Dry-season climate drives interannual variability in tropical tree growth

<p>Tropical and subtropical ecosystems are primarily responsible for the large inter-annual variability (IAV) in the global carbon land sink. The response of tropical vegetation productivity to climatic variation likely drives this IAV, but the climate sensitivity of key productivity components are poorly understood. Tree-ring analysis can help fill this knowledge gap by estimating IAV in woody biomass growth, the major carbon accumulation process in tropical vegetation.</p><p> </p><p>Here, we evaluate the climate responses of woody biomass growth throughout the global tropics. Using an unprecedented compilation of tropical tree-ring data, we test hypotheses that (1) precipitation (P) and maximum temperature (T<sub>max</sub>) have opposite and additive effects on annual tree growth, (2) these climate responses amplify with increasing aridity and (3) wet-season climate is a more important driver of growth than dry-season climate.</p><p> </p><p>We established a network of 347 tree-ring width chronologies compiled from (sub-)tropical latitudes, representing 99 tree species on five continents and obtained from contributors (n=112) and the International Tree-Ring Data Bank (ITRDB; n=235). Our network is climatologically representative for 66% of the pantropical land area with woody vegetation.</p><p> </p><p>To test hypotheses we re-developed standardized ring-width index (RWI) chronologies and assessed climate responses using SOM cluster analysis (monthly P and T<sub>max</sub>) and multiple regression analysis (seasonal P and T<sub>max</sub>). Our results were consistent with hypothesis 1: effects of monthly or seasonal P and T<sub>max</sub> on tree growth were indeed additive and opposite, suggesting water availability to be the primary driver of tropical tree growth. In accordance with hypothesis 2, these climate responses were stronger at sites with lower mean annual precipitation or a larger annual water deficit. However, our results contrast those expected under hypothesis 3. Three of the four clusters show a dominant role of dry-season climate on annual tree growth and regression analyses confirmed this strong dry-season role.</p><p> </p><p>The strong dry-season effect on tropical tree growth seemingly contrasts the general notion that tropical vegetation productivity peaks during the wet season but is consistent with studies showing that climatologically benign dry seasons increase reserve storage and xylem growth. We posit that dry-season climate constrains the magnitude of woody biomass growth that takes place during the following wet season, and thus contributes to IAV in tree growth.</p><p> </p><p>By providing field-based insights on climate sensitivity of tropical vegetation productivity, our study contributes to the major task in Earth system science of quantifying, understanding, and predicting the IAV of the carbon land sink.</p>

Sensitivity of tropical tree growth to climatic variation: a global meta-analysis of tree-ring data

<p>Tropical forests are a crucial component of the global carbon cycle and importantly contribute to the global carbon land sink. Stem growth of tropical trees is a key component of carbon dynamics in tropical forests, but our understanding of how this is driven by climatic variation is poor. Such understanding is needed for predictive vegetation modelling of climate change effects.</p><p>Here, we help to fill this knowledge gap by conducting a meta-analysis of published tropical tree-ring width chronologies. We compiled >350 tropical chronologies (30°N - 30°S) from all tropical climate zones. We used this data set to explore i) common patterns in the tree-growth responses to monthly rainfall and temperature (Tmax) patterns (cluster analysis), ii) the relative importance of temperature and rainfall in determining tropical tree growth (glm), iii) how these climatic drivers shift along gradients of temperature and precipitation.</p><p>Our cluster analysis revealed 6-8 primary types of responses to monthly climate variables. These clusters are associated with mean climate, elevation, or geographic location. The seasonality of growth responses to temperature and rainfall differed clearly among clusters, but the signs of responses were consistent: higher Tmax reduces growth, more precipitation increases growth. Multiple regression analyses of growth responses to seasonal climate further confirmed the negative effects of temperature and positive effects of rainfall. Rainfall during the dry season had the strongest relative importance. Finally, we found that seasonal drivers of tropical tree growth are modified by mean climate. In drier regions, growth sensitivity to temperature increases; in warmer regions, growth sensitivity to rainfall increases. The latter may imply that global warming leads to stronger drought effects on tree growth and possibly enhances mortality risks of tropical trees.</p><p>Our meta-analysis shows that tree-ring studies help to improve understanding of climate-driven carbon dynamics in tropical forests. Insights from this study can be used to benchmark global vegetation modelling and to better understand responses of tropical tree species to climate change.</p>