Net precipitation assessment in a grassland and soil moisture response at plot scale in a temperate climate

Abstract

<div> <p>Evidence on spatial variation of net precipitation in grasslands is scarce. Challenges arise due to a small-scale canopy structure of grasslands.</p> <p>In this study, we designed and tested a new in-situ measurement device (interception grid) to assess net precipitation in grasslands. The collector allows the natural development of the canopy. We tested the device both in the lab for splash loss and in the field to test its capacity to assess net precipitation. In the field, we installed 25 collectors on a grassland within the Hainich Critical Zone Exploratory (Thuringia, Germany), 23 of which were paired with soil moisture sensors. We conducted weekly measurements gross and net precipitation (above and below the canopy), along with grass height in 2019 (March-August) and 2020 (January -February). We categorized the data into two groups (‘covered,’ ‘uncovered’), accounting for canopy development.</p> <p>In the lab, we found that the drop size strongly affects splash loss. Drops of ca. 2 mm, created more than 16% splash loss, decreasing to less than 3% for drops <1.5 mm. Drop sizes <1.75 mm during the sampling period (2019) suggest low to intermediate splash loss in the field, further decreased in the covered period as the canopy contact slows down the drops. Grid measurements corrected with estimated splash loss during the uncovered period agreed well with gross precipitation. Using linear mixed effect models, we found that wind speed and grass height significantly affected the grid measurements of covered periods. Therefore, grids were able to capture net precipitation variation due to grass development. These steps encouraged us to examine the canopy effect in the soil moisture response to rainfall.</p> <p>Soil moisture response over the entire period was not related to the spatial variation of net precipitation. However, for the drier period (June-August 2019), when the spatial variation in soil moisture is higher, and the overall response to rain events stronger, net precipitation slightly affected soil moisture response. LMEM analysis to estimate factors on soil moisture response showed that grass height, net precipitation are significant predictors. Yet, there is no remarkable difference between using net precipitation and gross precipitation as potential drivers for soil moisture response, indicating that the spatial effects are comparatively small. Overall, our findings suggest that the grids are cable to catch canopy effects on the precipitation, while the effect of wind on under-catch still needs to be investigated further.</p> </div>