Reflectance and temperature of bare field soil were measured to determine the relationship between soil water content and spectral reflectance and temperature. Reflectance in the six reflective Thematic Mapper (TM) wavebands plus a 1.15 to 1.30 μm waveband (referred to as MMR5) was measured using a ground-based radiometer across a soil water gradient provided by a line source sprinkler system. Temperature was measured using a hand-held infrared thermometer. The results of this study show that TM band 7 reflectance (2.05-2.30 μm) provided improved estimates of surface soil water content (0-0.5 cm depth) over estimates using reflectance information from all seven TM bands. Good correlations were found between band ratio spectral indices of TM5/TM7, MMR5/TM7, and MMR5/TM5 and surface soil water content. A soil temperature normalization utilizing daily values of maximum vapor pressure deficit was found to provide improved estimates of soil water content to depths greater than 0.5 cm over estimates using other temperature normalizations.

Hitherto, most research conducted to monitor agricultural drought on the African continent has focused only on meteorological aspects, with less attention paid to soil moisture, which describes agricultural drought. Satellite missions dedicated to soil moisture monitoring must be used with caution across various scales. The rainfed sector of Sudan takes great importance due to it is high potential to support national food security. El Gedaref state is significant in Sudan given its potentiality of the agricultural sector under a mechanized system, where crop cultivation supports livelihood sources for about 80% of its population and households, directly through agricultural production and indirectly through labor workforce. The state is an essential rainfed region for sorghum production, located within Sudan's Central Clay Plain (CCP). Enhancing soil moisture estimation is key to boosting the understanding of agricultural drought in the farming lands of Sudan. Soil moisture measuring stations/sensors networks do not exist in the El Gedaref agricultural rainfed sector. The literature shows a significant gap in whether soil moisture is sufficient to meet the estimated water demands of cultivation or the start of the growing season. The purpose of this study is to focus principally on agricultural drought. The soil moisture data retrieved from the Soil Moisture Active Passive (SMAP) mission launched by NASA in 2015 were compared against in situ data measurements over the agricultural lands. In situ points (at 5 cm, 10 cm, and 20 cm depths) corresponding to 9×9 km SMAP pixel foot-print are rescaled to conduct a point-to-pixel evaluation of SMAP product over two locations, namely Samsam and Kilo-6, during the rainy season 2018. Four errors were measured; Root Mean Squared Error (RMSE), Mean Bias Error (MBE), unbiased RMSE (ubRMSE), Mean Absolute Bias Error (MABE), and the coefficient of determination R2. SMAP improve (significantly at the 5% level for SM). The results indicated that the SMAP product meets its soil moisture accuracy requirement at the top 5 cm and in the root zone (10 and 20 cm) depths at Samsam and Kilo-6. SMAP demonstrates higher performance indicated by the high R2 (0.96, 0.88, and 0.97) and (0.85, 0.94, and 0.94) over Samsam and Kilo-6, respectively, and met its accuracy targeted by SMAP retrieval domain at ubRMSE 0.04 m3m-3 or better in all locations, and most minor errors (MBE, MABE, and RMSE). The possibility of using SMAP products was discussed to measure agricultural drought and its impacts on crop growth during various growth stages in both locations and over the CCP entirely. The croplands of El Gedaref are located within the tropical savanna (AW, categorization following the Köppen climate classification), warm semi-arid climate (BSh), and warm desert climate (BWh). The areas of interest are predominantly rainfed agricultural lands, vulnerable to climate change and variability. The Climate Hazards Group Infrared Precipitation with Station data (CHIRPS), SMAP at the top surface of the soil and the root zone, and Soil Water Deficit Index (SWDI) derived from SMAP were analyzed against the Normalized Difference Vegetation Index (NDVI). The results indicate that the NDVI val-ues disagree with rainfall patterns at the dekadal scale. At all isohyets, SWDI in the root zone shows a reliable and expected response of capturing seasonal dynamics concerning the vegetation index (NDVI) over warm desert climates during 2015, 2016, 2017, 2018, and 2019, respectively. It is concluded that SWDI can be used to monitor agricultural drought better than rainfall data and SMAP data because it deals directly with the available water content of the crops. SWDI monitoring agricultural drought is a promising method for early drought warning, which can be used for agricultural drought risk management in semi-arid climates. The comparison between sorghum yield and the spatially distributed water balance model was assessed according to the length of the growing period. Late maturing (120 days), medium maturing (90-95 days), and early maturing variety (80-85 days). As a straightforward crop water deficit model. An adapted WRSI index was developed to characterize the effect of using different climatic and soil moisture remote sensing input datasets, such as CHIRPS rainfall, SMAP soil moisture at the top 5 cm and the root zone, MODIS actual evapotranspiration on key WRSI index parameters and outputs. Results from the analyses indicated that SMAP best captures season onset and length of the growing period, which are critical for the WRSI index. In addition, short-, medium-, and long-term sorghum cultivar planting scenarios were con-sidered and simulated. It was found that over half of the variability in yield is explained by water stress when the SMAP at root zone dataset is used in the WRSI model (R2=0.59–0.72 for sorghum varieties of 90–120 days growing length). Overall, CHIRPS and SMAP root zone show the highest skill (R2=0.53–0.64 and 0.54–0.56, respectively) in capturing state-level crop yield losses related to seasonal soil moisture deficit, which is critical for drought early warning and agrometeorological risk applications. The results of this study are important and valuable in supporting the continued development and improvement of satellite-based soil moisture sensing to produce higher accuracy soil moisture products in semi-arid regions. The results also highlight the growing awareness among various stakeholders of the impact of drought on crop production and the need to scale up adaptation measures to mitigate the adverse effects of drought.

Graduation date: 1992

This study compared two remote sensing water indices: the Normalized Difference Water Index (NDWI) and the Modified NDWI (MNDWI). Both indices were calculated using publically-available data from the Landsat 8 Operational Land Imager (OLI). The research goal was to determine whether the indices are effective in locating open water and measuring surface soil moisture. To demonstrate the application of water indices, analysis was conducted for freshwater wetlands in the Tualatin River Basin in northwestern Oregon to estimate hydrologic connectivity and hydrological permanence between these wetlands and nearby water bodies. Remote sensing techniques have been used to study wetlands in recent decades; however, scientific studies have rarely addressed hydrologic connectivity and hydrologic permanence, in spite of the documented importance of these properties. Research steps were designed to be straightforward for easy repeatability: 1) locate sample sites, 2) predict wetness with water indices, 3) estimate wetness with soil samples from the field, 4) validate the index predictions against the soil samples from the field, and 5) in the demonstration step, estimate hydrologic connectivity and hydrological permanence. Results indicate that both indices predicted the presence of large, open water features with clarity; that dry conditions were predicted by MNDWI with more subtle differentiation; and that NDWI results seem more sensitive to sites with vegetation. Use of this low-cost method to discover patterns of surface moisture in the landscape could directly improve the ability to manage wetland environments.