Drivers of root water uptake patterns in a beech-dominated mixed forest

Throughfall constitutes the majority of water  entering most forest ecosystems' root zones. Previous studies showed that throughfall patterns are temporally stable and influence soil moisture response to rainfall. However, their impact on soil water distribution ceases rapidly. The spatial variation in root water uptake was proposed as a reason for this decoupling throughfall and soil water patterns, but,  to the best of our knowdeldge experimental evidence is lacking. Therefore, we investigated root water uptake patterns with comprehensive field observations in an unmanaged forest site in the 2019 (April-August) growing season. The research site (1 ha) is a part of Hainich CZE in Thuringia, Germany. In the site, the tree community consists of 574 individuals of various ages (diameter at breast height ≥ 5cm). The European beech dominated site also hosts other temperate species such as Sycamore maple, European ash, and Norway maple. The field observation setup was composed of closely paired (within 1 m) throughfall and soil water content measurements at 34 locations. While soil water content was recorded every six minutes, throughfall was measured weekly. Moreover, we measured open rainfall in an adjacent open grassland (distance 250 m)  at the same time as  throughfall .

We derived root water uptake at each location from diurnal variations within the soil moisture time series. While daily average transpiration ranged between 0.9 mm d and 3 mm potential evapotranspiration changed between 1.8 mm and 3.1 mm. Further, we applied a linear mixed-effect model to identify controlling factors for horizontal patterns of root water uptake throughout the growing season. We found that temporally stable throughfall patterns do not influence root water uptake patterns. Instead, soil water distribution and vegetation features significantly influence local water uptake. We show that greater local soil water storage promoted root water uptake, slightly modulated by field capacity. Further, seasonally declined soil water storage, on average, likely shifted water extraction depth to deeper layers. A higher number of species is also related to higher root water uptake, which possibly signifies water competition among trees. Our findings suggest that elevated throughfall is neither taken up by roots nor retained in the soil matrix, probably due to local processes such as fast flow. Ultimately, the soil water availability and adaptation of co-existing trees to changes in accessible water storage regulate root water uptake patterns.