Ecohydrological linkages in forest ecosystems: the case of the Re della Pietra catchment

The exchange of fluxes between surface and subsurface water pools and vegetation is highly complex due to the dynamic - in space and time- interactions among several biotic (physiological) and abiotic (hydrometeorological, geological, geomorphological, pedological) factors. Efforts that go beyond the analysis of individual processes and aim at capturing how different processes and factors are strictly connected are essential to achieve a wider understanding of how forest ecosystems work and respond to climate stress. In this work, I capitalize and build on an increasing knowledge deriving from field observations in the mountain forested Re della Pietra experimental catchment (2 km²) in Italy, to investigate the main ecohydrological linkages governing the functioning of this ecosystem, specifically focusing on the role of hillslope topography.

Field measurements and statistical modelling analyses carried out through wavelet and machine learning applications revealed that the high slope of a 120m-long monitored hillslope in the headwater of the catchment controlled the spatial distribution of water in the vadose zone and affected the occurrence of subsurface preferential flow.

Higher soil water contents in the lower part of the hillslope promoted a faster and more efficient growth of trees that had larger diameters compared to trees in the upper part of the hillslope, although being of the same age, clearly reflecting local differences in water availability that impacted on growth rates. This behaviour was confirmed by sapflow measurements and isotope data that showed more reduced sapflow velocity of trees in the upper part of the hillslope during dry conditions compared to trees at the hillslope bottom, despite soil water in the first 40-cm was the main source for all trees. In turn, these differences in tree size and canopy expansion along the hillslope affected canopy interception, with larger and temporally stable patterns of throughfall in the upper hillslope, characterized by less dense canopies, than the hillslope bottom. Moreover, the relation between sap flow velocity and vapour pressure deficit varied along the hillslope as well, with larger hysteresis loops as a function of increased solar radiation, temperature, and soil moisture in the upper and middle part of the hillslope but erratic and more complex patterns at the hillslope bottom.

In the headwater, preferential flow occurred preferably in the middle hillslope position and more frequently during wet antecedent conditions, revealing a feedback relation between preferential flow and soil moisture. The initiation of preferential flow contributed to developing subsurface hillslope-stream connectivity that promoted sustained streamflow during large events. However, in the lower part of the catchment, where the hillslope slope is gentler, preferential flow was mainly controlled by soil properties (particularly, bulk density) and occurred more frequently than in the headwaters indicating that other factors interact with topography.

These results contribute to a more thorough understanding of ecohydrological linkages in mountain forested catchments and pave the way for further analyses aimed at disentangling the combined role of different but complementary factors driving the ecological and hydrological response of forest ecosystems.