University of Twente Student Theses


Impact of Tibetan land surface conditions on ASAR backscattering

Asnake, Seife Gedamu (2011) Impact of Tibetan land surface conditions on ASAR backscattering.

[img] PDF
Abstract:Tibetan plateau is the highest plateau in the world and is characterized by mountainous ranges with the average altitude of more than 6000 meter above sea level. Understanding the land surface-atmosphere interactions on Tibetan plateau is critical to better understand the Asian monsoon system and its link to climate change. This study analyse the response of ASAR WSM backscattering in relation to variations in land surface conditions over Tibetan plateau. The land surface states analysed are soil moisture, soil temperature and vegetation changes. The study focuses on Maqu Catchment which is located in the north-eastern edge of Tibetan plateau. Grassland, wetland, and hill slope geographical regions are representative for the study area. Soil moisture and temperature data collected in the period 2008 to 2009 were obtained from soil moisture measurement network installed in Maqu catchment and precipitation data were collected from Maqu meteorological station. SPOT NDVI product was used to analyse the effect of vegetation on ASAR WSM backscattering. ASAR WSM images have been processed using ENVI and IDL programming. The average ASAR WSM backscattering coefficient from homogeneous areas (3 × 3 pixel square window) surrounding each soil moisture stations were extracted. The temporal ASAR WSM backscattering signature over grassland wetland and hill slope shows similar trend. The lowest backscattering value was found during winter season. The backscattering increases from winter to summer season when the monsoon sets in and frozen soil water melts. Compared to the grassland and hill slope stations, wetland stations show the lowest backscattering signature during winter time. The seasonal variation, particularly, the frozen condition affects the backscattering. The spatial profile at three transects across yellow river during winter season shows high backscattering signature. High correlation exists between soil moisture and ASAR backscattering (R2 up to 0.88) and this suggests the backscattering response was sensitive to soil moisture. Towards the summer the area vegetation cover (NDVI) increases gradually and reaches maximum when the rainfall is at the peak intensity. The soil moisture and soil temperature declines during winter time. The backscattering coefficient increases in 10 dB towards summer season. The increase in backscattering signature follows the monsoon rainfall, the soil moisture state and vegetation biomass change. The correlation between the NDVI of grassland stations versus the mean NDVI was high (R2 greater than 0.9) and the root mean square difference (RMSD) and bias was low (approximately 0.05). The high coefficient of determination (R2 ) and low RMSD and bias suggest the homogeneity of the grassland cover. The correlation between backscattering and NDVI in all grassland, wetland and hill slope sites was low. This shows the effect of vegetation on backscattering was minimum compared to soil moisture. The result of this study show that ASAR WSM data can be used to monitor variations in land surface conditions on Tibetan grassland, wetland and hill slope areas, particularly sites with homogeneous grass cover. In addition, ASAR WSM observation could be a good choice for estimating soil moisture and monitoring the effect of seasonal climatic variation on land surface state such as freezing/thawing condition for hydrological modelling and integrated watershed management. Keywords: ASAR WSM, backscattering, soil moisture, soil temperature, NDVI, grassland, wetland, hill slope, ENVI/IDL
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page