<p> The population and industrial complexes of Choctaw County obtains much of its water from an aquifer system in the Tertiary age Wilcox unit of the Mississippi Embayment. Utilizing 20 years of physical chemistry (P-Chem) analysis, potentiometric groundwater records of Choctaw County public water wells as well as industrial P-Chem analysis and surface and ground water level records from an industrial complex, this study examined the changes to the hydrosphere that has taken place since 1995. Analysis of the hydrosphere shows that over the last 20 years, there has been a drop in the potentiometric surface of the Wilcox aquifer system. The analysis also shows changes in the P-Chem of the hydrosphere, changes such as a decrease in the concentration of free CO2 and chloride, and fluctuations of Alkalinity. Comparisons between groundwater records taken from the industrial complex and other locations around Choctaw County, show little variation in the physical chemistry.</p>

<p> Groundwater is an essential input to agriculture world-wide, but it is clear that current rates of groundwater use are unsustainable in the long term. This dissertation assesses both current use of groundwater for country- to global-scale agriculture, and looks at the future of groundwater. The focus is on 1) quantifying food directly produced as a result of groundwater use across spatially-varying agricultural systems, 2) projecting future groundwater demands with consideration of climate change and human decision-making, and 3) understanding the system dynamics of groundwater re-use through surface water systems. All three are addressed using a process-based model designed to simulate both natural and human-impacted water systems. </p><p> Irrigation can significantly increase crop production. Chapter 1 combines a hydrology model (WBM) with a crop model to quantify current crop production that is directly attributed to groundwater irrigation in China. Unsustainably-sourced groundwater &mdash; defined as groundwater extracted in excess of recharge &mdash; accounted for a quarter of China&rsquo;s crop production, and had significant spatial variability. Climate variability and groundwater demand magnified one another in hot and dry years, causing increased irrigation demand at the same time as limited surface water supplies. </p><p> Human decisions about water resource management can impact both the demand and sustainability of groundwater use. Chapter 2 takes an interdisciplinary approach to projecting India&rsquo;s future (to 2050) groundwater demands, combining hydrology and econometric modeling. The econometric model projects how humans make decisions to expand or contract the irrigated land area of crops in response to climate change. Even in areas with precipitation increases, human decisions to expand irrigated areas led to increasing demands for groundwater. We additionally assessed the potential impact of a large water infrastructure project to alleviate groundwater demands in India, and found that maximum alleviation (up to 16%) was dependent upon the storage volume and location of new reservoirs. </p><p> One proposed method for reducing the world&rsquo;s demand for groundwater is to increase the efficiency of agricultural water use. However, these same inefficiencies cause a portion of extracted groundwater to enter surface water systems; it can then be reused, creating a complex system in which groundwater demand does not linearly decline with increased water use efficiency. Chapter 3 quantifies the amount of groundwater that enters surface water systems, the number of times this water is reused for agriculture, and the minimum amount of groundwater required by current agricultural systems in the hypothetical scenario of perfect irrigation efficiency.</p>

<p> The goal of this study is to characterize spatial variability of snow water equivalent (SWE) at the meter scale. The study includes measurement of SWE with a new pressure sensor and use of meteorological sensor data to investigate physical properties within the snowpack that can affect sensor measurement. The new sensor, which can continuously measure a load equivalent to up to 5.5 meters of snow, is designed to be smaller and less expensive (&lt; $1,500) than traditional pressure sensors (> $10,000). Manual snow cores and detailed snow pit analyses were performed to assess accuracy of the sensors and identify physical properties that may be related to sensor measurement error. SWE sensor response and accuracy were assessed between sensors and through comparison with bulk precipitation gage, manual SWE measurements, and snow pillows. SWE sensor readings compared favorably to other measurement methods, particularly in early and peak season. Spatial variability of SWE during the melt season of the two low-snow years during the study period confounded our ability to compare multiple sensor readings for validation. Spatial variability of SWE at study sites was calculated from sets of manual SWE measurements. The correlation length of 80 cm, determined using semi-variograms, highlights the small scale variability in SWE. Statistical resampling of manual measurements suggests that a minimum of ten manual measurements are needed to get within 10% of the spatial average of SWE. Although SWE can remain relatively stable during the melt period, this can be a result of increased density with decreasing snow depth, suggesting that simple inferences about SWE from depth measurements are not appropriate.</p>

<p> With future climate change and increased water demand and scarcity in the Colorado River Basin, the Bureau of Reclamation estimates that the costs of salinity damage will increase for Colorado River users and will exacerbate the current salinity challenges. This study focuses on saline and sodic soils associated with the Mancos Shale formation in order to investigate the mechanisms driving sediment and salinity loads in the Price-San Rafael River Basin of the upper Colorado River. A Walnut Gulch rainfall simulator was operated with a variety of slope angles and rainfall intensities at two field sites (Price, Dry-X) near Price, Utah in order to evaluate how the amount and spatial distribution of vegetation affects salinity in runoff. For each simulated rainfall event, the time-varying concentrations of major cations, anions, and sediment in runoff were measured. Principal component analysis revealed that the two field sites are generally different in runoff water chemistry and soil chemistry, likely due to the difference in parent material and soil indicative of their location on different geologic members. The Dry-X site also has substantially greater total dissolved solids (TDS) and sediment in runoff, soil sodium absorption ratio (SAR), and soil cation exchange capacity (CEC) than the Price site. Despite these differences, a consistent positive linear relationship between the plot-averaged sediment and TDS concentration was found across both sites. The Rangeland Hydrology Erosion Model (RHEM) was calibrated to provide unbiased estimates of sediment in runoff from 23 runs of the rainfall simulator. RHEM simulated the plot-plot variability best at Dry-X compared to Price. Sensitivity analysis of the RHEM input parameters showed that the splash and sheet erodibility coefficient (Kss) and the effective saturated conductivity coefficient (Ke) had the largest influence on the model&rsquo;s sediment and discharge outputs, respectively. The regression that predicted TDS concentration from sediment was applied to RHEM outputs to show that the model could be used to provide salinity estimates for different storm intensities on this part of the Mancos Shale. The potential influence of vegetation canopy cover on sediment production from these two sites was inferred by running RHEM with canopy cover values ranging from 0% to 100%. This changed sediment output by 111% to -91% relative to the present vegetation cover. Measures of the geometry of soil and vegetation patches at Dry-X, such as fractal dimension index and proximity index, showed a relationship to error residuals from RHEM. As the vegetation becomes less isolated, more uniform, and the tortuosity of the bare soil area increases, observed sediment decreases relative to RHEM predictions. The results of this study will help land management agencies assess the feasibility of mitigation strategies for reducing sediment and salinity loads from the saline and sodic soils of the Mancos Shale formation and indicate a possible benefit to incorporating the parameters that describe the spatial pattern of vegetation in RHEM.</p>

<p>Southwest Foods in Lafayette, Louisiana has multiple underground storage tanks that leaked petroleum products over a period of time, releasing numerous toxic constituents of concern into the surrounding soil and groundwater. During the assessment phase of risk evaluation, the Synthetic Precipitation Leaching Procedure (SPLP) test was used to measure the leaching potential of possibly contaminated soils and lithologies when exposed to acidic rainfall. However, during the site investigation, SPLP data produced erratic results that apparently are related to the type of geologic material (?soil?) involved. These erratic results could have major implications for assessment, remediation, and the establishment of Risk Evaluation/Corrective Action Program (RECAP) standards in the state of Louisiana. Incorrectly applied RECAP standards could ultimately mean contaminants of concern could reach points of exposure (POEs), and potentially harm nearby residents and the environment. Previously, lithological composition had not been considered when evaluating the environmental impact of the test. At the site, numerous geotechnical soil boring logs were completed, and soil samples were sent to analytical laboratories for analysis of contamination. Soil analytical data compiled by the laboratories along with samples acquired from multiple groundwater monitoring wells at the site provided an in-depth view into the degree of contamination at Southwest Foods and produced the first signs of erratic SPLP test results. Contour maps of contamination and groundwater potentiometric levels were created to show both the level of contaminants in soil and the possible migration paths of affected groundwater. The SPLP sample data were evaluated and categorized into lithological subgroups for further examination. The lithologies, or soil types (with USCS symbols), that produced erratic results (i.e., some ?passed? and some ?failed?) for the SPLP analyses included ?clay, high plasticity? (CH) and ?clayey silt/sandy silt/silt? (ML). The results for ?silty clay/sandy clay? (CL) were more consistent. The inconsistent SPLP data could result in incorrect application of site-specific RECAP standards and the subsequent cleanup at Southwest Foods and conceivably at other contaminated sites as well.

<p> Aquifer mergence zones are erosional unconformities that hydraulically join interlaying aquifers. In the East Newport Mesa in Orange County, Southern California, aquifer mergence zones may provide a pathway for potentially impaired low-quality groundwater of the shallow, semi-perched aquifer to migrate into the underlying regional, potable, confined aquifers. Major ion and stable isotope results imply that vertical mixing is occurring locally between the discrete shallow and deep groundwater endmembers. Vertical mixing is suggested by anomalously young radiocarbon age-dates of deep groundwater units. Radon-222 results show that shallow groundwater is discharging from the mesa, and is also actively intruded by surface water. Mixing of shallow and deep waters at these mergence zones may pose a threat to the deeper regional aquifer system. </p><p>

<p> Hyperspectral and polarimetric imaging of the ocean, both below and above the water surface, has increased the interest of the Ocean Color (OC) scientific community for decades in an attempt to answer questions related to climate change, monitoring of water quality, assessment of the impact of anthropogenic activities on marine life and underwater ecosystems, detection and characterization of underwater targets. These needs are recognized by worldwide efforts devoted to acquiring accurate time series measurements in open-ocean and coastal waters by OC satellite imagery to produce reliable high-quality data records, which can be used both in support of operations and in climate studies. The reliability of satellite observations of the open-ocean and coastal zones requires these remote instruments to be regularly assessed and validated against actual in-situ measurements along with related atmospheric corrections. However, despite improvements in satellite data and retrieval algorithms, accurate retrievals in coastal waters remain challenging. </p><p> Modern hyperspectral imagers usually carry out their measurements on moving platforms, aircrafts or orbiting satellites, using push-broom scanning techniques for the acquisition of 3-D data cubes (along-track, cross-track and spectral). These data, however, may not always reflect accurately the temporal variability of measurements in a very dynamic atmosphere-ocean environment, especially in coastal areas. In recent years, new technologies have made possible the exploration of snapshot hyperspectral and polarimetric imaging of the ocean in an attempt to improve the accuracy of existing methods and exploring the pixel-by-pixel variability of the signal measured, often neglected in standard approaches. </p><p> The work presented in this thesis investigates and discusses imaging of underwater polarimetric targets in various water types and the estimation of parameters of the veiling light and the attenuation coefficients responsible for image degradation and blurring as a function of the water properties. In above water observations the effects of atmospheric aerosols and wind speed on the surface reflectance coefficients are studied along with the pixel-by-pixel variability of water-leaving radiance (<i>L_w</i>) and its relationship to water constituents for different coastal waters and atmospheric conditions, the impact of this variability on the uncertainties in above water measurements and satellite retrievals. Validation of all results has been achieved by the comparison with the comprehensive Vector Radiative Transfer simulations of the Atmosphere-Ocean System, as well as measurements by a number of other collocated radiometric and polarimetric instruments. </p><p>

<p> California Department of Water Resources (DWR) Snow Surveys Section has recently explored the potential use of recently developed hydrologic models to estimate snow water equivalent (SWE) for the Sierra Nevada mountain range. DWR Snow Surveys Section&rsquo;s initial step is to determine how well these hydrologic models compare to the trusted regression equations, currently used by DWR Snow Surveys Section. A comparison scheme was ultimately developed between estimation measures for SWE by interpreting model results for the Feather River Basin from: a) National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (JPL) gridded SWE reconstruction product, b) United States Geological Survey (USGS) Precipitation-Runoff Modeling System (PRMS), and c) DWR Snow Surveys Section regression equations. Daily SWE estimates were extracted from gridded results by computing an average SWE based on 1,000 ft elevation band increments from 3,000 to 10,000 ft (i.e. an elevation band would be from 3,000 to 4,000 ft). The dates used for processing average SWE estimates were cloud-free satellite image dates during snow ablation months, March to August, for years 2000&ndash;2012. The average SWE for each elevation band was linearly interpolated for each snow sensor elevation. The model SWE estimates were then compared to the snow sensor readings used to produce the snow index in DWR&rsquo;s regression equations. In addition to comparing JPL&rsquo;s SWE estimate to snow sensor readings, PRMS SWE variable for select hydrologic response units (HRU) were also compared to snow sensor readings. Research concluded with the application of statistical methods to determine the reliability in the JPL products and PRMS simulated SWE variable, with results varying depending on time duration being analyzed and elevation range.</p>

<p> Since 1865 California has practiced underground water storage through artificial recharge; however, in many parts of the state these efforts have been insufficient to meet its growing water demands, particularly for irrigated agriculture. During dry periods, vast agricultural areas depend upon groundwater for irrigation. In these areas, groundwater banking (GB) should be an essential strategy of their water management operations. GB is the practice of using surface water for percolation or injection into aquifers for later recovery. One variation of GB currently being studied in California is the use of agricultural lands for this practice (Ag-GB). Economic implications of Ag-GB need to be analyzed to inform water agencies and farmers interested in implementing this practice. This study proposes a conceptual model for determining the economic feasibility of Ag-GB at the irrigation district level. The Orland-Artois Water District (OAWD) in Glenn County is considered as the case study, and alfalfa as the test crop due to its tolerance to flooding and low use of pesticides and fertilizers which could leach into the aquifer. The proposed model consists of four components. The first component, the agricultural water demand calculator, calculates agricultural water demands based on historic land use, monthly reference evapotranspiration (ETo), monthly average precipitation, and average crop coefficient (Kc) values for the region. The second component, the aquifer mass balance model, is a one-bucket mass balance model that quantifies inflows and outflows to the simplified aquifer. The third component, the agronomic model, estimates costs and benefits of Ag-GB in terms of energy savings from pumping and crop production. The fourth component, the economic feasibility output, evaluates costs and benefits are evaluated to determine economic feasibility. The period of analysis is from 1993 through 2013. </p><p> Two policies (A and B) for implementation of Ag-GB are proposed and tested. Policy A proposes that all growers in OAWD pay for the implementation of the Ag-GB program. Policy B proposes that alfalfa growers using their lands for Ag-GB (Ag-GB alfalfa growers) are exempted from paying for Ag-GB implementation and the rest of the growers (non Ag-GB growers) pay for it. The economic analysis suggests that Policy A brings more costs than benefits to the Ag-GB alfalfa growers and hence is rejected as feasible. Policy B seems to bring more benefits than costs to all growers in OAWD and therefore it has potential to be economically feasible under the assumptions and limitations of the model. </p>

<p> Snow is a key component of the Earth&rsquo;s energy balance, climate, environment, and a major source of freshwater in many regions. Seasonal and perennial snow cover affect up to 50% of the Northern Hemisphere landmass, which accounts for vast regions of the Earth that influence climate, culture, and commerce significantly. Information on snow properties such as snow cover, depth, and wetness is important for making hydrological forecasts, monitoring climate change, weather prediction, and issuing snowmelt runoff, flash flood, and avalanche warnings. Hence, adequate knowledge of the areal extent of snow and its properties is essential for hydrologists, water resources managers, and decision-makers. </p><p> The use of infrared (IR) and microwave (MW) remote sensing (RS) has demonstrated the capability of estimating the presence of snow cover and snowpack properties with accuracy. However, there are few publicly accessible, operational RS-based snow depth products, and these only provide the depth of recently accumulated dry snow because retrievals lose accuracy drastically for wet snow (late winter - early spring). Furthermore, it is common practice to assume snow grain size and wetness to be constant to retrieve certain snow properties (e.g. snow depth). This approach is incorrect because these properties are space- and time- dependent, and largely impact the MW signal scattering. Moreover, the remaining operational snow depth products have not been validated against in-situ observations; which is detrimental to their performance and future calibrations. </p><p> This study is focused on the discovery of patterns in geospatial data sets using data mining techniques for mapping snow depth globally at 10 km spatial resolution. A methodology to develop a RS MW-based snow depth and water equivalent (SWE) product using regression tree algorithms is developed. The work divided into four main segments includes: (1) validation of RS-based IR and MW-retrieved Land Surface Temperature (LST) products, (2) studying snow wetness by developing, validating, and calibrating a Snow Wetness Profiler, (3) development of a regression tree algorithm capable of estimating snow depth based on radiative (MW observations) and physical snowpack properties, and (4) development of a global MW-RS-based snow depth product built on the regression tree algorithm. </p><p> A predictive model based on Regression Tree (RT) is developed in order to model snow depth and water equivalent at the Cooperative Remote Sensing Science and Technology Center &ndash; Snow Analysis and Field Experiment (CREST-SAFE). The RT performance analyzed based on contrasting training error, true prediction error, and variable importance estimates. The RT algorithm is then taken to a broader scale, and Japan Aerospace Exploration Agency (JAXA) Global Change Observation Mission &ndash; Water 1 (GCOM-W1) MW brightness temperature measurements were used to provide snow depth and SWE estimates. These SD and SWE estimates were evaluated against twelve (12) Snow Telemetry (SNOTEL) sites owned by the National Resources Conservation Service (NRCS) and JAXA&rsquo;s own snow depth product. Results demonstrated that a RS MW-based RT algorithm is capable of providing snow depth and SWE estimates with acceptable accuracy for the continental United States, with some limitations. The major setback to the RT algorithm is that it will only provide estimates based on the data with which it was trained. Therefore, it is recommended that the work be expanded, and data from additional in-situ stations be used to re-train the RT algorithm. The CREST snow depth and water equivalent product, as it was named, is currently operational and publicly accessible at https://www.noaacrest.org//snow/products/. </p><p>