Article: Temporal evolution of vegetation activity by multitemporal Landsat imagery

We analyzed the temporal evolution of vegetation activity on various land cover classes in the Spanish Pyrenees for the period 1984 to 2007.
Two time series of the normalized difference vegetation index (NDVI) were used, corresponding to March (early spring) and August (end of summer). The series were generated from Landsat TM and Landsat ETM+ images, which were subject to a process of atmospheric correction using a radiative transfer model (6S) and non-Lambertian topographic correction to avoid differences in the illumination conditions, and a relative normalization between dates using an automated process.

A statistically significant increase was found in the March NDVI for vegetated areas, and the opposite trend was found in both March and August for degraded areas (badlands and erosion risk areas). 

Temporal evolution of average March and August NDVI values for various land cover classes, and temporal evolution of driving climatic factors.

Multivariate regression analysis showed that a progressive increase in minimum temperatures and the inter-annual variation of the cloud cover were the most important factors explaining the increase of the NDVI in the vegetated areas, especially in March. In the degraded areas, however, no climatic or topographic variables were associated with the negative NDVI trend, so it may be related to active erosion processes taking place in these regions.

The article has been published in Earth Surface Processes and Landforms, a leading international journal in the field of Geosciencies, published by John Wiley & Sons, Ltd.

Reference: Alatorre L.C., Beguería S., Vicente-Serrano S. (2011) Evolution of vegetation activity on well covered, eroded, erosion risk areas by multitemporal Landsat TM imagery, central Spanish Pyrenees. Earth Surface, Processes, Landforms 36(35), 309319.

Sin comentarios

Deja una respuesta

Tu email nunca será compartido con nadie.Los campos obligatorios están marcados con *