Process representation of conifer tree-ring growth is improved by incorporation of climate memory effects

Abstract

Process modeling of tree growth responses to climate can improve our understanding of the functioning of tree growth under climate change, but these models do not consider so-called memory effects. Further, memory effects are unstudied in temperate forests of China, which is showing high vulnerability to climate warming. Here, we integrate a process-based Vaganov–Shashkin–Lite (VS-Lite) model of tree ring formation with the stochastic antecedent modeling (SAM) approach, to quantify the current and antecedent climatic drivers of the tree-ring width chronologies of 109 conifers in temperate China. Based on climate-response patterns, tree-ring width series were clustered into “warm-dry,” “cold-dry,” and “cold-wet.” The VS-Lite results indicated growth in the “warm-dry” and “cold-dry” sites were primarily constrained by current soil moisture, while growth in the “cold-wet” sites was mainly constrained by cotemporaneous temperature. Compared to VS-Lite alone, we found significant improvements in model fit for the SAM framework in all forest types. Notably, antecedent drivers explained 14%, 11%, and 21% of the growth variations in “warm-dry,” “cold-dry,” and “cold-wet” sites, respectively, challenging the assertion that long memory is only important in dry forests. Commonality analysis revealed that growth in “cold-dry” and “cold-wet” sites was determined by antecedent climate while growth in the “cold-dry” and “warm-dry” clusters were determined by the endogenous (autoregressive) memory. In “warm-dry” and “cold-wet” sites, growth was significantly affected by antecedent temperature in opposite directions, whereas in the “cold-dry” sites it was affected by antecedent soil moisture conditions and soil moisture-temperature interactions. Our combined process-empirical model demonstrates that memory is prevalent across temperate forests in China and also improves representation of tree growth in diverse forests under novel climate conditions of the future.