Natural and human-induced environmental changes have been altering the earth's surface and hydrological processes, and thus directly contribute to the severity of flood hazards. To understand these changes and their impacts, this research developed a GISbased hydrological and hydraulic modeling system, which incorporates state-of-the-art remote sensing data to simulate flood under various scenarios. The conceptual framework and technical issues of incorporating multi-scale remote sensing data have been addressed. This research develops an object-oriented hydrological modeling framework. Compared with traditional lumped or cell-based distributed hydrological modeling frameworks, the object-oriented framework allows basic spatial hydrologic units to have various size and irregular shape. This framework is capable of assimilating various GIS and remotely-sensed data with different spatial resolutions. It ensures the computational efficiency, while preserving sufficient spatial details of input data and model outputs. Sensitivity analysis and comparison of high resolution LIDAR DEM with traditional USGS 30m resolution DEM suggests that the use of LIDAR DEMs can greatly reduce uncertainty in calibration of flow parameters in the hydrologic model and hence increase the reliability of modeling results. In addition, subtle topographic features and hydrologic objects like surface depressions and detention basins can be extracted from the high resolution LiDAR DEMs. An innovative algorithm has been developed to efficiently delineate surface depressions and detention basins from LiDAR DEMs. Using a time series of Landsat images, a retrospective analysis of surface imperviousness has been conducted to assess the hydrologic impact of urbanization. The analysis reveals that with rapid urbanization the impervious surface has been increased from 10.1% to 38.4% for the case study area during 1974 - 2002. As a result, the peak flow for a 100-year flood event has increased by 20% and the floodplain extent has expanded by about 21.6%. The quantitative analysis suggests that the large regional detentions basins have effectively offset the adverse effect of increased impervious surface during the urbanization process. Based on the simulation and scenario analyses of land subsidence and potential climate changes, some planning measures and policy implications have been derived for guiding smart urban growth and sustainable resource development and management to minimize flood hazards.

Irrigation practices in the Rincon Valley and Mesilla Basin of the Lower Rio Grande have evolved over the last century into a complex setting of transboundary conjunctive use. Three major water users have surface and groundwater appropriation rights regulated by compact, treaty, and operating rules and agreements. The analysis of complex relationships between supply/demand components and the effects of surface-water and groundwater use requires an integrated hydrologic model to track all of the use and movement of water. Models previously developed for the region relied on a priori estimates of net irrigation flux or externally-calculated landscape water budgets. This study instead utilizes a MODFLOW model with the Farm Process (MF-FMP), which directly couples the surface-water and groundwater regimes through simulation of landscape processes. This allows the assessment of stream-aquifer interactions in the context of fulfilling irrigation demands with variable supplies of surface water allotments and supplemental groundwater pumping. MF-FMP also simulates direct uptake of groundwater by crops, an important utility for modeling a region with significant acreage dedicated to pecan orchards, a phreatophytic crop. The abilities and limitations of this new model are explored through scenario simulations meant to estimate streamflow depletions caused by historic pumping levels.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23855号 / 工博第4942号 / 新制||工||1772(附属図書館) / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 立川 康人, 教授 中北 英一, 講師 萬 和明 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Surface depressions are important topographic features, which affect overland flow, infiltration, and other hydrologic processes. Specifically, depressions undergo filling-spilling-merging-splitting processes under natural rainfall conditions, featuring discontinuity in hydrologic connectivity and variability in contributing area. However, a constant and time-invariant contributing area is often assumed in traditional hydrologic modeling, and consequently, the real threshold-controlled overland flow dynamics cannot be captured. The overall goal of this dissertation research is to improve hydrologic modeling, especially for depression-dominated areas, by quantifying the hydrologic effects of depressions. The specific objectives are to analyze the hydrotopographic characteristics of depressions and identify the intrinsic relationships of hydrologic variables, develop new modeling methods to simulate the depression-oriented dynamics in overland flow and variations in contributing area, and reveal the influence of spatially distributed depressions on the surface runoff generation and propagation processes. To achieve these objectives, three studies were conducted: (1) the frequency distribution of depression storage capacities was determined and a puddle-based unit (PBU)-probability distribution model (PDM) was developed; (2) the intrinsic changing patterns of contributing area and depression storage were identified, based on which a new depression-oriented variable contributing area (D-VCA) model was developed; and (3) a modified D-VCA (MD-VCA) model was further developed by introducing a depressional time-area zone scheme and a new variable contributing area-based surface runoff routing technique to account for the spatial distribution of depressions. These three models (PBU-PDM, D-VCA, and MD-VCA) were evaluated through the applications to depression-dominated watersheds in North Dakota, and simulation results demonstrated their capabilities in simulating the variations of contributing areas and threshold-controlled overland flow dynamics. In addition, these three studies emphasized the important roles of depressions in the evolution of contributing areas as well as surface runoff generation and propagation. Without considering the spatial distribution of depressions, the formation of contributing area and the timing and quantity of runoff contributions cannot be characterized.

No description available.

No description available.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23480号 / 工博第4892号 / 新制||工||1764(附属図書館) / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 立川 康人, 准教授 市川 温, 教授 堀 智晴 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Excessive nutrient concentrations from a different point and non-point sources are the main cause of water impairment in the United States. Appropriate management practices, according to the source and quantity of pollutions, need to be implemented to control excessive nutrient influx in the water body. Various types of hydrological and water quality models with diverse function, capability and degree of complexity are employed to quantify watershed hydrologic processes and nutrient pollution. Multiple models can be applied to a watershed but the suitable model must be selected based on watershed type and simulation need. Two watershed-scale models, Soil and Water Assessment Tool (SWAT) and Hydrologic Simulation Program-Fortran (HSPF) were chosen for this study to simulate runoff, sediment yield, and nutrient load from the Big Sunflower River Watershed (BSRW) of Mississippi. The objectives of this study are to access the nutrient sources within the watershed, determine the appropriate model to quantify them, develop and evaluate model considering spatial and temporal variations in input data, and evaluate the effectiveness of different Best Management Practices (BMPs) on surface runoff, sediment yield and nutrient load at watershed scale. This study has identified a potential source of nutrients in BSRW and provided a suitable BMP for its management. Similarly, the study found both SWAT and HSPF were efficient in the simulation of streamflow, sediment yield and nutrient load, where SWAT was more efficient during simulation streamflow and sediment yield. Likewise, the study established that both water-quantity and water-quality are sensitive to the change in LULC data layers and thus, seasonal LULC data applied to SWAT will better explain variation in hydrology and water quality as compared to the annual cropland data layer. Moreover, the study showed that well managed vegetative filter strip was very efficient in reducing sediment yield, TN, and TP at both field and watershed scale among different BMPs evaluated at field and watershed scale. This study will be beneficial in developing efficient nutrient management strategy at field and watershed scale, selecting appropriate model and input according to the need and type of watershed, and providing further research opportunities to the scientific community.

No description available.

<p>The thesis aims to investigate the impact of climate change on the hydrology of four semi-urban watersheds in southern Ontario. The study is mainly concerned with future changes in climate variables and flow regimes. The study also assesses future changes in the frequency and magnitude of peak and low flows. The hydrologic effects of climate change were assessed using a couple of climate and hydrological models. Three regional climate models (RCMs), namely, Canadian Regional Climate Model (CRCM), United States Regional Climate Model 3 (RCM3), United Kingdom Hadley Regional Model 3 (HRM3) were used to extract raw climate variables. The raw RCM data were corrected using a bias correction method. The method performance statistics and the nonparametric test results revealed that the bias corrected climate variables followed the patterns of the observed climate variables for all weather stations. Future climate scenario was then simulated and analyses show increases in annual precipitation about 5-8% and increases in mean annual daily mean temperature about 2.6-3.2 oC. Three hydrological models (namely HBV, MAC-HBV, and SAC-SMA) were used for flow simulation. The models' validation results show a good agreement with the observed flow with a Nash Sutcliffe efficiency around 0.49-0.75 and a correlation coefficient of around 0.7-0.8 for all sub-basins. The three hydrologic models coupled with the bias corrected RCMs data were used to simulate current and future flow. For future period (2050s), the models predicted increasing winter flow and decreasing spring, summer and autumn flows. Mean annual flow shows slight to moderate changes. Significant increases in peak and low flow magnitude are predicted for higher return periods (20-100 years). Overall, the effects of projected future changes in precipitation and temperature clearly govern the significant changes in seasonal and annual flows, peak and low flow magnitudes and frequencies. Using three hydrologic and three climate models projections, a comprehensive picture of probable hydrologic impact of climate change was assessed in the study area. The wide range of predicted changes will have significant implications for future water resources development in the selected semi-urban watersheds.</p> / Master of Applied Science (MASc)