Know your limits? Climate extremes impact the range of Scots pine in unexpected places.

ABSTRACT

BACKGROUND AND AIMS: Although extreme climatic events such as drought are known to modify forest dynamics by triggering tree dieback, the impact of extreme cold events, especially at the low-latitude margin ('rear edge') of species distributional ranges, has received little attention. The aim of this study was to examine the impact of one such extreme cold event on a population of Scots pine (Pinus sylvestris) along the species' European southern rear-edge range limit and to determine how such events can be incorporated into species distribution models (SDMs). METHODS: A combination of dendrochronology and field observation was used to quantify how an extreme cold event in 2001 in eastern Spain affected growth, needle loss and mortality of Scots pine. Long-term European climatic data sets were used to contextualize the severity of the 2001 event, and an SDM for Scots pine in Europe was used to predict climatic range limits. KEY RESULTS: The 2001 winter reached record minimum temperatures (equivalent to the maximum European-wide diurnal ranges) and, for trees already stressed by a preceding dry summer and autumn, this caused dieback and large-scale mortality. Needle loss and mortality were particularly evident in south-facing sites, where post-event recovery was greatly reduced. The SDM predicted European Scots pine distribution mainly on the basis of responses to maximum and minimum monthly temperatures, but in comparison with this the observed effects of the 2001 cold event at the southerly edge of the range limit were unforeseen. CONCLUSIONS: The results suggest that in order to better forecast how anthropogenic climate change might affect future forest distributions, distribution modelling techniques such as SDMs must incorporate climatic extremes. For Scots pine, this study shows that the effects of cold extremes should be included across the entire distribution margin, including the southern 'rear edge', in order to avoid biased predictions based solely on warmer climatic scenarios.