Authors: Md. Shahabul Alam, Ryan Johnson, Steven Burian – Alabama Water Institute, The University of Alabama
Presentation Type: Poster and lightning talk
Title: Community Streamflow Evaluation System: Developing a Python-based tool and Tethys-based web app
Abstract: The field of hydrological sciences is experiencing rapid advancement, fueled by enhanced computational capabilities, richer datasets, and collaborative open-source initiatives. Integrating physics with artificial intelligence/machine learning (AI/ML) in nationwide hydrological modeling encounters challenges, underscoring the importance of a consistent and standardized evaluation framework. The Community Streamflow Evaluation System (CSES), in alignment with the objectives of NOAA and the Cooperative Institute for Research to Operations in Hydrology (CIROH), simplifies the evaluation process and effectively overcomes constraints related to data acquisition, such as accessing NWIS observations or NWC/NWM predictions. This tool aims to bolster large-scale hydrological modeling efforts, facilitating enhancements in streamflow prediction accuracy. This study seeks to investigate two avenues: (i) the development of a Python-based tool to facilitate advancements in hydrological modeling, and (ii) the creation of a web application designed to disseminate model outcomes and proficiency to various users of hydrological models, including water resources managers, public water utilities, and stormwater management entities, among others. Our tool reveals that while NWM v2.1 performs well in headwater catchments, its accuracy diminishes in other basin areas. Future steps would include refining the Tethys-CSES app workflow and integrating data from other hydrological models like NWM v3.0.