Uncertainties in measuring precipitation hinders precise evaluation of loss of diversity in biomes and ecotones

Abstract

Assessing changes in the distribution of biological communities that share a climate (biomes) is essential for estimating their vulnerability to climate change. We use CMIP6 climate models to calculate biome changes as featuring in classifications such as Holdridge’s Life Zones (climate envelopes). We found that transitional zones between biomes (known as ecotones) are expected to decline under all climate change scenarios, but also that model consensus remains low. Accurate assessments of diversity loss are limited to certain areas of the globe, while model consensus is still poor for half of the planet. We identify where there are robust estimates of changes in biomes and ecotones, and where consensus is lacking. We argue that caution should be exercised in measuring biodiversity loss in the latter, but that greater confidence can be placed in the former. We find that shortcomings in the life zone classification are related to inter-model variability, which ultimately depends on a larger problem, namely the accurate estimation of precipitation compared to CRU. Application of the methodology to other climate classifications confirms the findings.