The lower Mississippi River alluvial valley (LMRAV) in northwestern Mississippi is characterized by a flat landscape and predominantly agricultural land use. The fluctuations in surface heat flux throughout the crop cycle due to land cover modifications are thought to have an impact on the regional weather. This research analyzes changes in convective patterns over the LMRAV based on the rapid variations in land cover as a result of the seasonal harvest cycle. Focusing on synoptically weak days between 2012-2016, data from the GOES 13-15 satellite visible imagers were used due to their 1-km spatial resolution and ability to distinguish lower clouds over a warm surface. By comparing the spatial and temporal patterns of convective clouds, the study confirmed that convective patterns do change based on land cover evolution resulting from the harvest cycle. These changes were likely a result of low-level thermal and moisture changes resulting from variations in evapotranspiration.

Classification of remotely sensed imagery for reliable land use and land cover (LULC) change information remains a challenge in areas where spectrally similar LULC features occur. Dissolved organic matter (DOM) influences the biogeochemistry of aquatic environments and its quantity and quality are due, in large part, to the surrounding LULC. Thus, objectives were to improve the accuracy of LULC classification and quantify seasonal variations of water quality in a watershed dominated by agriculture and determine the controls for the variations in water quality. Support vector machine classification scheme with post classification correction yielded highest accuracy for LULC classifications and four distinct DOM components were found that changed seasonally and were controlled by hydrology and LULC. The microbial component was the main fraction of the DOM pool due in large part to agricultural practices. This DOM can influence the water quality significantly as it moves downstream and causes increased biological activity.

No description available.

No description available.

No description available.

No description available.

Construction of land use and land cover (LULC) maps was accomplished through the use of remote sensing and GIS. Remote sensing and GIS were used to classify 1990 Landsat 5 and 2000 Landsat 7 Mississippi Gulf Coast imagery into six LULC classes: urban, barren, forested vegetation, non-forested vegetation, marsh, and water. An accuracy assessment was performed on the 2000 LULC map to determine the reliability of the map. Finally, GIS software was used to quantify and illustrate the various LULC conversions that took place over the ten year span of time. The paper concludes that remote sensing and GIS can be used to create LULC maps. It also notes that the maps generated can be used to delineate the changes that take place over time.

The land surface and atmosphere interact through complex feedback loops that link energy and water cycles. Effectively characterizing these linkages is critical to modeling weather and climate extremes accurately. Seasonal variability in vegetation growth and human-driven land cover changes (LCC) can alter the biophysical properties of the land surface, which can in turn influence the water cycle. We quantified the impacts of seasonal variability in vegetation growth on land surface energy and water balances using ecosystem-scale eddy covariance and large aperture scintillometer observations. Our results indicated that the monthly precipitation and seasonal vegetation characteristics such as leaf area index, root length, and stomatal resistance are the main factors influencing ecosystem land surface energy and water balances when soil moisture and available energy are not limited. Using a regional-scale climate model, we examined the effect of LCC and irrigation on summer water cycle characteristics. Changes in biophysical properties due to LCC reducing the evapotranspiration, atmospheric moisture, and summer precipitation over the contiguous United States (CONUS). The combined effects of LCC and irrigation indicated a significant drying over the CONUS, with increased duration and decreased intensity of dry spells, and reduced duration, frequency, and intensity of wet spells. Irrigated cropland areas will become drier due to the added effect of low-precipitation wet spells and long periods (3-4% increase) of dry days, whereas rainfed croplands are characterized by intense (1-5% increase), short-duration wet spells and long periods of dry days. An analysis based on future climate change projections indicated that 3–4 °C of warming and an intensified water cycle will occur over the CONUS by the end of the 21st century. The results of this study highlighted the importance of the accurate representation of seasonal vegetation changes and LCC while forecasting present and future climate. / Doctor of Philosophy / The land surface and atmosphere interact through complex feedback loops that link energy and water cycles. Effectively characterizing these linkages is critical to accurately model weather and climate extremes. We quantified the influence of human-driven land cover change (LCC), in this case, LCC associated with irrigated agriculture, and seasonal vegetation growth on the water cycle using a regional climate model and ecosystem-level observations. Our results indicated that monthly precipitation and seasonal vegetation growth are the main factors influencing land surface energy and water balances when soil moisture and solar energy are not limited. Our results showed that irrigation-related LCC reduced summer precipitation over the contiguous United States (CONUS), with an increased number of dry days (days with less than 1 mm precipitation) and reduced hourly, daily, and summer precipitation totals. Irrigated cropland areas are becoming drier due to the combined effects of low precipitation and long dry days, whereas rainfed croplands are characterized by intense short-duration precipitation and long dry days. Climate change analyses indicate that 3–4 °C of warming and an intensified water cycle will occur over the CONUS by the end of the 21st century. The results of this study highlight the importance of the accurate representation of LCC while forecasting future climate.

Previous studies on the effects of land cover influence on stream nitrogen have focused on base flow conditions or were conducted specifically within urbanized or primarily agricultural watersheds. While these studies have shown relationships between land cover and nitrogen, this relationship and the scale of influence could change during storms. The purpose of my study was to understand how land cover influences nitrogen in streams during storms. This was address using nine watersheds within the Little Tennessee Basin in North Carolina. While this basin is primarily forested, the nine watersheds have mixed agricultural, built, and forest land cover. Land cover influences were addressed through nitrogen concentration/discharge patterns, nitrogen concentration relationship to land cover, and comparison of storm and base flow nitrogen concentrations over time. Weekly base flow samples and samples from six storm were collected in 2010-2011. Total dissolved nitrogen (TDN), nitrate (NO??), dissolved organic nitrogen (DON), and ammonium (NH?⁺) concentrations were compared among sites. During most storms, DON peaked before the peak of the discharge while NO?? peaked after the peak of the storm. This suggest that DON could be coming from a near stream source or surface runoff while NO?? could be from longer pathways such as subsurface flow or from sources further away on the watershed. NO?? concentration varied among sites, while DON concentration varied more between base flow and storm samples. Examining the different landscape scales from 200-m local corridor, 200-m stream corridor, and entire watershed, watershed land cover was the best predictor for all the nitrogen concentrations. Agricultural and built combined best predicted TDN and NO??, while agricultural land cover was a better predictor of DON. For storms, nitrogen concentrations did not show seasonal patterns but was more related to discharge. Nitrogen concentration increased with discharge during storms and the more intense and longer storms had higher TDN and NO?? concentrations. However, conflicting seasonal trends were seen in monthly base flow. The more forested watersheds had high NO?? during the summer and low NO?? in the winter. For sites with higher NO??, the seasonality was reversed, with higher winter NO?? concentration. The least forested site had relatively constant nitrogen through the year at base flow and concentration decreased for most storms. Further studies on storms and nitrogen transport are needed to understand better the seasonal patterns of nitrogen input during storms. / Master of Science

This study used two modelling approaches to predict future urban landscape for the Baltimore-Washington metropolitan areas. In the first approach, we implemented traditional SLEUTH urban simulation model by using publicly available and locally-developed land cover and transportation data. Historical land cover data from 1996, 2001, 2006, and 2011 were used to calibrate SLEUTH model and predict urban growth from 2011 to 2070. SLEUTH model achieved 94.9% of overall accuracy for a validation year of 2014. For the second modelling approach, we predicted future county-level population (e.g., 2050) using historical population data and time-series forecasting. We then used future population projection of 2050, aided by strong population-imperviousness statistical relationship (R2, 0.78-0.86), to predict total impervious surface area for each county. These population-predicted total impervious surface areas were compared to SLEUTH model output, at the county-aggregated spatial scale. For most counties, SLEUTH generated substantially higher number of impervious pixels. An annual urban growth rate of 6.24% for SLEUTH model was much higher than the population-based approach (1.33%), suggesting a large discrepancy between these two modelling approaches. The SLEUTH simulation model, although achieved high accuracy for 2014 validation, may have over-predicted urban growth for our study area. For population-predicted impervious surface area, we further developed a lookup table approach to integrate SLEUTH out and generated spatially explicit urban map for 2050. This lookup table approach has high potential to integrate population-predicted and SLEUTH-predicted urban landscape, especially when future population can be predicted with reasonable accuracy. / Master of Science