Branch initiation and mortality
Over the period studied (1929-2013) branch initiation frequency was negatively correlated with mean summer temperatures in the year prior to branch initiation (r=-0.23, p<0.05; Fig. 3c), i.e. warm summers were linked with a lower amount of new branches formed in the year after. Over the period 1987-2013, branch mortality frequency was positively correlated with mean early winter temperatures following the growing season in the year of death (r=0.59, p<0.002; Fig. 3d) and with mean temperatures in spring prior to the growing season (0.50, p<0.01). Higher temperatures before and after the growing season were thus related with higher branch mortality.