The Next Generation Water Resources Modeling Framework: Open Source, Standards Based, Community Accessible, Model Interoperability for Large Scale Water Prediction

The NOAA-National Weather Service, Office of Water Prediction (OWP), working with federal water prediction partners, has designed and developed the Next Generation Water Resources Modeling Framework (Nextgen). The motivating objectives of the framework include increasing water resources model interoperability, intercomparison, testing of research hypotheses, and deploying into operations science-driven, evidence-based models while enabling rapid adoption of research advances. The current operational National Water Model (NWM) applies a single model formulation, using calibration and regionalization to emphasize stormflow generation process. The NWM performance varies regionally. Calibration has improved in some regions but not all. The literature supports the hypothesis that appropriately applied models formulated for specific dominant local processes consistently outperform general models (uniqueness of place); given the lack of a comprehensive stormflow generation theory, there is no one model to rule them all. By 2024, the NWM will consist of a particular configuration of the Nextgen framework. Nextgen is model agnostic with maximum flexibility, allowing the framework to adapt as models, data sources, and water prediction needs change. Nextgen uses unifying standards. The Basic Model Interface (BMI) model coupling standard for a common architecture that avoids duplication and promotes interoperability. The WaterML 2.0 HY_FEATURES data model provides standard description of surface water features. The framework, which is written in C++ and supports parallel computing, links models written in C, C++, Fortran and Python. Nextgen development occurs transparently using open-source practices that promote code reuse and development efficiency and encourage participation by our federal partners and the research community. The Nextgen code exists on the NOAA OWP GitHub repository, and we invite community involvement. The repository includes examples, sample data sets, and documentation explaining the step-by-step process to adapt models to use the BMI coupling standard and interoperate with the framework. User friendliness is a goal with a target of two weeks for a new employee or graduate student with programming skills to add new domain science capabilities to Nextgen.