Extended Data Fig. 4: Suitability predicted using macro- and microclimate (forest grid cells not masked).
From: Macroclimate data overestimate range shifts of plants in response to climate change
In (a) and (b) the relationship between the current (2003–2021) and historic (1977–1995) predicted probability of occurrence on the Lizard Peninsula is shown for all taxa, demonstrating that when microclimate (green) data are used, the relationship is shallower than when macroclimate (purple) data are used. In consequence, when modelled using microclimate data, fewer extirpations are predicted. The green and purple dashed lines are the line-of-best fit for the modelled relationship for microclimate and macroclimate respectively. In (a) macroclimate occurrence is derived from 0.5° grid resolution models and in (b) from 5 km grid resolution models. In (c) the relationship between the current (2003–2021) and historic (1977–1995) predicted probability of occurrence of Erica tetralix is shown in every grid cell with a predicted probability of occurrence >0.1 across Europe (purple) and for each 100 m grid cell with a historic record in each of 73 40 × 40 km selected focal areas (green).
