Comment on hess-2022-418

Abstract

<strong class="journal-contentHeaderColor">Abstract.</strong> Throughfall heterogeneity induced by the redistribution of precipitation in vegetation canopies has repeatedly been hypothesized to affect the variation of soil water content and runoff behavior, especially in forests. However, we are not aware of any observational study relating the spatial variation of soil water content directly to net precipitation to confirm modelling hypotheses. Here, we investigate whether throughfall patterns affect the spatial heterogeneity of soil water response in the main rooting zone. We assessed rainfall, throughfall and soil water contents (two depths: 7.5 cm and 27.5 cm) on a 1‐ha temperate mixed beech forest plot in Germany 2015&ndash;2016 during the growing seasons in independent high‐resolution stratified random designs. Because throughfall and soil water content cannot be measured at the same location, we used kriging to derive the throughfall values at the locations where soil water content was measured. We first explore the spatial variation and temporal stability of throughfall and soil water patterns, and next evaluate the effects of input (throughfall), soil properties (field capacity and air capacity), and vegetation parameters (canopy cover and distance to the next tree) on soil water content and dynamics. Throughfall spatial patterns were related to canopy density. Although spatial auto-correlation decreased with increasing event sizes, temporally stable throughfall patterns emerged, leading to reoccurring high and lower input locations across precipitation events. A linear mixed effect model analysis showed, that soil water content patterns were only poorly linked to throughfall spatial patterns, and it was rather shaped by unidentified but time constant factors. Instead of soil water content itself, the patterns of its increase after rainfall corresponded more closely to throughfall patterns, in that more water was stored in the soil where throughfall was elevated. Furthermore, soil moisture patterns themselves enhanced or decreased water storage in the soil, and probably fast drainage and runoff components. Locations with low topsoil water content tended to store less of the input water, indicating preferential flow. In contrast in subsoil, locations with high water content stored less water. Also, distance to the next tree and air capacity modified how much water was retained in soil storage. In this comprehensive study we show that throughfall patterns imprint less on soil water contents and more on soil water dynamics shortly after rainfall events, therefore only partly confirming previous modelling with data. Our findings highlight at the same time systematic patterns of times and locations where the capacity to store water is reduced and water probably conducted quickly to greater depth. Our results indicate that not soil moisture patterns but rather percolation may depend on small scale spatial heterogeneity of canopy input patterns.