Do Lateral Flows Matter for the Hyperresolution Land Surface Modeling?

Abstract

Hyperresolution land surface modeling provides an unprecedented opportunity to simulate locally relevant water and energy cycle, but lateral surface and/or subsurface flows that are essential at fine scale are often neglected by most one‐dimensional land surface models (LSMs). To analyze effects of lateral flows across scales, a Conjunctive Surface‐Subsurface Process model, which considers soil moisture‐surface flow interaction and quasi‐three‐dimensional subsurface flow, is implemented over a mountainous HyperHydro test bed in southwestern USA at different resolutions. Validation over more than 70 International Soil Moisture Network stations shows that there are significant improvements in soil moisture simulations from 30 km to 4 km as finer soil property and precipitation data are used, with correlation increased by 5%–16% and error decreased by 5%. Lateral surface flow has a significant influence on surface soil moisture and ground evaporation even at coarse resolution. Effect of lateral subsurface flow on soil moisture is nontrivial at 1 km or finer resolution especially over wet areas. At 100 m resolution, topography‐induced lateral subsurface flow causes drier peaks and wetter valleys, decreases latent heat by 8% at peaks, while increases it by 12% at valleys. Furthermore, influences of lateral subsurface flow on ground evaporation and vegetation transpiration are more significant during dry season due to a stronger coupling between soil moisture and evapotranspiration. Therefore, it is worthy to incorporate lateral flow processes in hyperresolution LSMs to better represent water and energy heterogeneity even with limited hyperresolution meteorological and surface data.

High‐Resolution Land Surface Modeling of Hydrological Changes Over the Sanjiangyuan Region in the Eastern Tibetan Plateau: 1. Model Development and Evaluation

“…The comparison between CSSPv2 simulations driven by GLDAS1 and high‐resolution forcings (red bars) shows the importance of meteorological forcings in soil moisture simulation (Figure ). This is also consistent with our previous study, where gradual improvements were found by incorporating high‐resolution meteorological and surface data information (Ji et al, ). In short, high‐resolution forcing data are critical for a successful high‐resolution land surface modeling.…”

“…The LSM used in this study is the Conjunctive Surface-Subsurface Process (CSSP) model (Ji et al, 2017;Liang et al, 2012;Yuan & Liang, 2011). The CSSP model was developed from the Common Land Model (Dai et al, 2003(Dai et al, , 2004, but it has better representations in lateral surface and subsurface hydrological processes and their interactions (Choi et al, 2013).…”

“…Generally, CSSP model includes a quasi-three-dimensional subsurface soil water transport model (Choi et al, 2007), a one-dimensional diffusive surface water model (Choi, 2006), and a one-dimensional groundwater model (Yuan & Liang, 2011). The lateral surface and subsurface flows were found to be nontrivial for high-resolution land surface modeling (Ji et al, 2017).…”

“…The latter direction requires developing high-resolution LSMs that can provide locally relevant information (Bierkens et al, 2015). As LSMs go into high or even hyper resolution (Wood et al, 2011), many hydrological processes that are neglected in coarse resolution Earth system models may have significant influence on the simulations (Melsen et al, 2016), such as the lateral flows, overland flow, ponding, and surface-subsurface interactions (Ji et al, 2017;Maxwell & Condon, 2016;Niu et al, 2014).…”

Ecohydrological dynamics in the Alps: Insights from a modelling analysis of the spatial variability

“…The T&C model simulates lateral surface and subsurface water flows, which are currently neglected in many land surface models. Recent studies showed that models incorporating these processes (e.g., Ivanov et al, ) can simulate the heterogeneity of water and energy fluxes more realistically (Ji, Yuan, & Liang, ; Maxwell & Condon, ). This model advantage is reflected in the simulated patterns of soil effective saturation and subsequently on the spatial patterns of transpiration, LAI, and NPP (Figures , , S6, S7, and S8).…”