The Department of Civil and Environmental Engineering at Colorado State University seeks a highly motivated Research Scientist to work under the supervision of Prof. Mazdak Arabi. The CSU One Water solutions Institute (OWSI) is recruiting a highly qualified candidate to develop deep learning techniques along with artificial intelligence (AI) using large language models for the following projects: Enhancing characterization of trends in large floods and flood forecasting: The Flood Potential Portal, developed by the U.S. Forest Service and OWSI, enhances the understanding of flood variability and quantifies design flood discharges and flood frequency relationships. The software assists practitioners with assessments to support infrastructure decisions, including designing road-stream crossings. This project develops and implements advanced ensemble machine learning to improve forecasting of temporal trends in large floods due to changes in the underlying climatic, physiographic, and ecohydrologic conditions. Physics-informed machine learning for agroecosystem modeling: This project funded by USDA focuses on developing deep learning algorithms that incorporate scientific understanding of hydrological, water quality, and soil and crop processes. Specifically, physics-informed techniques will be developed based on the SWAT+ model for agroecosystem modeling at the field to watershed scales. The deep learning model will be trained with the outputs of the calibrated and tested SWAT+ model to represent landscape-level hydrologic and water quality responses. The outputs of this landscape-level model will include crop water use, soil erosion, phosphorus runoff, water quality, and crop yield/productivity. The performance validity of the model will be corroborated using extensive observed hydrologic, water quality, and crop yield data across the study region. Urban water modeling software: Over the past decade, OWSI has developed data analytics and modeling software for integrated water and land use planning in cities. Specifically, the Polaris, IUWM, and CLASIC tools provide capacities for water supply-demand assessments, lifecycle assessment of green and gray stormwater practices, and evaluation of fit-for-purpose use of alternative water sources. This project will incorporate advanced machine learning methods and tools in Polaris, IUWM, and CLACIS tools for enhanced characterization and forecasting purposes.