Electronic Theses and Dissertations

Date of Award


Document Type


Degree Name

M.S. in Engineering Science

First Advisor

Lance D. Yarbrough

Second Advisor

Gregory L. Easson

Third Advisor

Zahra Ghaffari


University of Mississippi

Relational Format



Remote sensing systems such as multispectral and radar imaging can provide detailed information about soil moisture levels, vegetation cover, and topography. These data can be used to identify areas of high soil moisture, monitor changes in soil moisture over time, and assess the impact of human activities on watersheds. A commonly used system in orbit for monitoring soil moisture is the Soil Moisture Active Passive Mission (SMAP). For SMAP and all spaceborne systems, one of the major limitations for users to implement satellite-based data is the coarse resolution of the pixels (~9 km). Downscaling approaches are introduced by many researchers to overcome the low resolution of the surface soil moisture data. In this project, the random forest approach is used to downscale surface soil moisture derived from SMAP level 4 root zone soil moisture geophysical (SPL4SMGP) data product to a 1-km spatial resolution for a region in northeastern Mississippi. Normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and diurnal temperature are used as independent variables for the random forest model. Due to the lack of fine-resolution surface soil moisture data in the study area, 9 km SPL4SMGP product is disaggregated to a 1km spatial resolution without altering original pixel values and used as the dependent variable for the model training. Field data were collected from 25 locations within the study region for six days throughout the year to validate the downscaling output. While results of the downscaling showed poor correlation to the field-collected surface soil moisture data, land cover types and surface geology showed a good match to the downscaled data. According to the results, Croplands and Cropland/Natural Vegetation Mosaics show the highest surface soil moisture values for any day considered.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.