Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Israel → Detecting litter accumulation and vegetation dynamics with remote sensing techniques in a semi-arid landscape

Ben-Gurion University of the Negev (2011)

Detecting litter accumulation and vegetation dynamics with remote sensing techniques in a semi-arid landscape

Guagnin, Johannes

Titre : Detecting litter accumulation and vegetation dynamics with remote sensing techniques in a semi-arid landscape

Auteur : Guagnin, Johannes

Université de soutenance : Ben-Gurion University of the Negev

Grade : Master of Science (MS) 2011

Résumé
In the semi-arid landscape of the Sayeret Shaked Park in the northern Negev Desert of Israel, shrub patches were observed to be sinks for water, soil, and nutrients that are washed from the surrounding crusted areas. Furthermore, they caused an accumulation of dry organic matter that improves soil hydraulic properties and control nutrient cycling. After several years of drought, most of shrubs in the research area died, and only accumulation of snail shells on dry organic matter remained. The accumulation of litter is assumed to be an important factor for the growth of vegetation because of its ameliorating effects on soil water household and the effects on nutrient cycling. The objectives of the research were to discover organic matter dynamics after shrubs death ; whether the accumulation of organic matter under shrub remains and snail shells have an impact on the erosion of dry organic matter, from the hillsides to the streambed and if there is an impact on the dynamics of new annual and perennial higher vegetation. Images were obtained with a hyperspectral camera at the Sayeret Shaked Park at three different plots. Two plots were located close to the bank of the watershed (Plot A, B) and one (Plot C) in its very streambed. Of these plots, images were obtained at five different dates : At the end of the dry season, in August 2009, after the first rainfalls in November 2009, in the midst of the rainy season, January 2010, and at the end of the wet season in April 2010. Finally, an image was obtained in September 2010, again at the end of the dry season. The images were processed and analyzed for ground coverage with ENVI software. Spectral Angel Mapper was applied to classify the pixels to one of five chosen classes. For that purpose it was distinguish between dry organic matter, green vegetation, snail shells and biological soil crusts in both a younger and an older stage. After classification class statistics were calculated and analyzed for changes. The first plot, plot A, at the northern bank of the watershed, was divided into two parts, one homogenous part above and one heterogeneous part below shrub remains. The two parts of the first plot had a similar trend in accumulation over the year but differences in its dynamic during the period. Plot B shows that there is indeed an increase in dry organic matter in the watershed from the first image in August 2009 to the last image in 2010. Plot C did not show any remarkable changes in litter accumulation. Litter was determined to be the main source for the growth of new vegetation in all the plots. It is concluded, that shrub remains and snail shell accumulations still serve as barrier for erosion and provide accumulation of litter that is necessary for the new growth of perennial and annual higher vegetation.

Mots Clés : Arid regions — Remote sensing. ; Plant litter — Arid regions — Remote sensing. ; Vegetation dynamics — Arid regions — Remote sensing. ; Plant litter — Arid regions — Remote sensing ; Vegetation dynamics — Arid regions — Remote sensing.

Présentation

Version intégrale

Page publiée le 19 mai 2012, mise à jour le 13 septembre 2017