data in eastern Connecticut
for regional land use and land cover classification. The results were
found most satisfactory for regional planning in the sate of
Connecticut.. Ulbricht et al. (1992 used the supervised
classification for the soil and land use studies for a part of semi-
arid regions of Brazil. Teotia et al. (1996) did a very
comprehensive work for land use planning in semi-arid regions
of NE Brazil, usiusing SPOT HRV data. According to Silva
Junior (2003) the remote sensing and gIS are economic and
adequate for the land survey and land use classification of semi-
arid regions of Paraiba state of Brazil. Ribeiro G. do N ( 2006)
studied some remotely sensed data and found that the ERDAS
Imagine Software and SPOT data is a good combination for land
use and land cover mapping in order to develop the Agreste
region of the Paraiba sate of Brazil.
STUDY AREA: (Figures-1 and 2); (Photos 1,2, 3 & 4 ):
The study area falls in the states of Paraiba, Ceara, and Piaui,
which is mainly semi-arid. The area has various types of dry lands
and altitudes varying from 200 to 900 meters. Various types of
physiological relief features and géomorphologie forms are
encountered in the study area. Hydrologically, the area is covered
by a network of rivers, many of which are intermittent. Soils
derived from recent streams sediments are rich, whereas others
are poor to moderate in fertility. The major soil groups of the area
