USAGE OF DIFFERENT SPECTRAL BANDS IN AGRICULTURAL ENVIRONMENTAL 
PROTECTION 
A T ' E ] 
P. Burai * , J. Tamas ^ , Cs. Lenart ^ , I. Pechmann ' 
* Department of Water and Environmental Management, University of Debrecen, Hungary,4032 Debrecen, Bószórményi üt 138. - 
pburai@gisserverl.date.hu 
? Department of Water and Environmental Management, University of Debrecen, Hungary,4032 Debrecen, Bószórményi üt 138. - 
tamas(@gisserverl.date.hu 
* Department of Water and Environmental Management, University of Debrecen, Hungary,4032 Debrecen, Bószórményi üt 138. - 
lenart@gisserver1 .date.hu 
Department of Water and Environmental Management, University of Debrecen, Hungary,4032 Debrecen, Bôszôrményi ût 138. - 
ipechmann@gisserver | .date.hu 
KEY WORDS: Agriculture, Hyperspectral, Multispectral, Classification, GIS 
ABSTRACT: 
Hyper and multispectral imaging systems widely used in agricultural and environmental protection. Remote sensing techniques are 
suitable for evaluating environmental protection hazards, as well as for agriculture resource exploration. In our research we compared 
aerial hyper and multispectral images, as well as multispectral digital camera images with the background data in the test site. 
Hyperspectral records were obtained by a new 79-channeled aerial /DAIS 7915/. We made additional images with a TETRACAM 
ADC wide band multispectral camera. We have chosen an farm as test sites, where intensive crop cultivation takes place, so soil 
degradation and spreading of weeds can be intensive as a result of land use and irrigation. The first objective of this study was to 
evaluate the applicability and usability of the multi- and hyperspectral images in large spatial resolution according to the prescription 
of NAKP (National Program of Agi- Environmental Protection). Multispectral analyses were complied for to evaluate the solidity of 
supervised classification. During the analyses of the images, the spectral characters of various vegetation types and possibilities of 
their classification as well as the spectral identification of the ecological buffer zones and filtering areas and their GIS interpretation 
were investigated. 
INTRODUCTION For environmental state assessment and score of the areas 
involved in the Program, data sources with large spatial 
resolution are worth to apply. To distinguish the commonly 
Among the aims of the National Program for Agri- small fileds from the island-like vegetation spots inserted in 
Environmental Protection (NAKP) the principles of a the culture land the traditional methods (CORINE) are not 
sustainable landuse matched to the facilities of Hungary's suitable any more. Our study's aims are to evaluate the 
different regions are declared according to the Union's applicability and usability of the multi- and hyperspectral 
2078/92 directive (Angyan et al, 1999). The farmers who images in large spatial resolution according to the 
have joined the above mentioned program for to gain the prescription of NAKP. During the analyses of the photos, the 
support cannot lead their farms without considering its spectral characters of various vegetation types and 
principles in connection with environmental protection and possibilities of their classification as well as the spectral 
nature reservation. Assignation the suitable landuse system identification of the ecological buffer zones and filtering 
means a serious problem, but to control whether the areas and their GIS interpretation were investigated. 
regulations are kept or not is much more difficult. To the 
right application of this supporting system the investigation 
of the biotical and abiotical environment surrounding the MATERIALS AND METHODS 
cultivated lands is needed. Environmental information can be 
obtained by classifying air- and spaceborn remote sensing Test region 
images. Knowledge of changes in land cover is becoming 
increasingly important from both the ecological and The investigated area is situated in the eastern part of the 
economical point of view (Janssen et al, 1990). Hungarian Great Plain, next to a small rural town called 
Hajdünánás. It has a special location being on the border of 
There is an increase in the use of remote sensing two geographical regions. On one hand, it is continuous with 
technologies. Image interpretation is faster and less expensive the area of the Hortobágy, a typical solonetz-type salt- 
than conducting ground survey (Floyd, 1996). Remote . affected area of the country. On the other hand, the area of 
sensing data have been used to detect and classify vegetation the Hajdüság loess field characterized by chernozem-type 
using supervised classification, cluster analysis, decision soils also reaches the region. Its climate is mainly continental, 
trees, neutral networks, etc. (Carpenter et al., 1999; DeFries strictly saying wet-contitnetal with 9,9"C mean temperature. 
& Townshend, 1994; Friedl & Brodley, 1997; Martonchik, The annual average rainfall takes 580 mm, which’s 
1994; Running et al., 1995). inhomogeneous distribution often causes inauspicious 
conditions in agricultural point of view. Besides intensive 
cultivation forms, extensive grassland and water reservoir are 
86 
Inter 
— 
als 
and 
inst 
Da 
Ael 
fro 
aT 
Im: 
imc 
ban 
667 
pix 
gen 
sup 
con 
con 
wit 
the 
as \ 
Ou 
pla 
OPF 
ass