Enhancing the HYPE model for simulating dissolved organic carbon in inland surface waters

Dissolved Organic Carbon (DOC) in inland surface waters constitutes a significant component of the global carbon cycle, responsible for over half of the carbon transport from terrestrial ecosystems to the ocean. Computational modelling provides an effective method for monitoring spatial and temporal DOC dynamics in inland surface waters, beyond the limited information from in-situ measurements that are often sparse. Numerous process-based models have been developed for simulating DOC in inland surface waters. A common model structure for inland water DOC simulation is to simulate the soil carbon and its subsequent transport to the aquatic environment. Consequently, those models tend to have a complex terrestrial carbon module and comprehensive DOC transport processes from terrestrial to aquatic ecosystems, while their aquatic carbon simulation processes are often simplified. However, such simplification lead to insufficient representation of the interactions, which limits the model capability and undermines our understanding of complete DOC processes and dynamics.

The Hydrological Predictions for the Environment (HYPE) model, a process-based semi-distributed hydrological model developed by the Swedish Meteorological and Hydrological Institute (SMHI), is capable of simulating water quantity (e.g., daily streamflow) and water quality (e.g., nitrogen, phosphorus and carbon concentrations) at various scales. The HYPE model has been validated with reported good performance across the world and it is used by SMHI to provide many operational services in entire Sweden. The HYPE model is among the models that simplify the aquatic organic carbon cycle; it only considers primary production, mineralization, and sedimentation in the DOC simulation. This study aims to enhance the organic carbon module of the HYPE model by improving its presentation of aquatic carbon processes. Specifically, we will develop inclusion of additional key carbon pools and their interactions. For instance, two algae pools (upper and lower) would be added with consideration of algae mortality; the particulate organic carbon would be included in the carbon cycle; the inorganic carbon transport from the soil profile would be considered. As a result, the enhanced HYPE model will be able to represent more detailed aquatic carbon processes. The enhanced HYPE model will be tested in the Krycklan catchment in northern Sweden and several catchments in southern Sweden. Model performance will be evaluated at different timescales with commonly used metrics such as Kling-Gupta efficiency and Nash-Sutcliffe efficiency. We will also perform detailed analyses of parameter sensitivity and model uncertainty. Our study research presents a progressive step in the modelling efforts towards a better DOC simulation and prediction of carbon transport at the catchment scale, which helps us eventually obtain a deeper understanding of DOC dynamics in inland surface waters.