Civil & Environmental Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/2221
Browse
13 results
Search Results
Item BEYOND PEAK RATE FACTOR 484: USING RADAR RAINFALL, GAUGED STREAMFLOW, AND DISTRIBUTED WATERSHED MODELING TO INVESTIGATE PARAMETERS OF THE NATURAL RESOURCES CONSERVATION SERVICE CURVILINEAR UNIT HYDROGRAPH(2024) Shehni Karam Zadeh, Mani; Brubaker, Kaye L.; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Accurate estimation of runoff and peak discharge is crucial in hydrology for engineering design and flood management. The Natural Resources Conservation Service’s (NRCS) Unit Hydrograph (UH) is a widely used model to predict the runoff response of an ungauged watershed to a precipitation event. The NRCS UH model makes use of a Peak Rate Factor (PRF) to quantify the peak discharge. The standard value of PRF is 484; however, PRF can be adjusted as a user input variable in NRCS tools such as the WinTR-20 software. Little guidance is available to appropriately estimate PRF for specific regions and evaluate its overall usefulness in the runoff and peak discharge estimation. Time of concentration (tc) is another input variable in the NRCS UH model; inconsistent definitions of tc and diverse methods of calculating it contribute to uncertainty in hydrologic estimates and predictions. The NRCS UH approach assumes that the watershed’s temporal runoff response to each increment of precipitation is identical in shape and proportional to precipitation excess in that increment of time. The UH, PRF, and tc are often assumed to be time-invariant properties of a watershed. This dissertation sought to improve the knowledge and understanding of PRF and tc. First, it evaluated if a unique UH and tc exist for a given watershed from various storm events. It then assessed whether variations in PRFs can be explained by watershed predictor variables and if PRFs in neighboring watersheds followed a local trend. This phase of study employed a gamma function representation of the NRCS UH, with two parameters: time to peak (tp) and shape (m). Precipitation inputs were watershed-averaged time series of NEXRAD level III data, and streamflow data were obtained from the United States Geological Survey (USGS) National Water Information System (NWIS). The UHs were derived from a constrained optimization approach, and PRF and tc were estimated for each event. Subsequently, a fully distributed model was created to provide insight on PRF and tc, and investigate the impact of detailed soil profiles on runoff and peak discharge. Finally, a fully distributed model was applied to simple, synthetic watersheds to investigate the impact of selected watershed parameters on PRF, time to peak, peak discharge and overall shape of the UH. To the best of the author's knowledge, this study is the first attempt to generate UHs from a simple distributed model and estimate associated PRFs. The findings suggest that there is no unique UH and tc for a given watershed, and UH shape and parameters change for every event in a given watershed. Additionally, the variations in PRFs cannot be explained by variations in selected watershed predictor variables. The distributed model results provided insights about the application of detailed soil profiles in runoff and peak discharge estimation. The findings also suggest that, except for Manning's roughness, selected watershed characteristics cannot be used to estimate PRF in a synthetic V-shaped watershed. These findings suggest that the application of PRF to estimate peak discharge should be used with caution due to the inherent uncertainties and lack of physical meaning of the parameter.Item Characterizing a Multi-Sensor System for Terrestrial Freshwater Remote Sensing via an Observing System Simulation Experiment (OSSE)(2022) Wang, Lizhao; Forman, Barton A; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Terrestrial freshwater storage (TWS) is the vertically-integrated sum of snow, ice, soilmoisture, vegetation water content, surface water impoundments, and groundwater. Among these components, snow, soil moisture, and vegetation are the most dynamic (i.e., shortest residence time) as well as the most variable across space. However, accurately retrieving estimates of snow, soil moisture, or vegetation using space-borne sensors often requires simultaneous knowledge of one or more of the other components. In other words, reasonably characterizing terrestrial freshwater requires careful consideration of the coupled snow-soil moisture-vegetation response that is implicit in both TWS and the hydrologic cycle. One challenge is to optimally determine the multi-variate, multi-sensor remote sensing observations needed to best characterize the coupled snow-soil moisture-vegetation system. Different types of sensors each have their own unique strengths and limitations. Meanwhile, remote sensing data is inherently discontinuous across time and space, and that the revisit cycle of remote sensing observations will dictate much of the efficacy in capturing the dynamics of the coupled snow-soil moisture-vegetation response. This study investigates different snow sensors and simulates the sensor coverage as a function of different orbital configurations and sensor properties in order to quantify the discontinuous nature of remotely-sensed observations across space and time. The information gleaned from this analysis, coupled with a time-varying snow binary map, is used to evaluate the efficacy of a single sensor (or constellation of sensors) to estimate terrestrial snow on a global scale. A suite of different combinations, and permutations, of different sensors, including different orbital characteristics, is explored with respect to 1-day, 3-day, and 30-day repeat intervals. The results show what can, and what cannot, be observed by different sensors. The results suggest that no single sensor can accurately measure all types of snow, but that a constellation composed of different types of sensors could better compensate for the limitations of a single type of sensor. Even though only snow is studied here, a similar procedure could be conducted for soil moisture or vegetation. To better investigate the coupled snow-soil moisture-vegetation system, an observing system simulation experiment (OSSE) is designed in order to explore the value of coordinated observations of these three separate, yet mutually dependent, state variables. In the experiment, a “synthetic truth” of snow water equivalent, surface soil moisture, and/or vegetation biomass is generated using the NoahMP-4.0.1 land surface model within the NASA Land Information System (LIS). Afterwards, a series of hypothetical sensors with different orbital configurations is prescribed in order to retrieve snow, soil moisture, and vegetation. The ground track and footprint of each sensor is approximated using the Trade-space Analysis Tool for Constellations (TAT-C) simulator. A space-time subsampler predicated on the output from TAT-C is then applied to the synthetic truth. Furthermore, a hypothesized amount of observation error is injected into the synthetic truth in order to yield a realistic synthetic retrieval for each of the hypothetical sensor configurations considered as part of this dissertation. The synthetic retrievals are then assimilated into the NoahMP-4.0.1 land surface model using different boundary conditions from those used to generate the synthetic truth such that the differences between the two sets of boundary conditions serve as a realistic proxy for real-world boundary condition errors. A baseline Open Loop simulation where no retrievals are assimilated is conducted in order to evaluate the added utility associated with assimilation of one (or more) of the synthetic retrievals. The impact of the assimilation of a given suite of one or more retrievals on land surface model estimates of snow, soil moisture, vegetation, and runoff serve as a numeric laboratory in order to assess which sensor(s), either separate or in a coordinated fashion, yield the most utility in terms of improved model performance. The results from this OSSE show that the assimilation of a single type of retrieval (i.e., snow or soil moisture or vegetation) may only improve the estimation of a small part of the snow-soil moisture-vegetation system, but may also degrade of other parts of that same system. Alternatively, the assimilation of more than one type of retrieval may yield greater benefits to all the components of the snow-soil moisture-vegetation system, because it yields a more complete, holistic view of the coupled system. This OSSE framework could potentially serve as an aid to mission planners in determining how to get the most observational “bang for the buck” based on the myriad of different sensor types, orbital configurations, and error characteristics available in the selection of a future terrestrial freshwater mission.Item Estimating terrestrial water budget components across high mountain Asia using remote sensing, data assimilation, and machine learning(2021) Ahmad, Jawairia; Forman, Barton A.; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Contemporary studies have predicted a vulnerable future for key water budget components across high mountain Asia (HMA) and the adjoining areas. Considering the regional population and its dependence on agrarian economies, it is imperative that efforts be channelized towards improving the estimation of the hydrologic cycle across HMA. In this study, data assimilation methods were employed to assimilate remotely-sensed observations into land surface models to improve snow mass, soil moisture, and runoff estimates. The NASA Land Information System was used to simulate the hydrologic cycle across HMA and the adjoining areas using the Noah-MP land surface model. In an effort to improve snow mass estimation, passive microwave brightness temperature spectral differences (∆Tb) from the Advanced Microwave Scanning Radiometer-2 (AMSR2) were assimilated into Noah-MP snow mass estimates. Support vector machine regression, a supervised machine learning technique, was used as the observation operator to map the geophysical states into the observed ∆Tb space. Evaluation of the assimilation routine highlighted the decrease in domain-wide snow mass bias. The assimilation framework proved to be more effective during the (dry) snow accumulation season resulting in decreased snow mass bias and RMSE at 76% and 58% of the comparative locations, respectively. Diagnostic metrics such as the innovation sequence were studied to assess the snow-related observation error characteristics of AMSR2 ∆Tb. To improve the spatiotemporal variability of modeled soil moisture estimates, Soil Moisture Active Passive (SMAP) soil moisture retrievals were assimilated into Noah-MP. Assimilation was carried out using bias corrected (via CDF-matching) and raw (without CDF-matching) SMAP retrievals. Comparison against in-situ soil moisture measurements across the Tibetan Plateau highlighted the improvement in modeled soil moisture with reductions in mean bias and RMSE by 8.4% and 9.4%, respectively, even though assimilation occurred during <10% of the total study period across the Tibetan Plateau. More importantly, SMAP retrieval assimilation corrected biases that were generated due to unmodeled hydrologic phenomenon (i.e., surface irrigation associated with agricultural production). Improvements in soil moisture translated into changes in the modeled evapotranspiration. Further, the improvement in fine-scale (0.05 degree) modeled soil moisture estimates by assimilating coarse-scale soil moisture retrievals (36 km) indicated the potential of the described methodology for soil moisture estimation over data scarce regions. Soil moisture assimilation also increased the gridded total runoff (particularly baseflow) and volumetric streamflow across irrigated areas; however, limited impact was noted in terms of volumetric streamflow along high-flow river tributaries. In this study, data assimilation was leveraged to advance contemporary land surface modeling of the terrestrial water budget components across HMA. The study objectives explored how assimilation systems could be used to improve critical geophysical state estimation for a better informed future of regional water resources.Item Estimation of Terrestrial Water Storage in the Western United States Using Space-based Gravimetry, Ground-based Sensors, and Model-based Hydrologic Loading(2020) Yin, Gaohong; Forman, Barton A; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Accurate estimation of terrestrial water storage (TWS) is critically important for the global hydrologic cycle and the Earth's climate system. The space-based Gravity Recovery and Climate Experiment (GRACE) mission and land surface models (LSMs) have provided valuable information in monitoring TWS changes. In recent years, geodetic measurements from the ground-based Global Positioning System (GPS) network have been increasingly used in hydrologic studies based on the elastic response of the Earth's surface to mass redistribution. All of these techniques have their own strengths and weaknesses in detecting TWS changes due to their unique uncertainties, error characteristics, and spatio-temporal resolutions. This dissertation investigated the potential of improving our knowledge in TWS changes via merging the information provided by ground-based GPS, GRACE, and LSMs. First, the vertical displacements derived from ground-based GPS, GRACE, and NASA Catchment Land Surface Model (Catchment) were compared to analyze the behavior and error characteristics of each data set. Afterwards, the ground-based GPS observations were merged into Catchment using a data assimilation (DA) framework in order to improve the accuracy of TWS estimates and mitigate hydrologic state uncertainty. To the best of our knowledge, this study is the first attempt to assimilate ground-based GPS observations into an advanced land surface model for the purpose of improving TWS estimates. TWS estimates provided by GPS DA were evaluated against GRACE TWS retrievals. GPS DA performance in estimating TWS constituent components (i.e., snow water equivalent and soil moisture) and hydrologic fluxes (i.e., runoff) were also examined using ground-based in situ measurements. GPS DA yielded encouraging results in terms of improving TWS estimates, especially during drought periods. Additionally, the findings suggest a multi-variate assimilation approach to merge both GRACE and ground-based GPS into the LSMs to further improve modeled TWS and its constituent components should be pursued as a new and novel research project.Item Uncertainty quantification of a radiative transfer model and a machine learning technique for use as observation operators in the assimilation of microwave observations into a land surface model to improve soil moisture and terrestrial snow(2020) Park, Jongmin; Forman, Barton A; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Soil moisture and terrestrial snow mass are two important hydrological states needed to accurately quantify terrestrial water storage and streamflow. Soil moisture and terrestrial snow mass can be measured using ground-based instrument networks, estimated using advanced land surface models, and retrieved via satellite imagery. However, each method has its own inherent sources of error and uncertainty. This leads to the application of data assimilation to obtain optimal estimates of soil moisture and snow mass. Before conducting data assimilation (DA) experiments, this dissertation explored the use of two different observation operators within a DA framework: a L-band radiative transfer model (RTM) for soil moisture and support vector machine (SVM) regression for soil terrestrial snow mass. First, L-band brightness temperature (Tb) estimated from the RTM after being calibrated against multi-angular SMOS Tb's showed good performance in both ascending and descending overpasses across North America except in regions with sub-grid scale lakes and dense forest. Detailed analysis of RTM-derived L-band Tb in terms of soil hydraulic parameters and vegetation types suggests the need for further improvement of RTM-derived Tb in regions with relatively large porosity, large wilting point, or grassland type vegetation. Secondly, a SVM regression technique was developed with explicit consideration of the first-order physics of photon scattering as a function of different training target sets, training window lengths, and delineation of snow wetness over snow-covered terrain. The overall results revealed that prediction accuracy of the SVM was strongly linked with the first-order physics of electromagnetic responses of different snow conditions. After careful evaluation of the observation operators, C-band backscatter observations over Western Colorado collected by Sentinel-1 were merged into an advanced land surface model using a SVM and a one-dimensional ensemble Kalman filter. In general, updated snow mass estimates using the Sentinel-1 DA framework showed modest improvements in comparison to ground-based measurements of snow water equivalent (SWE) and snow depth. These results motivate further application of the outlined assimilation schemes over larger regions in order to improve the characterization of the terrestrial hydrological cycle.Item Predicting Low Probability Streamflow Using Lidar Data and Hydraulic Geometry(2019) Mardones, Javier; Brubaker, Kaye L; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Predicting stream flow is essential for safe and economic planning and design of hydraulic structures. This study uses the observed channel cross-section from LiDAR data and physical concepts of shear stress to estimate bankfull discharge (Qbf). Assuming that Qbf is the median of the annual peak flow distribution, a 2-parameter Extreme Value Type I distribution was fitted to predict discharge to a 200-year return period. The method was compared with gauged sites in low-order streams (less than 90-meter bankfull width) resulting in SE/SY=1.31 for Qbf and SE/SY=1.90 for the 200-year return period discharge; model precision is poor. However, the relative bias (-15\% to +15\%) demonstrates that on average results are similar to gauged data. Relationships between flow and channel geometry assure a quick way to estimate stream data and can serve as a tool used prior to applying conventional hydrologic methods such as flow routing and regional regression equations.Item ANALYSES OF INFLUENTIAL FACTORS FOR ACCURATE DETERMINATION OF PEAK RATE FACTORS AND TIMES TO PEAK OF UNIT HYDROGRAPHS(2017) Zhao, Tianming; McCuen, Richard H.; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Despite the availability of a number of sophisticated hydrologic models, the unit hydrograph (UH) is still one of the most widely used models for computing runoff hydrographs. The two-parameter gamma UH can be fully characterized by two parameters: the peak rate factor (PRF) and the time to peak (tpUH). Currently, obtaining accurate estimates of UH parameters is still a problem, especially for ungauged watersheds. The goal of this research was to analyze factors that influence estimates of UH parameters and to develop general guidelines that can assist in estimating UH parameters more accurately. A calibration model was developed for evaluating PRFs and tpUHs simultaneously from rainfall-runoff data. The effects of various influential factors were identified and investigated based on analyses of both synthetic and measured rainfall-runoff data. Results showed that the accuracy of calibrated UH parameters is affected by the rainfall characteristics, the time offset, the nonuniformity of rainfall, the extent of nonlinear watershed processes, and the flexibility of the gamma probability distribution function. Guidelines were developed to assist UH users in interpreting the calibration results and calibrating UH parameters more accurately.Item A DIAGNOSTIC DECISION SUPPORT SYSTEM FOR SELECTING BEST MANAGEMENT PRACTICES IN URBAN/SUBURBAN WATERSHEDS(2015) Wang, Yan; Montas, Hubert J; Brubaker, Kaye L; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Best Management Practices (BMPs) have become the most effective way to mitigate the non-point source pollution (NPS) problems. Much attention has been paid on NPS in rural areas, where agricultural activities increase the nutrients, toxics, and sediments in surface water. Urban and suburban areas are also major contributors of NPS, largely due to stormwater. For watersheds bearing various soil types and land uses, a single type of BMP cannot be the panacea to all stormwater and related water quality problems. There is a need for a series of spatially distributed small-scale BMPs aimed at reducing flow volume and improving urban stormwater quality. This research seeks to develop a Diagnostic Decision Support System (DDSS) for urban BMP selection. The process-based distributed hydrologic model, Soil and Water Assessment Tool (SWAT), was used to simulate the hydrologic processes, estimate water quality variables, and to model the urban BMPs. The DDSS consists of three parts: a Hotspot Identifier, which locates the water quality and quantity hotspots; a Diagnostic Expert System (DES), which identifies the most likely physical reasons for excessive pollutants; and a Prescriptive Expert System (PES), which selects a proper set of spatially distributed BMPs. SWAT was calibrated and validated first to simulate pre-BMP watershed responses. The DDSS was then applied for BMP recommendation. The prescribed BMPs were modeled back into SWAT to quantify their effectiveness. Total Cost for BMP implementation was calculated as a function of BMP coverage area, BMP numbers and types, and residents' preferences. Protocols for urban BMP modeling were developed based on the BMPs' mechanism and the hydrologic processes involved. The DDSS was tested in Watts Branch, a small urban watershed in metropolitan Washington D.C., and Wilde Lake, a suburban watershed in Columbia, MD. Comparisons were carried out in terms of hotspots distribution and BMP recommendation between the two study areas. The hotspots identified and BMPs prescribed by the DDSS were also examined under future climate scenarios. The prescribed BMPs and GIS maps will be useful in agency-level decision making and in developing appropriate educational material for residents and the general public.Item EVALUATION OF THE NASA MICROWAVE RADIATIVE TRANSFER MODEL FOR SOIL MOISTURE ESTIMATION USING AQUARIUS BRIGHTNESS TEMPERATURE OBSERVATIONS OVER THE CONTINENTAL UNITED STATES(2014) Tarik, Saad Bin; Forman, Barton A; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The implications of near-surface soil moisture (~5 cm) variability in land surface processes and land-atmosphere interactions is important in regional and global scale climatology since it controls the partitioning of precipitation and radiation fluxes that play a crucial role in dictating weather and climate. Passive microwave (PMW) remote sensing is an increasingly popular approach to measure soil moisture because of its global coverage of the Earth. This study evaluates the performance of the NASA Goddard Earth Observing System, Version 5 (GEOS-5) radiative transfer model (RTM) using Aquarius brightness temperature (Tb) observations with the eventual goal of integrating the RTM into a data assimilation (DA) framework for the purpose of improved soil moisture estimation. Statistics were calculated from two plus years of observations across different climate regions of the United States. Seasonal variations of soil moisture were also investigated. Results suggest the RTM reasonably reproduces Aquarius Tbs, but that systematic biases exist, which must be mitigated prior to inclusion into the DA framework.Item SENSITIVITY ANALYSIS OF MACHINE LEARNING IN BRIGHTNESS TEMPERATURE PREDICTIONS OVER SNOW-COVERD REGIONS USING THE ADVANCED MICROWAVE SCANNING RADIOMETER(2014) Xue, Yuan; Forman, Barton; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Snow is a critical component in the global energy and hydrologic cycle. Further, it is important to know the mass of snow because it serves as the dominant source of drinking water for more than one billion people worldwide. Since direct quantification of snow water equivalent (SWE) is complicated by spatial and temporal variability, space-borne passive microwave SWE retrieval products have been utilized over regional and continental-scales to better estimate SWE. Previous studies have explored the possibility of employing machine learning, namely an artificial neural network (ANN) or a support vector machine (SVM), to replace the traditional radiative transfer model (RTM) during brightness temperatures (Tb) assimilation. However, we still need to address the following question: What are the most significant parameters in the machine-learning model based on either ANN or SVM? The goal of this study is to compare and contrast sensitivity analysis of Tb with respect to each model input between the ANN- and SVM-based estimates. In general, the results suggest the SVM (relative to the ANN) may be more beneficial during Tb assimilation studies where enhanced SWE estimation is the main objective.