Infiltration from the Pedon to Global Grid Scales: An Overview and Outlook for Land Surface Modeling

Abstract

Core Ideas Land surface models (LSMs) show a large variety in describing and upscaling infiltration. Soil structural effects on infiltration in LSMs are mostly neglected. New soil databases may help to parameterize infiltration processes in LSMs. Infiltration in soils is a key process that partitions precipitation at the land surface into surface runoff and water that enters the soil profile. We reviewed the basic principles of water infiltration in soils and we analyzed approaches commonly used in land surface models (LSMs) to quantify infiltration as well as its numerical implementation and sensitivity to model parameters. We reviewed methods to upscale infiltration from the point to the field, hillslope, and grid cell scales of LSMs. Despite the progress that has been made, upscaling of local‐scale infiltration processes to the grid scale used in LSMs is still far from being treated rigorously. We still lack a consistent theoretical framework to predict effective fluxes and parameters that control infiltration in LSMs. Our analysis shows that there is a large variety of approaches used to estimate soil hydraulic properties. Novel, highly resolved soil information at higher resolutions than the grid scale of LSMs may help in better quantifying subgrid variability of key infiltration parameters. Currently, only a few LSMs consider the impact of soil structure on soil hydraulic properties. Finally, we identified several processes not yet considered in LSMs that are known to strongly influence infiltration. Especially, the impact of soil structure on infiltration requires further research. To tackle these challenges and integrate current knowledge on soil processes affecting infiltration processes into LSMs, we advocate a stronger exchange and scientific interaction between the soil and the land surface modeling communities.

A Simple Method to Design Irrigation Rate and Duration and Improve Water Use Efficiency

“…Data corresponding to the soil infiltration capacity in the field are relatively easily accessible, by means of analytical or numerical solutions of the flow equations (Assouline, ; Vereecken et al, ), or by means of infiltrometers or single or double ring infiltration experiments (Dirksen, ; Hillel, ). It is therefore possible to characterize a parcel under interest in terms of its infiltration capacity curve.…”

“…Water infiltrates the soil at a rate that depends on the application rate, the soil properties within the profile (homogeneous soil, heterogeneity, layering, …), the initial water content distribution in the soil profile, and the lower boundary condition of the system (deep profile, drainage system, water table or impervious layer at a prescribed depth, … ; Assouline, ; Vereecken et al, ). The process of infiltration is quantitatively described by the solution of the flow equation (Richards, ), expressed in equation for the case of vertical infiltration: where θ is the soil water content; ψ , the capillary head; z , the vertical coordinate taken positive upward; t , the time; and K (ψ), the soil unsaturated hydraulic conductivity function.…”

Drivers of regional soil water storage memory and persistence

“…The data we analyzed here do not account for changes in LULC (e.g., tile drainage, soil compaction, and agricultural expansion), which contribute to homogenization of the hydrologic properties of the landscape and would further emphasize hydroclimatic forcing as the key driver of variability in soil water storage dynamics. However, Vereecken et al (2019) argue that current land surface models, like the one we used here, do not fully represent the processes and soil properties involved in infiltration, which could affect the realism of the models. However, Wang-Erlandsson, Bastiaanssen, Gao, Jägermeyr, and Senay (2016) quantified global root-zone soil water storage capacity based on satellite-based precipitation and ET data and found that the LULC types common in our study region (cropland, grassland, and deciduous broadleaf forest) have similar storage and hydrologic buffering capacities (i.e., memory).…”

KEYLINK: towards a more integrative soil representation for inclusion in ecosystem scale models. I. review and model concept

“…Continuous advances in both numerical techniques and computation power are now making it increasingly possible to perform comprehensive simulations of non-equilibrium flow processes in the vadose zone (review: Vereecken et al, 2019 ). Such simulations, especially if paired with exhaustive field data sets (e.g., by data assimilation), are vital for better understanding and quantifying the effects of heterogeneities, fractures and macropores on flow and transport at the field scale ( Van Genuchten, Leij & Wu, 1999 ; review: Šimůnek et al, 2003 ).…”

“…Challenges in predicting soil water flow and solute transport beyond laboratory scale include: soil parameterization, handling structured soils including preferential flow, handling soil heterogeneity, temporally changing properties (e.g., soil bulk density, structural properties, etc. ), and description of root water uptake (e.g., Jury et al, 2011 ; review: Vereecken et al, 2019 ). Thus, it is clear that although the importance of soil structure and water are proven, their inclusion in models is hampered because of the lack of data on soil structure and the difficulties in measuring and simulating soil water at the ecosystem scale.…”