Laboratory observations for examining estimates of soil dry surface layer thickness with parsimonious models

Soil dry surface layer (DSL) thickness is often considered a key parameter for land surface resistance to gas exchange. Commonly-used, simple models for DSL thickness are typically empirical in nature and based on limited observational evidence. Laboratory experiments were performed to test soil condition and boundary effects on DSL formation. DSL thickness was analyzed in soil columns with varying texture, initial water content, and potential evaporation rate. DSLs formed to greater depth in fine-textured compared to coarse-textured soils, when beginning from similar initial water content. Based on experiments, we compared a simple, but physically-based mass balance DSL model to an empirical DSL model from the literature. The mass balance model performed better than the empirical relative-wetness based model, and is similar in structure to the current DSL parameterization in the Community Land Model. Results suggest soil resistance parameterizations can be improved by employing simple, but texture-dependent, physically-based DSL formulations.