563
In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B
HYPERSPECTRAL EVALUATION OF THE PEAR TREES ON THE
BASIS OF THE GENETIC COLLECTION OF THE DIFFERENT SPECIES
J. Tamas a , Z. Szabo b
a University of Debrecen, Water and Environmental Management, Boszormenyi 138., 4032, Debrecen, Hungary -
tamas@gisserverl .date.hu
b University of Debrecen, Department of Horticulture, Boszormenyi 138., 4032, Debrecen, Hungary-
zszabo@agr.unideb.hu
KEY WORDS: Hyper spectral, Agriculture, Precision, Sustainable, Vegetation, Land Use, GIS, High resolution
ABSTRACT:
The principle task of the sustainable development is the preservation of the genetic variety, which is similar challenge in the
horticulture regarding the sublimation of fruit species. The breeders of the traditional fruit strains give stock to the sustenance
diversity of the agri-environment on the species and landscape level. In 2009, hyperspectral images have been taken by AISA Dual
sensors from the pear gene pool in Ujfeherto, Hungary. The hyperspectral data cube (in the wavelength range of 400-2500 nm, with
1.5 m ground resolution) ensured possibility to make the spectral library of pear species. In the course of the simultaneously field
work the spatial position and individual extent of all pear trees was defined to set up a detailed GIS data base. The water stress
sensitivity of single species and the descriptive spectral curves were determined with common evaluation of the spectral and spatial
data. Based on the unique methodology processing and the hyperspectral data base suitable strains can be chosen for agri
environment and let take adaptive stocks regarding climate change into the genetic grafting work. Furthermore we could determine
and map the sparsely species in the region with the help of the hyperspectral data.
1. INTRODUCTION
On the basis of the United Nations Earth Summit in Rio de
Janeiro in 1992 the biodiversity conversation is the important
task for every country. Even so numerous species is
disappearing off the earth decreasing Earth’s biodiversity.
Plant and animal species collected by human genetic breeding
works and selected over more hundred years are less
emphasized contrary to wild species.
Our experimental respects the pear species (Pyrus communis
L.). The pear is the second more important fruit in temperate
zone following apple worldwide. It produces 4-4.5 million
tonnes per year in Europe especially in Italy, France and Spain,
while the pear is the fourth important fruit in Hungary with the
amount of 40000-80000 tonnes.
There are numerous species of these, which can only be found
in a certain landscape. These species mean irreparable genetic
sources same as wild and rare species. There are innumerable
genetically encoded qualities - eg. resistivity against the
diseases and frost, drought tolerance, unique colour and taste,
and storage tolerance, which lack partly or in gross from the
commercial species. Because of the continually changing
consumer demand, and environmental and landscape aspect the
preservation of every pear species having special or useful
quality is necessary.
The countries maintain this particular genetic property in the
special plantations - gene pool. In Hungary these examination
plantation can be found in 4 different production site - Cegled,
Erd, Fertod, Ujfeherto, where 7833 trees are cultivated with
professional technology and high cost. The items of the genetic
collections are fully documented and the data are registered by
international norm. The recorded data ensure the identification,
central registration, suitable access to provide using for the
national or international researchers, breeders and teachers.
The greater part of data was in analogous form and the smaller
part of it was in digital form until 2008. The most important
cause of the establishment of the geoinformatical system was
the dismembered database and the difficult management or
actualizing.
The aim was to create a new GIS - Remote Sensing data
system, which can be the basis of the later agri-ecological
decision support system.
2. MATERIAL AND METHODS
The Fruit growing Research and Consultant non-profit company
created the genetic collection of pomum fruits ( apple, pear,
quince, naseberry) at Ujfeherto (47° 49’ 30” N; 21° 40’ 27” E)
in 1980, which recorded 1876 trees in 2009. The area of the
research company is situated in north direction from the town of
Ujfeherto, in Nyirseg landscape NE Hungary. It consists of
fixed sand dunes accordingly the relief is heterogeneous. The
total cultivated area (fruit plantations and arable land) of the
research company is approximately 3 km 2 .
In the 5 ha plantation of genetic collection it can be found 673
apple, 480 pear, 57 quince and 28 naseberry species, variety and
hybrid at present. The pear plantation is situated in the east part
of the gene pool covered by 28000 m 2 , where the planting
started in 1982. According to the traditional cultivation methods
the row pace is 8 m and the space between the trees is 2 m.
In course of the establishment of the GIS logical model we
determined the types of the accessible data and the forms of
utilization in the geoinformatics system. The fruit tree was the
smallest unique entity, which has individually pollable
attributes (name of species, planting date, height, condition etc.)
It was also important to ensure the uploading of the
phenological data such us blossoming, ripening time, and
