Improved tree-ring sampling strategy enhances the detection of key meteorological drivers of rockfall activity

Abstract

To overcome the lack of historical archives at active rockfall environments, dendrogeomorphic techniques have been used extensively on forested slopes since the early 2000s and several approaches developed to extract rockfall signals from tree-ring records. Given the unpredictable nature of rockfall, these reconstructions are, in principle, of great help when it comes to relate fluctuations in rockfall activity to meteorological variables. Yet, so far, dendrogeomorphic time series were only rarely compared with meteorological records. Here, we ascribe this shortfall to the absence of clear guidelines on how to optimize the sampling strategy. In order to test this hypothesis, we capitalize on the extensive dataset of rockfall impacts recorded in trees growing in a mixed forest plot within the French Alps. We designed six different scenarios retrospectively and compared rockfall reconstructions with meteorological records. Our results demonstrate that reconstructions that include trees located in the most active segments of the plot (i.e. close to the cliff and over periods lacking multidecadal trends in the reconstruction) capture summer precipitation as the main driver of rockfall activity more clearly. This result is in line with monitoring studies from calcareous cliffs in the Alps located outside periglacial environments thus confirming the robustness of our approach.