Grazing by sheep and goats plays an important role in nearly all study areas investigated during the EPRODESERT project. Browsing as well as treading impede the regeneration of vegetation on many grazed areas in the Ebro Basin and the Pyrenees, and sheep trails encourage the development of erosion rills by exceptionally high runoff and erosion rates (Molinillo et al., 1997).
The Arnas catchment, located in the Upper Aragon River Basin of the Spanish Pyrenees, was cultivated totally with cereal until the middle of the twentieth century. Since its abandonment, large parts of the catchment were affected by a process of natural plant colonization with matorral composed of Genista scorpius, Buxus sempervirens, and Rosa gr. canina (García-Ruiz et al., 2005). The field covered by this image map (Fig. 16-6; see also Figs. 16-3D and 10-20) was abandoned in the late 1970s and subsequently used as sheep pasture. Close to the monitoring test site, a grazing control cage was installed in 1996. In the following years, plant species, vegetation cover and height were recorded in regular intervals both within the cage and on the neighboring reference sites (Ries et al., 2003, 2004). The test site as well as the grazing control sites were monitored with aerial photography, and vegetation cover maps were prepared for all sites. The small image ground sample distance (GSD), the presence of the cage grating and the high degree of shadowing between G. scorpius shrubs prevented the use of fully automatic image classification methods (see Chapter 11.5), and a hybrid method of multispectral thresholding and manual mapping had to be employed for the maps of the control sites (Fig. 16-7).
Results show that the exclusion of sheep from the area is able to boost vegetation cover significantly. While the reference-site cover is stable between 50% and 54%, vegetation cover within the cage increased from 60% to 92% between July 1996 and August 1998. The development of individual vegetation cover classes is equally interesting: 53% (cage) and 74% (reference site) of the areas initially show <60% cover by grass and herbs—the value shown to be an important threshold for reducing erosion processes from the aerial test-site monitoring. The percentage of these classes fell dramatically to 4% in the cage and at the same time increased slightly to 80% on the grazed reference site. At the time of writing, the control cage was completely filled with dense shrub growing out between the grating (Seeger 8/2009, pers. comm.).
Transferring the recovering rate of the vegetation from the control cage to the nearby monitoring test site showed that exclusion of sheep from the catchment would lead to nearly closed vegetation cover within only two years (Fig. 16-8; Ries et al., 2000). Interception by the protecting shrub canopy of G. scorpius matorral would reduce erosion rates even on the bare areas of the old trails beneath, where vegetation regeneration can be expected to be slower. In this study, high-resolution SFAP was able to capture patterns of vegetation exactly on the scale level where sheep have short-term influence on plant distribution.
Was this article helpful?