Large-sample assessment of spatial scaling effects of the distributed wflow_sbm hydrological model shows that finer spatial resolution does not necessarily lead to better streamflow estimates

Abstract

Abstract. Distributed hydrological modelling moves into the realm of hyper-resolution modelling. This results in a plethora of scaling related challenges that remain unsolved. In light of model result interpretation, finer resolution output might implicate to the user an increase in understanding of the complex interplay of heterogeneity within the hydrological system. Here we investigate spatial scaling in the realm of hyper-resolution by evaluating the streamflow estimates of the distributed wflow_sbm hydrological model based on 454 basins from the large-sample CAMELS data set. Model instances were derived at 3 spatial resolutions, namely 3 km, 1 km, and 200 m. The results show that a finer spatial resolution does not necessarily lead to better streamflow estimates at the basin outlet. Statistical testing of the objective function distributions (KGE score) of the 3 model instances show only a statistical difference between the 3 km and 200 m streamflow estimates. However, results indicate strong locality in scaling behaviour between model instances expressed by differences in KGE scores of on average 0.22. This demonstrates the presence of scaling behavior throughout the domain and indicates where locality in results is strong. The results of this study open up research paths that can investigate the changes in flux and state partitioning due to spatial scaling. This will help further understand the challenges that need to be resolved for hyper resolution hydrological modelling.

The eWaterCycle platform for open and FAIR hydrological collaboration

“…Current hydrological modeling studies often include many catchments in their analysis. The platform fully supports those more complex studies, as is shown in Aerts et al (2021), where the impact of different spatial resolutions of the same model are studied for 454 catchments from the CAMELS dataset.…”

HydroMT: Automated and reproducible model building and analysis

“…Several recent applications have already used HydroMT. It has been used to build instances of the hydrological Wflow model (Imhoff et al, 2020;Verseveld et al, 2022) for climate change impact studies (Sperna Weiland et al, 2021); for a large sample assessment of spatial scales in hydrological modeling (Aerts et al, 2021); to estimate water balance components based on global data in a data-scarce area (Rusli et al, 2021); and to build instances of the hydrodynamic SFINCS model (Leijnse et al, 2021) to simulate compound flood hazard from (global) datasets in any coastal delta globally (Eilander et al, 2022). While the package is sufficiently general to support model software outside the domain of hydro models, the focus of many workflows is on typical hydrological data and parameters and hence hydro models benefit most from the tool.…”

Hyperlocal Modeling: Reality or Wishful Thinking?

“…Generalmente, esta incerteza proviene de la falta de conocimiento sobre estos procesos y parámetros en resoluciones tan detalladas(Beven & Cloke, 2012). La incertidumbre de la medición se refiere a la duda que existe sobre el resultado de cualquier medición, especialmente cuando son adquiridas de forma indirecta, como en conjuntos de datos de alta resolución sobre topografía, uso del suelo, geología y propiedades del suelo, que pueden contener imprecisiones debido a métodos de teledetección, modelos de superficie terrestre y técnicas estadísticas de reducción de escala(Aerts et al, 2021).En relación a la incertidumbre sobre los métodos matemáticos utilizados en el modelo, estas pueden ocurrir por la variación de la resolución espacial, generando una necesidad por modelos conjuntos de múltiples escalas para mejorar la capacidad predictiva(Aerts et al, 2022). Por ejemplo, modelos a una escala de 0,5° (alrededor de 50 km en el ecuador) asumen que la demanda de agua se satisface localmente, pero a resoluciones de 10 km o más, es necesario considerar la redistribución del agua entre celdas desde puntos de extracción hasta lugares críticos de consumo o a través de flujos subterráneos(Krakauer et al, 2014).Más allá de la incertidumbre, los modelos hidrológicos hiper-locales pueden fallar en cuanto a sus resultados.…”