Contemporary tree growth shows altered climate memory
Drew Peltier, William R. L. Anderegg, Jessica Guo, Kiona Ogle
Abstract
Abstract Trees are long‐lived organisms, exhibiting temporally complex growth arising from strong climatic “memory.” But conditions are becoming increasingly arid in the western USA. Using a century‐long tree‐ring network, we find altered climate memory across the entire range of a widespread western US conifer: growth is supported by precipitation falling further into the past (+15 months), while increasingly impacted by more recent temperature conditions (−8 months). Tree‐ring datasets can be biased, so we confirm altered climate memory in a second, ecologically‐sampled tree‐ring network. Predicted drought responses show trees may have also become more sensitive to repeat drought. Finally, plots near sites with relatively longer precipitation memory and shorter temperature memory had significantly lower recent mortality rates ( R 2 = 0.61). We argue that increased drought frequency has altered climate memory, demonstrate how non‐stationarity may arise from failure to account for memory, and suggest memory length may be predictive of future tree mortality.