Current Research Projects

DOE, $999,987, 2024-2027 (Co-PI) 

The Coastal Bend Climate Resilience Center (CB-CRC) presents a convergent approach for basic and applied research and capacity building aimed at advancing the understanding of the drivers and impacts of hydrologic multi-hazards to support equitable adaptation planning and hazard mitigation in the Texas Coastal Bend. The specific objectives of the CB-CRC are to (1) leverage advanced earth systems simulations, multivariate and multisite downscaling approaches, and local-scale hydrologic, hydraulic, and hydrodynamic impact models to uncover the relationships between compound event drivers and hazards and predict their evolution in the future climate, (2) co-develop adaptation strategies with stakeholder partners and test these strategies within the model framework to assess their efficacy at providing effective and equitable risk reduction, and (3) engage in meaningful outreach with stakeholders and community groups to identify information needs, disseminate project outputs, and build capacity to apply project outputs to inform decision making.

USBR, $399,986, 2024-2026 (PI) 

In partnership with the Nueces River Authority, this project aims to generate medium- and long-range predictions of storage levels of the Nueces reservoir system. The project will use a combination of hydrological modeling, remote sensing, geographical information system, and publicly available datasets. A WebGIS interface will also be developed for the visualization and distribution of forecasts and data products and transform the project outputs into a decision support system. This project will allow decision makers, stakeholders, and the public to assess data through interactive online maps and generate recommendations on estuary inflow management and municipal drought contingency.

NSF, $599,826, 2023-2026 (Co-PI) 

The infiltration and inflow problem can lead to sewer overflows and have serious consequences for public health and the environment. There is a lack of efficient and cost-effective methods for the modeling of infiltration and inflow and automatic anomaly detection and prediction. The proposed research project bridges the gap between AI research and physical sensing systems, enabling putting AI into practice. The nature of an urban wastewater system, characterized by its temporal, spatial, and topological properties, can be represented as a graph with manholes represented as nodes and pipes as edges. Graph Neural Networks (GNN) are particularly well-suited for modeling graph relations with node embeddings preserving node features and edges carrying flow messages pass to each node. The GNN surrogate model will be augmented by an in-situ water pressure monitoring system that leverages emerging energy-harvesting technology for sustainable in-situ monitoring. Mapping the urban wastewater system into a GNN surrogate model, augmented with a physical sensing system, enables infiltration and inflow anomaly detection, predictions of cascading impacts with the GNN backbone, and taking corrective actions before there is harm to public health and safety or the environment.

TGLO, $99,819, 2023-2025 (Co-PI) 

This project will develop intelligent sensors conducting sensing, in-situ risk evaluation, and real-time prewarning sustainably for coastal communities to make thorough preventative measures. This project will develop a smart and self-sustaining early-warning system for coastal flooding with state-of-the-art low-power Tiny Machine Learning (TinyML) and energy harvesting solutions. In particular, we will embed sensors, Tiny ML, and connectivity into energy-harvesting-powered IoT devices for sustainable, accurate, and real-time monitoring, prediction, and pre-warning of flooding paths and risk levels. The outcomes of this project would bring great social-and-economic benefits to various communities. The techniques outcomes will be intellectually compelling and appealing to students as they bring together theoretical and engineering concepts. 

NSF, $199,966, 2021-2024 (PI) 

The description of watershed imperviousness has relied on the concepts of total impervious area and effective impervious area. Both are lumped and static measures that emphasize the extent of imperviousness, but they lack consideration of the spatial pattern of various impervious surfaces. This knowledge gap affects the reliability of hydrological models that tend to conceptualize pervious and impervious surfaces as two lumped components with no interactions. This project aims to answer: 1) How do spatial heterogeneity and hydrological connectivity of impervious surfaces affect stormwater runoff in small urban watersheds? 2) How can the understanding of this mechanism benefit the modeling of stormwater runoff? This project conceptualizes the impervious source area of stormwater runoff as a combination of a constant component and a dynamic component with its hydrological connectivity affected by pervious surfaces. Using Corpus Christi in south Texas as an urban laboratory, the project efforts consist of three phases: (i) UAS-aided data collection, (ii) characterization of hydrological connectivity, and (iii) development of a new multi-component curve number model. The findings of this project will lead to a spatially explicit understanding of the effects of the impervious area on stormwater runoff processes and provide an improved scientific basis for developing reliable stormwater management strategies. 

NSF, $4,995,354, 2021-2026 (Co-PI, Lead of Subproject 2) 

The Centers of Research Excellence in Science and Technology (CREST) program supports the enhancement of research capabilities of minority-serving institutions through the establishment of centers that effectively integrate education and research. This center articulates three research subprojects addressing the challenges of coastal resilience by developing new approaches that integrate remote and autonomous sensing with geospatial computing, artificial intelligence/deep learning, and big data analytics for comprehensive coastal zone monitoring at different scales. The focus is on improving resiliency to extreme coastal hazards and episodic and persistent events (e.g., hurricane and sea-level rise). Subproject 1 develops new approaches integrating remote and autonomous sensing with geospatial computing and artificial intelligence to improve coastal zone monitoring and resiliency decision-making. Subproject 2 contributes to the understanding of the urban water cycle and the resilience of water infrastructure through integrated characterization, simulation, and assessments. Subproject 3 investigates how emerging data sources and advanced geospatial computing can be applied to evaluate, assist, and improve a coastal community’s physical, behavioral, and social health after disasters.