| Titre : | Inventory and evaluation of pastures in Albania (2010) |
| Titre original: | Inventarizimi dhe vlerësimi i kullotave në Shqipëri |
| Auteurs : | E. Pasho |
| Type de document : | Article |
| Dans : | Albanian journal of agricultural science (vol. 9, n. 1, 01/01/2010) |
| Article en page(s) : | p. 79-86 |
| Langues : | Albanais |
| Langues du résumé : | Anglais ; Albanais |
| Catégories : |
Thésaurus IAMM PATURAGES ; TERRE DE PATURAGE ; DEGRADATION ; TELEDETECTION ; SYSTEME D'INFORMATION GEOGRAPHIQUE ; ALBANIECatégories principales 06 - AGRICULTURE. FORÊTS. PÊCHES ; 6.4 - Production Agricole. Système de Production |
| Résumé : | Pastures are an important resource in Albania but they were irrationally used in the past during the communist period leading to their present degradation. In this paper, a variety of abiotic, biotic and management variables were explored for their relative contribution to this degradation after implementing a national inventory with remote sensing and GIS techniques. Abiotic variables included altitude, slope, aspect and erosion status; The first three variables were derived by processing the digital elevation model of Albania (DEM) in Arc Map and classified into a number of classes. The latter variable was produced using a Multi Criteria Evaluation (MCE) decision rule which consisted of integrating several thematic maps such as soil type, precipitation, cover components, slope and aspects. Biotic variables included NDVI and ground cover (produced in Erdas Imagine); The NDVI was derived based on the ratio between the reflectance of the red (R) and near infrared (IR) spectral bands (NDVI = (NIR – RED) / (NIR + RED). Cover components of pastures were mapped by implementing a methogology which consisted on a combination of three spectral analysis of Landsat ETM+ images: NDVI, Principal Component Analysis and Supervised Classification. NDVI was performed with the purpose of enhancing the differences among pixels allocated to vegetation and non vegetation components of pastures, which were represented by a range of positive and negative values respectively (-1 ,+1). Principal Component Analysis was applied on the images with the purpose of improving data interpretability and enhancement of several features over the pasture areas. Supervised Classification was applied on the best combination of principal components by clustering pixels (representing homogeneous areas) in spectral classes. Management variables involved distances from roads, villages and watering points. All these variables were derived in Arc Map using “Calculate distances” tool. The base data for the study were collected in Albania and analysed in Greece (Crete), in 2007. It was found that the majority of pastures are distributed on a mountainous landscape with steep to very steep (79%) and south facing slopes (44%) and high level of erosion (30%) suggesting highly unfavorable abiotic conditions. This is reflected in the relative low vegetative cover (53%) and the relatively high cover with rocks (24%) and bare soil (23%). For the management variables, it was found that most pastures are distributed close to roads (67%), watering points (72%) and in a medium distance from villages (51%). It is concluded that degradation of pastures in Albania should be attributed to the interaction of unfavourable abiotic conditions and irrational grazing management practices. |
| Cote : | Réservé lecteur CIHEAM |
Accueil
Sélection de la langue
Adresse
ÉQUIPE
Nous contacter
Informations pratiques
contact
