Civil & Environmental Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/2221
Browse
7 results
Search Results
Item Machine Learning for Projecting Extreme Precipitation Intensity for Short Durations in a Changing Climate(MDPI, 2019-05-09) Hu, Huiling; Ayyub, Bilal M.Climate change is one of the prominent factors that causes an increased severity of extreme precipitation which, in turn, has a huge impact on drainage systems by means of flooding. Intensity–duration–frequency (IDF) curves play an essential role in designing robust drainage systems against extreme precipitation. It is important to incorporate the potential threat from climate change into the computation of IDF curves. Most existing works that have achieved this goal were based on Generalized Extreme Value (GEV) analysis combined with various circulation model simulations. Inspired by recent works that used machine learning algorithms for spatial downscaling, this paper proposes an alternative method to perform projections of precipitation intensity over short durations using machine learning. The method is based on temporal downscaling, a downscaling procedure performed over the time scale instead of the spatial scale. The method is trained and validated using data from around two thousand stations in the US. Future projection of IDF curves is calculated and discussed.Item DATA-DRIVEN ASSESSMENT FOR UNDERSTANDING THE IMPACTS OF LOCALIZED HAZARDS(2022) Ghaedi, Hamed; Reilly, Allison C.; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Both the number of disasters in the U.S. and federal outlays following disasters are rising. Thus, evaluating the impact of varying natural hazards on the built environment and communities rapidly and at various spatial scales is of the utmost importance. Many hazards can cause significant and repetitive economic and social damages. This dissertation is a collection of studies that broadly evaluates resilience outcomes in urban areas using data-driven approaches. I do this over three chapters, each of which explores a unique aspect of hazards and their impact on society. The first two chapters are devoted to federal disaster programs aimed at supporting recovery and building resilience. I especially seek to understand how characteristics of hazards intersect with aspects of the physical and social environment to drive federal intervention. The final chapter explores the heterogenous impacts of natural hazards in urban communities and how disparities correlate with various socioeconomic and demographic characteristics. The first two studies examine two major federal disaster programs in the U.S. – FEMA Public Assistance (PA) and FEMA National Flood Insurance Program (NFIP) – at varying spatial and temporal scales. Both leverage parametric and non-parametric statistical learning algorithms to understand how measures of hazard intensity and local factors drive federal intervention. These studies could be used by federal/state-level resource managers for planning the level of aid that may be required after a disaster. This study can also potentially be useful for decision-makers to identify the potential causes of increased disaster spending over time. In the final chapter, I evaluate the links between public transit disruptions, socioeconomic characteristics, and precipitation. By analyzing the spatial distribution and clustering of infrastructure disruptions, I identify the area(s) susceptible to a disproportionate amount of disruptions. Additionally, spatial statistical models are developed to investigate the relationship between infrastructure disruptions and the characteristics of the communities by including variables related to socioeconomic, demographics, social vulnerability, traffic volume, transit system, road connectivity, and the built environment characteristics. For the decision-makers with the goal of improving the performance and resilience of the transit system, this study can provide insight to locate critical areas impacted by such disruptions.Item Predicting flood damage using the Flood Peak Ratio and Giovanni Flooded Fraction - Code(2022-07-11) Ghaedi, Hamed; Baroud, Hiba; Ferreira, Celso; Perrucci, Daniel; Reilly, Allison; Reilly, AllisonA spatially-resolved understanding of the intensity of a flood hazard is required for accurate predictions of infrastructure reliability and losses in the aftermath. Currently, researchers who wish to predict flood losses or infrastructure reliability following a flood usually rely on computationally intensive hydrodynamic modeling or on flood hazard maps (e.g., the 100-year floodplain) to build a spatially-resolved understanding of the flood’s intensity. However, both have specific limitations. The former requires both subject matter expertise to create the models and significant computation time, while the latter is a static metric that provides no variation among specific events. The objective of this work is to develop an integrated data-driven approach to rapidly predict flood damages using two emerging flood intensity heuristics, namely the Flood Peak Ratio (FPR) and NASA’s Giovanni Flooded Fraction (GFF). This study uses data on flood claims from the National Flood Insurance Program (NFIP) to proxy flood damage, along with other well-established flood exposure variables, such as regional slope and population. The approach uses statistical learning methods to generate predictive models at two spatial levels: nationwide and statewide for the entire contiguous United States. A variable importance analysis demonstrates the significance of FPR and GFF data in predicting flood damage. In addition, the model performance at the state-level was higher than the nationwide level analysis, indicating the effectiveness of both FPR and GFF models at the regional level. A data-driven approach to predict flood damage using the FPR and GFF data offer promise considering their relative simplicity, their reliance on publicly accessible data, and their comparatively fast computational speed.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 EVALUATING MILITARY INSTALLATION OPERATIONS DEPENDENCE ON INFRASTRUCTURE USING MACHINE LEARNING AND SURVEY DATA(2021) Magoulick, Paul F; Reilly, Allison C; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Critical infrastructure on many Department of Defense (DOD) installations are increasingly threatened by extreme weather, in part due to climate change and also in part due to the location vulnerability of these assets. At the same time, they are being relied upon more to ensure overall mission success. While each installation has its unique mission, we explore the mission at one installation – marine recruit training at the U.S. Marine Corps Recruit Depot in Parris Island, South Carolina – and predict how day-to-day and extreme weather events lead to infrastructure failures through statistical modeling. We then quantify what this means for installation operability by surveying individuals tasked with mission training. The results are informative for how to allocate resources for infrastructure enhancements in a way that protects base operability in addition to base infrastructure.Item ASSIMILATION OF PASSIVE MICROWAVE BRIGHTNESS TEMPERATURES FOR SNOW WATER EQUIVALENT ESTIMATION USING THE NASA CATCHMENT LAND SURFACE MODEL AND MACHINE LEARNING ALGORITHMS IN NORTH AMERICA(2017) Xue, Yuan; Forman, Barton A.; 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. 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. To accurately estimate the depth of snow and mass of water within a snow pack across regional or continental scales is a challenge, especially in the presence of dense vegetations since direct quantification of SWE is complicated by spatial and temporal variability. To overcome some of the limitations encountered by traditional SWE retrieval algorithms or radiative transfer-based snow emission models, this study explores the use of a well-trained support vector machine to merge an advanced land surface model within a variant of radiance emission (i.e., brightness temperature) assimilation experiments. In general, modest improvements in snow depth, and SWE predictability were witnessed as a result of the assimilation procedure over snow-covered terrain in North America when compared against available snow products as well as ground-based observations. These preliminary findings are encouraging and suggest the potential for global-scale snow estimation via the proposed assimilation procedure.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.