Hotter drought and trade-off between fast and slow growth strategies as major drivers of tree-ring growth variability of global conifers

We found: a) growth variability was mainly affected by warm-induced drought and increased at lower latitudes. Climate warming in winter could decrease growth variability, but this effect is by far not enough to offset the threat of hotter drought; b) there existed a trade-off between fast- and slow-growing (drought tolerance) strategies for global conifer species, and abiotic and stand factors affected growth variability via functional traits. Contrary to common conjecture, species with higher drought tolerance revealed higher growth variability due to their occupation of more xeric sites, and may also because higher investment in drought tolerance leads to less investment remaining for growth; c) older trees revealed higher growth variability due to their more conservative growth strategy, while at large scales taller trees showed lower growth variability due to occupying more productive sites; and d) moderate N deposition could reduce growth variability by leading conifers to adopt a more fast-growing strategy (e.g., in Asia), but long-term and excessive N deposition led to increased growth variability (e.g., in North America and Europe).

Our results suggest that coniferous forests in water-limited regions should be more vulnerable to hotter drought, and the ‘fast-slow’ growth strategies may be key in regulating the effects of various abiotic and stand factors on ecosystem stability. Moreover, future hotter drought and N deposition will severely threaten conifer growth, especially for old trees and conifers at lower latitudes.

Methods
Raw tree-ring widths data were extracted from the International Tree-Ring Data Bank (ITRDB) V.7.23 (https://www1.ncdc.noaa.gov/pub/data/paleo/treering/) in October 2021. 

The monthly temperature, precipitation, and potential evapotranspiration during 1970-2010 were extracted for each tree-ring site based on its latitude and longitude, from the Climatic Research Unit (CRU) database (https://crudata.uea.ac.uk/cru/data/hrg/).

N deposition data were obtained from the N Deposition Database (Ackerman et al. 2019; https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2018GB005990), which used the GEOS-Chem Chemical Transport Model to estimate total (wet and dry) inorganic N deposition globally at a spatial resolution of 2°×2.5°.

For stand factors, the stand age of each chronology was calculated as the mean age for all tree rings used to build the chronology.

This global canopy height map was estimated with new satellite LiDAR (light detecting and ranging) data, with a high accuracy and a very high spatial resolution of 10×10 m (Lang et al. 2022; arXiv:2204.08322).