
HydroCAL: a computationally efficient continuous ISSHM developed through the Cellular Automata paradigm
Please login to view abstract download link
This study presents the new features of HydroCAL (Hydrological Cellular Automata Layer) [1], an innovative spatially distributed Integrated Surface-Subsurface Hydrological Model (ISSHM). HydroCAL is developed entirely within the Cellular Automata (CA) paradigm, utilising the latest available version of the OpenCAL software library [2]. The model addresses the computational challenges often associated with ISSHMs, which are crucial for accurately simulating the complex interactions between surface and subsurface hydrological processes. The CA paradigm offers significant advantages, particularly in reducing computational loads through parallelisation and advanced optimisation techniques. A key feature is its asynchronous functionality [3], which, for specific time steps, selectively identifies domain areas requiring to be updated, therefore lowering computational costs while maintaining numerical accuracy. This note introduces an updated version of HydroCAL, redesigned to be compatible with the latest version (2.0) of OpenCAL. This update introduces capabilities for transparently transforming the implemented serial version into a CUDA API version, allowing users to fully leverage the power of General Purpose Graphics Processing Units (GPGPUs). HydroCAL has been tested against classical benchmark study cases from existing literature [4], validating its reliability in replicating established hydrological processes. Additionally, the model has been applied to real-world events within a small headwater catchment covering 7 km², demonstrating its ability to reproduce streamflow dynamics with high statistical performance metrics accurately. Furthermore, HydroCAL provides detailed, fully spatially distributed outputs for multiple hydrological variables across both surface and subsurface domains, offering comprehensive insights into watershed behaviour. HydroCAL implementation in the new environment allows its efficient application as a continuous model incorporating evapotranspiration processes, enabling the investigation of hydrological cycle evolution encompassing extended time ranges even of hydrological years.