Detection and evaluation of anthropogenic impacts on natural forest ecosystems from long-term tree-ring observations

• Published in: Forest ecology and management Vol. 558; p. 121784
• Main Authors: Bogachev, Mikhail I., Grigoriev, Andrey A., Pyko, Nikita S., Gulin, Alexey N., Grigorieva, Alena V., Chindyaev, Alexander S., Kayumov, Airat R., Tishin, Denis V.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 15.04.2024
• Subjects: Climate variability; Detrended fluctuation analysis; Local trend detection; Long-term correlations; Tree ring width; Detrended fluctuation analysis; Local trend detection; Tree ring width; Long-term correlations; Climate variability
• ISSN: 0378-1127; 1872-7042
• DOI: 10.1016/j.foreco.2024.121784

Anthropogenic interventions lead to various direct and indirect impacts on natural ecosystems that are often hindered by natural long-term variability, and thus their detection and evaluation remain challenging. Ecological systems are strongly affected by climate variations that typically exhibit long-term correlations capable of imitating or hindering external trends in finite-time observations, thus complicating their detection and correct attribution to either anthropogenic interventions or natural variability. Here we focus on the quantitative assessment of the alterations in the tree-ring width (TRW) of four tree species in response to the changes in the soil water regime following a drainage experiment in a dwarf-shrub type peatland forest. We consider the long-term effects of the intervention, focusing on two characteristic quantities: the durations of clusters when significant discrepancies with relevant controls could be observed in every single consecutive year, and relative trends in the data reflecting long-term, gradual changes in the ecosystem. By extrapolating pre-drainage TRW dynamics and adjusting for recent climate variations using a multivariate model, we simulate surrogate data series that act as additional controls for the post-drainage time period. By comparing the long-term dynamics of the observational TRW data series against both natural and surrogate controls over several decades following the drainage experiment, we evaluate long-term alterations and gradual trends in the tree growth dynamics and reassess the statistical significance of these effects, taking into account long-term correlations in the natural TRW variations. Our results also indicate pronounced alterations in the drought stress response characterized by significant negative trends in the tree growth dynamics following the 2010 heatwave and associated flash drought in the drained area, while no similar effect could be observed in the undrained area, indicating that the increased productivity of the forest ecosystem following the drainage likely comes at the cost of its reduced drought stress resilience. •Long-term correlations in tree-ring and climate records can imitate or hinder trends.•Extrapolated pre-experimental tree-ring data series can be used as surrogate control.•Significance of multi-year discrepancies can be re-evaluated from cluster durations.•Drainage increases productivity of forests at the cost of drought stress resilience.