The dataset presented here was collected by the GenTree project (EU-Horizon 2020), which aims to improve the use of forest genetic resources across Europe by better understanding how trees adapt to their local environment. This dataset of individual tree-core characteristics including ring-width series and whole-core wood density was collected for seven ecologically and economically important European tree species: silver birch (Betula pendula), European beech (Fagus sylvatica), Norway spruce (Picea abies), European black poplar (Populus nigra), maritime pine (Pinus pinaster), Scots pine (Pinus sylvestris), and sessile oak (Quercus petraea). Tree-ring width measurements were obtained from 3600 trees in 142 populations and whole-core wood density was measured for 3098 trees in 125 populations. This dataset covers most of the geographical and climatic range occupied by the selected species. The potential use of it will be highly valuable for assessing ecological and evolutionary responses to environmental conditions as well as for model development and parameterization, to predict adaptability under climate change scenarios.

Genetic association studies in forest trees would greatly benefit from information on the response of trees to environmental stressors over time, which can be provided by dendroecological analysis. Here, we jointly analysed dendroecological and genetic data of surviving silver fir trees to explore the genetic basis of their response to the iconic stress episode of the 1970s and 1980s that led to large-scale forest dieback in Central Europe and has been attributed to air pollution. Specifically, we derived dendrophenotypic measures from 190 trees in the Bavarian Forest that characterize the resistance, resilience and recovery during this growth depression, and in the drought year in 1976. By focusing on relative growth changes of trees and by standardizing the dendrophenotypes within stands, we accounted for variation introduced by micro- and macroscale environmental differences. We associated the dendrophenotypes with single nucleotide polymorphisms (SNPs) in candidate genes using general linear models (GLMs) and the machine learning algorithm random forest with subsequent feature selection. Most trees at our study sites experienced a severe growth decline from 1974 until the mid-1980s with minimum values during the drought year. Fifteen genes were associated with the dendrophenotypes, including genes linked to photosynthesis and drought stress. With our study, we show that dendrophenotypes can be a powerful resource for genetic association studies that permit to account for micro- and macroenvironmental variation when data are derived from natural populations. We call for a wider collaboration of dendroecologists and forest geneticists to integrate individual tree-level dendrophenotypes in genetic association studies.

No abstract