Population growth, urbanization and climate change threaten urban water systems. The rise of demands caused by growing urban areas and the potential decrease of water availability caused by the increase of frequency and severity of droughts challenge the continued well-being of society. Due to increasing environmental and financial constraints, water management paradigms have shifted from supply augmentation to demand management, and water conservation initiatives may efficiently decrease water demands to more sustainable levels. To provide reliable assessment of the efficiencies of different demand management strategies, new modeling techniques are needed that can simulate decentralized decisions of consumers and their interactions with the water system. An integrated simulation-optimization framework, based on the paradigm of Complex Adaptive Systems, is developed here to model dynamic interactions and adaptations within social, built, and natural components of urban water systems. The framework goes beyond tradition engineering simulations by incorporating decentralized, heterogeneous and autonomous agents, and by simulating dynamic feedback loops among modeling components. The framework uses modeling techniques including System Dynamics, Cellular Automata, and Agent-based Modeling to simulate housing and population growth, a land use change, residential water consumption, the hydrologic cycle, reservoir operation, and a policy/decision maker. This research demonstrates the applicability of the proposed framework through a series of studies applied to a water supply system of a large metropolitan region that is located in a semi-arid region and suffers recurrently from severe droughts. A set of adaptive demand management strategies, that apply contingency restrictions, land use planning, and water conservation technologies, such as rainwater harvesting systems, are evaluated. A multi-objective Evolutionary Algorithm is coupled with the CAS simulation framework to identify optimal strategies and explore conflicting objectives within a water system. The results demonstrate the benefits of adaptive management by updating management decisions to changing conditions. This research develops a new hydrologic sustainability metric, developed to quantify the stormwater impacts of urbanization. The Hydrologic Footprint Residence captures temporal and spatial hydrologic characteristics of a flood wave passing through a stream segment and is used to assess stormwater management scenarios, including Best Management Practices and Low Impact Development.

No description available.

No description available.

Coastal regions are at risk of environmental threats. Flooding in coastal rivers is the result of intense precipitation which is triggered by climate change. Coastal watersheds are prone to losing significant amounts of sediment and nutrients because of the shorter transport pathway that drains directly into the coastal water. In this study, the hydrology, flood frequency, and water quality assessment of two coastal watersheds, Wolf River watershed (WRW) and Jourdan River watershed (JRW), were conducted using the Soil and Water Assessment Tool (SWAT). Since WRW and JRW are the main tributaries to fetch freshwater to Saint Louis Bay (SLB) of Western Mississippi Sound, an integrated approach to assess the influence of freshwater influx into the coastal water is also performed by coupling SWAT with hydrodynamic visual Environment Fluid Dynamics Code (v-EFDC). An auto-calibration tool, SWAT Calibration and Uncertainty Programs (SWAT-CUP) was used to calibrate and validate the flow, total suspended solids and mineral phosphorous for obtaining satisfactory statistical results. While comparing the flood frequency of historical, baseline and projected scenario in both watersheds, the results illustrated that using annual maximum series, 1% exceedance probability was the highest for WRW baseline scenario, whereas for JRW, 1% exceedance probability was the highest for projected scenario. The water quality assessment study of WRW and JRW suggested that ponds and wetlands were more effective in reducing TSS and riparian buffers were more effective in reducing MinP at the outlet of both the watersheds. The integrated approach of coupling SWAT-vEFDC model result indicated that major impact on water quality was observed at the location where the freshwater inflow into the SLB, and the impact was diminished while moving further along the Western Mississippi Sound. Overall, this study gives an insight for integrated coastal watershed management which includes prediction of future flood frequency, the application of best management practices for reducing sediment and nutrient load, and estimation of upstream watershed pollutant load draining along with runoff including its effect on the coastal water quality.

The aim of this thesis is to examine and develop new techniques in stormwater Best Management Practices (BMP) for nutrient and erosion reduction and monitoring by incorporation of low impact green technologies and sensor networks. Previous research has found excessive nutrient loading of nitrogen and phosphorus species from urban stormwater runoff can lead to ecological degradation and eutrophication of receiving lakes and rivers (Fareed and Abid, 2005). In response, the Florida Department of Environmental Protection (FDEP) has set forth reduction goals as established in Total Maximum Daily Load (TMDL) reports to reduce nutrient loading and restore, or maintain, Florida water bodies to reasonable conditions. Often times current stormwater management practices are not sufficient to attain these goals and further improvements in system design are required. In order to reach these goals, affordable technologies designed for both nutrient reduction and monitoring of system performance to deepen and improve our understanding of stormwater processes are required. Firstly this thesis examines the performance of three types of continuous-cycle Media Bed Reactors (MBRs) using Bio-activated Adsorptive Media (BAM) for nutrient reduction in three retention ponds located throughout the Central Florida region. Chapter 2 examines the use of a Sloped and Horizontal MBRs arranged in a baffling configuration, whereas Chapter 3 examines the field performance of a Floating MBR arranged in an upflow configuration. Each MBR was analyzed for performance in reducing total phosphorus, soluble reactive phosphorus, total nitrogen, organic nitrogen, ammonia, nitrates + nitrites, turbidity and chlorophyll a species as measured from the influent to effluent ends of the MBR. The results of the experiments indicate that MBRs may be combined with retention ponds to provide "green technology" alternatives for inter-event treatment of nutrient species in urban stormwater runoff by use of recyclable sorption media and solar powered submersible pumps. Secondly the thesis focusses on three new devices for BMP monitoring which may be integrated into wireless networks, including a Groundwater Variable Probe (GVP) for velocity, hydraulic conductivity and dispersion measurements in a retention pond bank (Chapter 4), an affordable Wireless Automated Sampling Network (WASN) for sampling and analysis of nutrient flux gradients in retention ponds (Chapter 5), and finally an Arc-Type Automated Pulse Tracer Velocimeter (APTV) for low velocity and direction surface water measurements in retention ponds and constructed wetlands (Chapter 6). The GVP was integrated with other environmental sensing probes to create a remote sensing station, capable of real-time data analysis of sub-surface conditions including soil moisture, water table stage. Such abilities, when synced with user control capabilities, may help to increase methods of monitoring for applications including erosion control, bank stability predictions, monitoring of groundwater pollutant plume migration, and establishing hydraulic residence times through subsurface BMPs such as permeable reactive barriers. Advancement of this technology may be used by establishing additional sub-stations, thereby creating sensing networks covering broader areas on the kilometer scale. Two methods for velocity calculation were developed for the GVP for low flow (Pe < 0.2) and high flow (Pe > 0.6) conditions. The GVP was found to operate from a 26-505 cmd-1 range in the laboratory to within ±26% of expected velocities for high-flow conditions and effectively measure directional flow angles to within ±14° of expected. Hydraulic conductivity measurements made by the GVP were confirmed to within ±12% as compared to laboratory measurements. The GVP was found capable of measuring the dispersion coefficient in the laboratory, however turbulent interferences caused during injection was found to occur. Further advancement of the technology may be merited to improve dispersion coefficient measurements. Automated water sampling can provide valuable information of the spatial and temporal distribution of pollutant loading in surface water environments. This ability is expanded with the development of the WASN, providing an affordable, ease-of-use method compared to conventional automated water samplers currently on the market. The WASN was found to effectively operate by text activation via GSM cellular networks to an activation module. Propagation of the signal was distributed to collection units via XBee modules operating on point-to-point star communication using an IEEE 802.15.4 protocol. Signal communications effectively transmitted in the field during a storm event to within a range of 200 feet and collected 50 ±4 ml samples at synced timed increments. A tracer study confirmed that no mixing of samples occurs when a factor of safety of 2 is applied to flush times. This technology provides similar abilities to current market devices at down to 10% of the cost, thereby allowing much more sampling locations for a similar budget. The Arc-Type APTV is useful in establishing both low range horizontal velocity fields and expanding low range velocity measurements below detection ranges of mechanical velocity meters. Installation of a field station showed system functionality, which may be integrated with other environmental sensing probes for surface water testing. This may assist in nutrient distribution analysis and understanding the complex behavior of hydraulic retention times within wetland systems. The device was found to work effectively in both lab and field environments from a 0.02 – 5.0 cms-1 range and measure velocity within approximately ±10% of an acoustic Doppler velocimeter and within an average of ±10° of directional measurements. A drop in accuracy was measured for velocity ranges > 4.5 cms-1. The field station operated on 3G CDMA cellular network two-way communication by installation of a Raven cellular modem. Use of LoggerNet software allowed control and data acquisition from anywhere with an internet connection. This thesis also introduces brief discussions on expanding these "point" measurement technologies into sensing networks. Installation of sub-stations with communication protocols to one central master node station may broaden the sensing system into much larger kilometer-scale ranges, thus allowing large spatial analysis of environmental conditions. Such an integration into controllable sensing networks may help bridge the gap and add calibration and verification abilities between fine-resolution "point" measurements and large scale technologies such as Electrical Resistivity Tomography and satellite remote sensing. Furthermore, application of sensing networks may assist in calibration and verification of surface and groundwater models such as ModFlow, SVFlux and FEHM.

Green adsorption media with the inclusion of renewable and recycled materials can be applied as a stormwater best management practice for copper removal. A green adsorption media mixture composed of recycled tire chunk, expanded clay aggregate, and coconut coir was physicochemically evaluated for its potential use in an upflow media filter. The results found that the use of the green adsorption media mixture in isolation or the coconut coir with an expanded clay filtration chamber could be an effective and reliable stormwater best management practice for copper removal. A suite of tests were conducted on the media mixture and the individual media components including studies of isotherm, reaction kinetics, column adsorption and reaction kinetics. Batch adsorption tests revealed that the media and media mixture follow both the Freundlich and Langmuir isotherm models and that the coconut coir had the highest affinity for copper. A screening of desorbing agents revealed that hydrochloric acid has good potential for copper desorption, while batch tests for desorption with hydrochloric acid as the desorbing agent showed the data fit the Freundlich isotherm model. Reaction kinetics revealed that the adsorption reaction took less than 1 hour to reach equilibrium and that it followed pseudo-second order kinetics for the mixture and coconut. Desorption kinetic data had high correlation with the pseudo-second order model and revealed a rapid desorption reaction. Batch equilibrium data over 3 adsorption/desorption cycles found that the coconut coir and media mixture were the most resilient and demonstrated that they could be used through 3 or more adsorption/desorption cycles. The coconut coir also performed the best under dynamic conditions, having an equilibrium uptake of 1.63 mg?g-1, compared to 0.021 mg?g-1 at an influent concentration of 1.0 mg?L-1 and a hydraulic retention time of 30 minutes. A physical evaluation of the media found the macro-scale properties, such as particle size distribution and mass-volume relationships, and observed the micro-scale properties such as surface and pore microstructures, crystalline structures, and elemental composition. FE-SEM imaging found a strong correlation between the porosity of the micro pore structure and the adsorptive capacity. The equilibrium and dynamic adsorption testing results were confirmed by elemental analysis, which showed measureable quantities of copper in the coconut coir and media mixture after adsorption followed by partial desorption. A new scaling-up theory was developed through a joint consideration of the Damköhler and Péclet numbers for a constant media particle size such that a balance between transport-controlled and reaction-controlled kinetics can be harmonized. A series of column breakthrough tests at varying hydraulic residence times revealed a clear peak adsorptive capacity for the media mixture at a Damköhler number of 2.7. The Péclet numbers for the column breakthrough tests indicated that mechanical dispersion is an important effect that requires further consideration in the scaling-up process. However, perfect similitude of the Damköhler number cannot be maintained for a constant media particle size, and relaxation of hydrodynamic similitude through variation of the Péclet number must occur.

Environmental concerns are becoming increasingly relevant during a period of hemorrhaging ecosystem goods and services. Restoring these would result in positive outcomes for public health and economic benefit. This thesis seeks to address two environmental concerns: (1) accurate soil mapping and (2) treatment of nitrogen to affect water quality change.The current method of soil mapping, SSURGO (USDA‐NRCS Soil survey), is often erroneous and misleading. Two studies in this dissertation are conducted to evaluate the potential that different resolution digital elevation models (DEMs) have to distribute soil characteristics successfully. These studies are conducted in southwest Virginia and western Vermont. The aforementioned studies evaluated 36 and 59 soil samples, respectively. Spatial characteristics, including slope, catchment area, and topographic wetness, are derived from several DEMs. In chapter 2, these characteristics are spatially compared, and we found that small resolution rasters result in narrow flow paths relative to coarser rasters. In chapter 3, we isolate the analysis to focus on resolution size, instead of a mix of both resolution size and generation method. This is done by recursively coarsening small rasters, deriving spatial attributes from said rasters and evaluating their potential to fit the soil characteristics of interest. Here we found that slopes generated from resolutions smaller than 11m were poor predictors of soil characteristics. Both chapters are finished by proposing and evaluating a soil map. Proposed regressions beat SSURGO in all investigated properties. Furthermore, proposed maps consistently beat out uninformed smallest resolution derived maps.Chesapeake bay water quality managers are struggling to achieve targets for nitrogen loading. This is in part due to the widespread presence of legacy nitrogen. Legacy nitrogen is an emerging issue, and springs exporting high levels of nitrogen are not uncommon in northern Virginia. This thesis explores, in part, a novel concept of treating large loads of nitrogen exported from a spring with a bioreactor. Bioreactors are a young science that most typically pair carbon heavy subterranean receptacles to agricultural drainage. This provides a location for nitrogen fixing bacteria to consume nitrate/nitrite, turning these into inert nitrogen gas. A spring fed bioreactor is studied for 10 months, and bioreactor conditions including influent and effluent nitrogen concentrations, bioreactor flow, and temperature are collected. A model driven by first order reaction equations is found to be most accurate with inputs of temperature and bioreactor age. The resulting marginal effects of these inputs were consistent with previously reported studies. / Doctor of Philosophy / Centuries of industrialization have resulted in widespread human progress but have, at times, adversely impacted the environment. Constituents rely heavily on environmental services, such as clean air and water, to subsist. Environmental degradation has resulted in detrimental effects to public health, and remediation is currently economically viable. As such, there are strong incentives for researchers to understand environmental processes at a fundamental level. One such process is soil characteristic distribution. The distribution of soil characteristics, such as soil texture or organic matter, is especially important for agriculturalists, hydrologists and geotechnicians. Soil texture and organic matter distribution can affect crop yield, nitrogen export to surface waters, and structural stability of soils. Thus, accurate characterization of measured soil properties is paramount to multiple fields. The most typically used soil map is USDA‐NRCS Soil survey (commonly referred to as SSURGO). Currently, the SSURGO database is a poor predictor of soil characteristics. There is an opportunity to improve soil characteristic distribution using digital elevation models (DEMs). As DEMs become cheaper to develop, they are typically available in multiple resolutions and generation methods. In this research, several DEMs are used to better soil maps for watersheds in Southwest Virginia and Western Vermont. Both studies showed that DEMs can better distribute soils when compared to the current SSURGO maps. Additionally, we showed that the finest resolution dataset was not always best, and mixed resolution topographic wetness indices to be most advantageous for distributing soils.Another such process is remediation of surface waters from high loads of nitrogen and phosphorus. The Haber-Bosch method of producing nitrogen fertilizer is one of the most important human innovations in recent history. This method is likely responsible for the aversion of widespread famine in the early 1900s. However, residents of multiple river systems, including the Chesapeake Bay and the Mississippi River, are suffering from the adverse effects of widespread hypoxic/anoxic (with little/no oxygen, respectively) zones within water. These have partially been responsible for the decline of commercial ventures such as fisheries and tourism. These zones are caused by eutrophication, a process of unsustainable plant growth in the presence of nitrogen and phosphorus. Water quality managers typically target agricultural runoff and point source polluters when trying to eliminate anthropogenic nitrogen. However, legacy nitrogen (nitrogen stored in groundwater in excess of a year) has become an emerging concern for water quality. It is not uncommon for springs in karst areas to be contaminated with high concentrations of nitrogen. These springs present a point source that can be treated by an emerging technology: bioreactors. Bioreactors are subterranean, woodchip filled basins that provide a location for microbes to exchange water soluble nitrogen for inert nitrogen gas. The consistency in nitrogen loading and constant flow provide stability relative to more traditional bioreactor installations. Most typically, bioreactors are installed downstream of agricultural drainage systems, and influent flow and nitrogen load depend wholly on precipitation/irrigation and nitrogen application. In this thesis, a novel spring fed bioreactor is studied. Removal rates of nitrogen are quantified using a regression driven by reaction kinetics. The analysis showed bioreactor efficiency was intimately related to hydraulic residence time, nitrogen loading, bioreactor bed temperature, and bioreactor age. The spring fed bioreactor is found to be advantageous because of its consistency, and disadvantages because springs are colder and thus less efficient than typical irrigated runoff.

No description available.

Forestry Best Management Practices (BMPs) include guidelines, recommendations, and protocols utilized to protect forest water quality from nonpoint source pollutants (NPSP). Sediment is the most common NPSP associated with forest operations, and BMPs are implemented primarily to reduce erosion and potential sediment delivery to streams. Skid trails, stream crossings, forest roads, decks, and harvest areas are major forest operational features that have the potential to erode and deliver sediment to streams. These five features are also common BMP categories evaluated by states across the southeastern U.S. Although BMPs are designed to minimize erosion and sediment delivery, the exact relationship between BMP implementation rates, erosion rates, and potential sediment delivery is largely unexamined. Specifically, the inherently intuitive but unverified concept that increasing levels of BMP implementation decreases erosion and sediment delivery associated with forest harvesting. This hypothesis was tested in this project at five operational features located within three physiographic regions, including the Mountains, Piedmont, and Coastal Plain, across clearcut harvest sites in the southeastern U.S. First, BMP implementation rates, audit questionnaires, and state guidelines were compared across 13 states in the southeastern region at 116 clearcut harvest sites. Overall, BMPs were implemented at an average rate of 90.1% in the southeastern United States, thus demonstrating that BMPs are currently being implemented consistently at high levels throughout the region. Across all regions, average BMP implementation rates were highest at harvest areas (95.6%), followed by decks (92.7%), haul roads (91.9%), stream crossings (88.2%), and skid trails (82.9%). Average BMP implementation rates for Mountain stream crossings (83.9%) and skid trails (76.1%) were significantly lower than rates calculated in the Piedmont and Coastal Plain, and had the lowest implementation rate for any feature in any region. These findings indicate that skid trails and stream crossings, especially in the Mountains, may benefit the most from enhanced BMP implementation and resources. In the second manuscript, the relationship between BMP implementation, estimated erosion, and potential sediment delivery were examined in three regions of Virginia and North Carolina. This study is one of the only forestry studies that have presented sediment delivery ratios by operational feature and physiographic region. BMP implementation rates and sediment delivery ratios were poorly correlated, however, a significant inverse relationship existed between BMP implementation and the total sediment mass delivered to streams (Spearman ρ = -0.2206, p-value = 0.0027). Generally, as BMP implementation increased, erosion rates and the amount of sediment delivered to streams also decreased. Additionally, this study demonstrated that most of the erosion generated by clearcutting operations in the southeast is trapped in either the harvest area or in Streamside Management Zones (SMZs) prior to reaching the stream. In the third manuscript, BMP implementation rates and erosion estimates were categorized into three BMP levels (BMP−, BMP-standard, BMP+) which represent low, moderate, and high levels of BMP implementation, respectively. Skid trails and haul roads generally had the highest erosion estimates, regardless of BMP level and physiographic region. Non-parametric correlation analyses indicated that significant inverse relationships existed between BMP implementation rates and erosion estimates at skid trails (Spearman ρ = -0.589, p-value < 0.0001), haul roads (Spearman ρ = -0.388, p-value < 0.0001), and harvest areas (Spearman ρ = -0.2305, p-value = 0.0169), while decks and stream crossings were more poorly correlated with erosion estimates. This reinforces the need for BMP audit questions that specifically address ground cover and bare soil, water control structures, gradients, and stabilization to better address potential erosion and sedimentation. Clearcut areas, erosion and sediment estimates, and sediment removal efficiencies were presented for the Mountains, Middle/Lower Coastal Plain, and Piedmont/Upper Coastal Plain for the fourth, fifth, and sixth manuscripts, respectively. Regardless of BMP level and physiographic region, a combination of harvest areas, skid trails, and haul roads were responsible for over 95% of potential sediment delivery. Increasing site-wide BMP implementation from BMP− to BMP+ could reduce sediment delivery by 70% in all physiographic regions. High levels of BMP implementation were most effective at reducing potential erosion and sediment delivery from skid trails and haul roads throughout the southeast. Findings from these studies demonstrate that current BMPs are highly effective at mitigating sediment. In the southeastern U.S., increasing levels of BMP implementation effectively reduce both potential erosion and sedimentation associated with forest harvesting. Generally, both estimated erosion and sedimentation associated with clearcutting in the region is much lower than rates associated with other land uses such as development or agriculture, especially when BMPs are implemented at standard or high levels. However, several opportunities exist to improve the effectiveness of BMPs in the southeastern U.S. Skid trails, haul roads, and stream crossings consistently had the lowest BMP implementation rates and highest estimated erosion rates and sediment delivery ratios. While these features only represent a small proportion of total clearcut area in the southeast, they are responsible for a disproportionate amount of sediment delivery and should receive more attention and resources during the pre-harvest planning and closure processes. Conclusively, this project addresses several knowledge gaps pertaining to water quality impacts resulting from harvesting operations in the southeastern U.S. For instance, this is only the third project that has presented sediment delivery ratios associated with forest operations in the southeastern U.S., and the first to do so for the Mountains and Coastal Plain regions. On average, SMZs and harvest areas trap 66-96% of sediment on-site before it can be deposited into streams. Additionally, this project provides one of the first and most comprehensive regional comparisons of state BMP manuals, audits, and programs in-field using a third-party approach. Several significant differences existed among state BMP programs and protocol, and states may need to design specific BMP guidelines and audit protocol for major physiographic regions to address the challenges and variation of on-site conditions inherent of each region. Additionally, this project presents one of the only regional-scale estimates of sediment and efficiencies of increased levels of BMP implementation at mitigating sediment associated with forest operations in the southeastern U.S. Conclusively, this project provides forest managers, state and federal agencies, and policymakers with a robust assessment on the effectiveness of forestry BMPs in the southeast. / Doctor of Philosophy / Forestry Best Management Practices (BMPs) are used throughout the southeastern U.S. to minimize the impacts that harvesting has on soil and water resources. Eroded soil that is eventually deposited into streams as sediment is the most important pollutant that BMPs address in forestry. Common BMP guidelines utilized to minimize sediment include leaving riparian buffers along streams, providing ground cover, minimizing slopes on roads, and using water control structures to divert runoff from road systems. The exact relationship between forestry BMPs, erosion, and sediment delivery is largely unexamined. Objectively, this study was designed to provide a better understanding of this relationship, and to present estimates of erosion and sediment delivery resulting from clearcut harvesting on over 100 harvest sites across 13 states with diverse conditions and topography in the southeastern U.S. Forestry BMPs are being implemented at an average rate of 90.1% across the southeast. BMPs are implemented at higher rates in the Coastal Plain, followed by the Piedmont and Mountains. Generally, harvest areas had the highest BMP implementation rates, followed by decks, haul roads, stream crossings, and skid trails, respectively. This relationship was consistent across most regions and states. Logging decks, which are areas where wood was transported for processing and loading onto logging trucks, were generally located distantly from streams and followed most state-approved BMPs. Whereas skid trails, which are low-standard temporary roads trafficked during primary transport and require water diversion structures such as waterbars, occupied a much larger area and were generally located on much steeper slopes. Mountain skid trails leading to stream crossings are concerning because of their low BMP implementation rates and high erosion potential. These features would greatly benefit from increased ground cover and water control structures. Forestry BMPs reduce both erosion and sedimentation associated with clearcutting. The highest level of BMP implementation reduces potential sediment by over 70% in all regions, and the highest BMP level is most effective at reducing sediment from skid trails and forest roads. Average sedimentation rates, especially at the highest BMP level, calculated for Mountains, Piedmont, and Coastal Plain clearcuts are considerably lower than sedimentation rates associated with agricultural and developmental land uses. Clearcut forestland, which is generally the most erodible time period during a forest's cycle, only makes up 1-2% of total forestland in the southeast annually. Both erosion and sedimentation rates should decrease further in the years following harvesting as the site revegetates and trees begin to reestablish. This study verifies that BMPs are highly effective at reducing erosion and sediment while subsequently providing estimates of erosion and sediment delivery based on ranges of BMP implementation that state forestry agencies can use to better quantify the effectiveness of their BMPs.

Climate change impacts hydrology, nutrient cycling, agricultural conservation practices, and greenhouse gas (GHG) emissions. The Chesapeake Bay and its watershed are subject to the largest and most expensive Total Maximum Daily Load (TMDL) ever developed. It is unclear if the TMDL can be met given climate change and variability (e.g., extreme weather events). The objective of this dissertation is to quantify the impact of climate change and climate on water resources, nutrient cycling and export in agroecosystems, and agricultural conservation practices in the Chesapeake Bay watershed. This is accomplished by developing and employing a suite of modelling tools. GHG emissions from agroecosystems, particularly nitrous oxide (N2O), are an increasing concern. To quantify N2O emissions a routine was developed for the Soil and Water Assessment Tool (SWAT) model. The new routine predicts N2O and di-nitrogen (N2) emissions by coupling the C and N cycles with soil moisture, temperature, and pH in SWAT. The model uses reduction functions to predict total denitrification (N2 + N2O production) and partitions N2 from N2O using a ratio method. The SWAT nitrification routine was modified to predict N2O emissions using reduction functions. The new model was tested using GRACEnet data at University Park, Pennsylvania, and West Lafayette, Indiana. Results showed strong correlations between plot measurements of N2O flux and the model predictions for both test sites and suggest that N2O emissions are particularly sensitive to soil pH and soil N, and moderately sensitive to soil temperature/moisture and total soil C levels. The new GHG model was then used to analyze the impact of climate change and extreme weather conditions on the denitrification rate, N2O emissions, and nutrient cycling/export in the 7.4 km2 WE38 watershed in Pennsylvania. Climate change impacts hydrology and nutrient cycling by changing soil moisture, stoichiometric nutrient ratios, and soil temperature, potentially complicating mitigation measures. To quantify the impact of climate change we forced the new GHG model with downscaled and bias-corrected regional climate model output and derived climate anomalies to assess their impact on hydrology, nitrate (NO3-), phosphorus (P), and sediment export, and on emissions of N2O and N2. Model-average (± standard deviation) results indicate that climate change, through an increase in precipitation, will result in moderate increases in winter/spring flow (2.7±10.6 %) and NO3- export (3.0±7.3 %), substantial increases in dissolved P (DP, 8.8±19.8 %), total P (TP, 4.5±11.7 %), and sediment (17.9±14.2 %) export, and greater N2O (63.3±50.8 %) and N2 (17.6±20.7 %) emissions. Conversely, decreases in summer flow (-12.4±26.7 %) and the export of P (-11.4±27.4 %), TP (-7.9±24.5 %), sediment (-4.1±21.4 %), and NO3- (-12.2±31.4 %) are driven by greater evapotranspiration from increasing summer temperatures. Increases in N2O (20.1±29.3 %) and decreases in N2 (-13.0±14.6 %) are also predicted in the summer and driven by increases in soil moisture and temperature. In an effort to assess the impact of climate change at a regional level, the model was then scaled-up to the entire Susquehanna River basin and was used to evaluate if agricultural best management practices (BMPs) can offset the impact of climate change. Agricultural BMPs are increasingly and widely employed to reduce diffuse nutrient pollution. Climate change can complicate the development, implementation, and efficiency of BMPs by altering hydrology, nutrient cycling, and erosion. We select and evaluate four common BMPs (buffer strips, strip crop, no-till, and tile drainage) to test their response to climate change. We force the calibrated model with six downscaled global climate models (GCMs) for a historic period (1990-2014) and two future scenario periods (2041-2065) and (2075-2099) and quantify the impact of climate change on hydrology, NO3-, total N (TN), DP, TP, and sediment export with and without BMPs. We also tested prioritizing BMP installation on the 30% of agricultural lands that generate the most runoff (e.g., critical source areas-CSAs). Compared against the historical baseline and excluding the impact of BMPs, the ensemble model mean (± standard deviation?) predictions indicate that climate change results in annual increases in flow (4.5±7.3%), surface runoff (3.5±6.1%), sediment export (28.5±18.2%) and TN (9.5±5.1%), but decreases in NO3- (12±12.8%), DP (14±11.5%), and TP (2.5±7.4%) export. When agricultural BMPs are simulated most do not appreciably change the overall water balance; however, tile drainage and strip crop decrease surface runoff generation and the export of sediment, DP, and TP, while buffer strips reduced N export substantially. Installing BMPs on critical source areas (CSAs) results in nearly the same level of performance for most practices and most pollutants. These results suggest that climate change will influence the performance of BMPs and that targeting BMPs to CSAs can provide nearly the same level of water quality impact as more widespread adoption. Finally, recognizing that all of these model applications have considerable uncertainty associated with their predictions, we develop and employ a Bayesian multi-model ensemble to evaluate structural model prediction uncertainty. The reliability of watershed models in a management context depends largely on associated uncertainties. Our Objective is to quantify structural uncertainty for predictions of flow, sediment, TN, and TP predictions using three models: the SWAT-Variable Source Area model (SWAT-VSA), the standard SWAT model (SWAT-ST), and the Chesapeake Bay watershed model (CBP-model). We initialize each of the models using weather, soil, and land use data and analyze outputs of flow, sediment, TN, and TP for the Susquehanna River basin at the Conowingo Dam in Conowingo, Maryland. Using these three models we fit Bayesian Generalized Non - Linear Multilevel Models (BGMM) for flow, sediment, TN, and TP and obtain estimated outputs with 95% confidence intervals. We compare the BGMM results against the individual model results and straight model averaging (SMA) results using a split time period analysis (training period and testing period) to assess the BGMM in a predictive fashion. The BGMM provided better predictions of flow, sediment, TN, and TP compared to individual models and the SMA during the training period. However, during the testing period the BGMM was not always the best predictor; in fact, there was no clear best model during the testing period. Perhaps more importantly, the BGMM provides estimates of prediction uncertainty, which can enhance decision making and improve watershed management by providing a risk-based assessment of outcomes. / Ph. D.