Continental-scale Projection of Future Douglas-fir Growth from Tree Rings: Testing the Limits of Space-for-Time Substitution

A central challenge in global change research is the projection of the future behavior of a system based upon past observations. Over the last decade, tree rings have been used to project tree growth under future climate conditions. But how can the response of tree growth to past climate variation predict the future, when the future does not look like the past? Space-for-time substitution is one way to overcome the problem of extrapolation: the response at a given location in a warmer future is assumed to follow the response at a warmer location today. We compiled a very large tree-ring dataset for Douglas-fir (Pseudotsuga menziesii Franco; >30,000 time series), a species that occupies an exceptionally large geographic and environmental space in North America, then fit to these data a generalized linear mixed effects model capturing variation in absolute ring widths and climate sensitivity. We expected and found opposing gradients of productivity and climate sensitivity, with largest growth rings and weakest response to interannual climate variation in the mesic coastal part of its range vs. narrower rings and stronger climate sensitivity across the semi-arid interior. We also found responses to temperature variation that were opposite in sign across space vs. time. The response to purely spatial variation in mean annual temperature was positive, whereas the response to purely temporal variation in 2- or 3-month mean maximum temperatures was negative. That is, Douglas-fir achieves higher growth rates at warmer locations, but is negatively sensitive to interannual temperature variation in the vast majority of its range and in most seasons. This latter result suggests that spatial and temporal variation are not substitutable. Indeed, projection of future growth based on all model terms leads to a more optimistic picture of future Douglas-fir growth compared to projection that presumes climate sensitivity but not productivity is substitutable. The latter implementation of space-for-time substitution projects widespread decline in the near-term future (2011-2040), with largest relative decreases in the semiarid U.S. Interior West and smallest in the coastal U. S. Pacific Northwest. Uncertainty about growth change is greatest in the Pacific Northwest, both because of a paucity of samples and weak climate sensitivity. We discuss the generality of the positive response to spatial variation vs. negative response to temporal variation in temperature, and the future of space-for-time substitution in ecological forecasting.