Remote Sensing Techniques for Soil Moisture and Agricultural Drought Monitoring
Drought is the most complex and least understood of all natural hazards, affecting more people than any other hazard. Soil moisture is a primary indicator for agricultural drought. This dissertation is aimed at evaluating and investigating soil moisture and drought monitoring using remote sensing techniques. Recent technological advances in remote sensing have shown that soil moisture can be measured by a variety of remote sensing techniques, each with its own strengths and weaknesses. This research is designed to combine the strengths of optical/infrared as well as microwave remote sensing approaches for soil moisture estimation. A soil moisture estimation algorithm at moderate resolution was developed based on the well known ‘Universal Triangle’ relation by using MODIS land parameters as well as ground measured soil moisture. Though lower in spatial resolution, AMSR-E microwave measurements provides daily global soil moisture of the top soil layer, which are typically less affected by clouds, making them complementary to MODIS measurements over regions of clouds. Considering that the ‘Universal Triangle’ approach for soil moisture estimation is based on empirical relations which lack solid physical basis, a new physics based drought index, the Normalized Multi-band Drought Index (NMDI) was proposed for monitoring soil and vegetation moisture from space by using one near-infrared (NIR) and two shortwave infrared (SWIR) channels. Typical soil reflectance spectra and satellite acquired canopy reflectances are used to validate the usefulness of NMDI. Its ability for active fire detection has also been investigated using forest fires burning in southern Georgia, USA and southern Greece in 2007. Combining information from multiple NIR and SWIR channels makes NMDI a most promising indicator for drought monitoring and active fire detecting. Given the current technology, satellite remote sensing can only provide soil moisture measurements for the top soil profile, and these near-surface soil moisture must be related to the complete soil moisture profile in the unsaturated zone in order to be useful for hydrologic, climatic and agricultural studies. A new numerical method was presented to solve the governing equation for water transport in unsaturated soil by matching physical and numerical diffusion. By applying a new numerical scheme with which to discrete the kinematic wave equation on the space-time plane, this method shows the capability to simulate the physical diffusion of the diffusive wave with the numerical diffusion generated in the difference solution under certain conditions. Compared with other numerical methods with the first-order finite differences scheme, this method has enhanced the solution precision to the second order. An example application shows a good agreement with the observed data and suggests this new approach can be appropriate for soil moisture profile estimation. By combining the proposed soil moisture and drought estimation techniques, the daily soil moisture profile at high resolution can be gained, and is thus expected to be helpful not only in drought monitoring and active fire detecting, but also in agricultural applications and climate studies.
Remote Sensing, Soil moisture, Agricultural drought, Drought index, Fire detection