Browse by author
Lookup NU author(s): Viet Hoang, Dr Meredith Williams, Professor David ManningORCiD
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
This paper presents the initial findings of an investigation into multi-sensor remote sensing as a cost effective means of monitoring desertification in a semi-arid coastal environment. The project aims to develop a means of providing annually updated information at a range of spatial scales for local government and land use planners. A twin scale approach is employed to facilitate mapping at national and local scale. MODIS and ASAR wide swath data provide a generalized assessment for the whole country, whilst ASTER and ENVISAT ASAR image mode imagery are used to investigate desertification problems at a more detailed level. Three parameters were selected to develop a desertification index: land surface temperature, vegetation index, and soil moisture. Land surface temperature was extracted from MODIS and ASTER thermal bands, utilizing standard level 2 data. Vegetation indices extracted from MODIS and ASTER VNIR for evaluation as model inputs included NDVI, SAVI and EVI. Soil moisture is estimated from SAR data following a delta backscatter approach, in which the effects of soil roughness are accounted for by differencing the SAR backscatter from a given image and backscatter from a dry season image (σ – σdry). The relationship between vegetation density, soil moisture, and surface temperature, and the role of these parameters in the desertification process are still under investigation. It has been shown that vegetation index and surface temperature are strongly related to moisture stress and can explain the dynamic of desertification. An index based on relation between vegetation density and surface temperature was tested. The Vegetation Temperature Angle Index (VTAI) measures the angle between the line from a particular pixel to the convergence point and the line defining maximum vegetation temperature limit. The smaller the angle the more extreme the drought level. Soil moisture estimation from delta backscatter (σ – σdry) showed a strong relation with field measurements (r2 = 0.89) for bare land and sparsely vegetated areas. When the vegetation density is higher (NDVI>0.5), the relation is weak (r2 = 0.58) therefore soil moisture estimation is not possible.
Author(s): Hoang VA, Williams M, Manning DAC
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Unknown
Conference Name: Annual Conference of the Remote Sensing & Photogrammetry Society
Year of Conference: 2006
