The exchange of water and heat between the Earth and its atmosphere determines climate zones and ecosystems, which in turn influence where essential human activities take place. While land surface models are an effective tool to simulate these systems, they may not represent some key physical rules that control land processes. However, a research team from Japan has come up with a new strategy to integrate hillslope water dynamics with distribution of different vegetation types at fine scales across Africa.

The findings, recently published in Water Resources Research, demonstrated that the simulation was substantially more accurate in making predictions. This included estimating the division of precipitation into soil moisture and evaporation, alongside the release of water from plants and runoff. This new approach can be used to inform more sustainable land and water management in natural and human systems in the future.

"The empirical knowledge tells us that when rain falls, the subsequent water movement in terms of runoff, evaporation and transpiration is strongly influenced by local topography and vegetation. However, previous land surface models rarely account for these factors," said Shuping Li, the lead author of the study. "Accurate simulations are essential to understand and predict the effect of global change on ecosystems and human society."

The team developed a new scheme that links potential water movement with the distribution patterns of vegetation, and implemented this in a land surface model to represent realistic water dynamics and vegetation heterogeneity along hillslopes. As part of this, the model divided the land surface into different height band structures to simulate water flow on different parts of hillslopes.

There were four experimental settings, each with a different arrangement of slope and vegetation, which used standardized frameworks of land surface models. The simulations were applied in four desert and forest ecosystems across the African continent over a 10-year period.

"Applying the updated model, we found that simulated runoff, evapotranspiration and soil moisture were substantially influenced by the combination of hill-to-valley water dynamics and hillslope vegetation, although this varied with the type of ecosystem," explained senior author Dai Yamazaki. "Our complex models reproduced results that were much more consistent with the observation data than the simple ones, particularly for soil moisture."

The team reported clear differences in water and energy budgets using the new models, which are suggested to be a vast improvement on previous models. It is hoped that this development will be helpful in formulating sustainable land and water management policies.

The article, "Resolving Land Cover Heterogeneity Along Hillslope Improves Simulation of Terrestrial Water and Energy Budgets", is published in Water Resources Research at DOI: 10.1029/2025WR040706.

Research Contact

Shuping Li, Project Researcher
Institute of Industrial Science, the University of Tokyo
Tel：+81-4-7136-6965
E-mail：shuping (Please add "@iis.u-tokyo.ac.jp" to the end)