Genomic prediction Fagus sylvatica

We used 15 F. sylvatica stands in southern Germany selected along an ecological gradient from humid-cold to warm-dry climate. For each stand, we sequenced the whole genome of DNA pools (PoolSeq). Building on an individual genomic prediction model, we developed a genomic prediction framework for such pooled data. We used remote sensing to phenotype stands for either leaf area index (LAI) or moisture stress index (MSI). We then predicted this phenotypic data with meteorological data from the period 2016-2022 and a genomic population prediction score in a single Generalised Linear Model. Our results showed that while meteorological data alone explained ~68% of the phenotypic variance in LAI and 66% in MSI, the addition of genomic prediction significantly increased the explanatory power of the model by 11.4% and 14.8%, respectively. We then used this model to predict the reaction of the stands to future climate change under different evolutionary adaptation scenarios. Under the moderate RCP 4.5 future greenhouse gas concentration scenario, the future persistence of beech forests appeared possible if at least moderate evolutionary adaptation occurs. However, under the RCP 8.5 scenario, beech forests will likely disappear from most of the area investigated by 2050.