Layered Green and Ampt Infiltration with Redistribution, with Talbot - Ogden groundwater bi-directional coupling

Simulating precipitation partitioning into infiltration and runoff presents a substantial challenge for hydrologic modelers. Soils with different physical properties in layers are ubiquitous. Accurate calculation of infiltration into layered soils with a near surface water table typically requires numerical solution of the Richards equation. However, the Richards equation can require considerable computational effort and can fail to converge within desired mass balance tolerances. The recently developed Layered Green and Ampt Infiltration with Redistribution (LGAR) method calculates infiltration in deep, well-drained, layered soils. LGAR successfully emulated infiltration and runoff simulations as generated by a popular Richards equation solver. The primary limitation of the LGAR method is that it can only be used in situations where groundwater does not affect near-surface soil moisture. To remedy this, we extended the LGAR approach to simulate groundwater - surface water interactions by coupling it to a multilayer extension of the Talbot - Ogden (TO) model that allows for bi-directional movement of water between the vadose zone and groundwater table. In doing so, we introduce the Layered Green and Ampt with Redistribution and Talbot - Ogden model (LGARTO). LGARTO results compare well against simulations using the Richards equation, while offering substantial advantages in speed, stability, and mass conservation. This model will improve our ability to simulate infiltration, runoff, and recharge in layered soils in transitional sub-humid catchments and in riparian portions of arid- and semi-arid catchments, especially in the context of hydrologic models.