Hamid Moradkhani Alton N. Scott Professor of Civil and Environmental Engineering · Director, Center for Complex Hydrosystems Research · The University of Alabama

Downscaled SMAP Product

A Downscaled Soil Moisture Active Passive (SMAP) product at 1-km resolution built using machine learning, providing remotely sensed soil moisture data at fine spatial scale.

Open the SMAP product page

This personal website presents the academic work of a hydrologist and water resources modeler whose research spans drought, flooding, climate variability, and the vulnerability of communities to extreme events. The emphasis is on computational methods that translate observations into usable forecasts, drawing on remote sensing, machine learning, and high-performance computing to make large-scale hydrologic analysis feasible. Rather than serving as a general information portal, the site organizes a faculty research program around questions of water availability, risk, and sustainable management.

The research focuses on advancing hydrologic forecasting through multi-model and ensemble approaches that explicitly account for uncertainty and model spread. Rather than relying on a single simulation, the work treats probability and inference as first-class components of prediction, exploring how probabilistic forecasts can be communicated to decision-makers who must act under uncertain conditions. Post-processing and downscaling techniques are developed to translate coarse model output into locally meaningful estimates, while multi-model averaging is used to capture structural differences among competing representations of the water cycle.

Data assimilation and inference form another pillar of the program, linking observations to model states so that predictions remain consistent with what monitoring networks actually observe. Techniques drawn from Bayesian reasoning are applied to hydrologic problems, allowing uncertainty to be quantified rather than hidden behind a single deterministic run. This line of inquiry supports improved estimates of streamflow, soil moisture, and related variables, and it underpins efforts to translate complex computational output into information that water managers can use for planning and risk assessment.

Remote sensing plays a central role in the group's methodology, providing spatially distributed observations of the land surface that complement ground-based monitoring. Satellite-derived information is combined with in-situ records and processed through data assimilation systems, producing analyses of drought conditions, flooding, and hydrologic variability across space and time. Interactive maps and data analysis tools are developed to make these products accessible, allowing researchers and the public to explore how hydrologic conditions evolve and how extreme events propagate through complex basin systems.

The research theme of complex hydrosystems emphasizes interactions among processes that are often studied separately, such as precipitation, evapotranspiration, runoff generation, and flow routing. Characterizing and quantifying these couplings is essential for understanding how changes in climate and land use ripple through catchments. The work examines the recovery of systems following disturbances, the efficiency of water resource use, and the behavior of extreme events, building insight into how interconnected components of the water cycle shape outcomes for the people and ecosystems that depend on them.

Machine learning and data-driven analytics are harnessed alongside process-based physical modeling, offering complementary strengths for pattern recognition, prediction, and rapid experimentation. Rather than treating these approaches as substitutes, the research explores how computational advances can improve forecast performance and how hybrid methods can better represent uncertainty in hydrologic applications. High-performance computing resources are used to support computationally intensive simulations, enabling multi-model ensembles and large-scale inference that would otherwise be impractical, and broadening the range of questions the group can address.

The site also serves as an archive of the group's scholarly output, including journal papers, conference contributions, and other publications documenting methodological advances in hydrology and water resources. Findings are presented with an emphasis on practical consequence: how probabilistic information, downscaled projections, and assimilation-based analyses can support sustainable water management. Access to these materials reflects an intent to share methods and results with the broader scientific community, providing a foundation for collaboration and for further development of predictive tools.

Supporting the research program are the people who carry it out, along with courses and structured training opportunities that build capacity in hydrologic modeling and data analysis. A dedicated team page describes the group's composition and contributions, while instructional offerings connect classroom learning with the computational techniques used in research. Together, these elements illustrate a workflow in which teaching, methodological development, and applied forecasting reinforce one another, cultivating the next generation of practitioners equipped to work with uncertainty and large environmental datasets.

Overall, the website documents a research agenda aimed at understanding complex interactions within the water cycle and at advancing the reliability of predictive hydrology. The connecting thread is that better forecasting requires both better computation and better treatment of uncertainty, and that observational data must be integrated carefully if models are to be trusted in practice. By publishing research, maintaining interactive resources, and sharing data and analysis tools, the program contributes to improved water resource management and to a clearer picture of how hydrologic systems respond to a changing climate.

USDM and IDI

USDM and IDI describe multivariate-based integrated drought monitoring systems used for drought assessment and analysis.

Open the USDM and IDI page