Teleconnections and edaphoclimatic effects on tree growth of Cedrela odorata L. in a seasonally dry tropical forest in Brazil

The Seasonally Dry Tropical Forests (SDTF) present very high biodiversity and a number of tree species that are adapted to prolonged periods of water stress. Considering tree ring formation is mainly driven by seasonal variation in precipitation in tropical environments, tree-ring studies from STDF can provide important contributions to understanding how these forests are responding to climate variations. In the present study, we demonstrate the influence of edaphoclimatic variables (precipitation, air temperature and soil water deficit-SWD) and the ocean teleconnections (Tropical Southern Atlantic-TSA, Atlantic Multidecadal Oscillation-AMO, Western Hemisphere Warm Pool-WHWP and El Niño 3.4) on Cedrela odorata L. growth from a SDTF of northeastern Brazil. We used standard dendrochronological methods to develop an 89-year-long ring-width index chronology. The climate sensitivity of C. odorata was assessed through Pearson's correlation tests and linear regressions, which allowed to identify the determinant months (cause-effect) of each variable on the chronology. Tree growth was positively correlated with precipitation and negatively correlated with air temperature and SWD, particularly during the rainy season (March to August). In parallel, we identified that extremely dry years can contribute to missing rings, exposing the lack of growth in C. odorata caused by water stress. Among the oceanic variables, all of them showed a negative effect on radial growth of C. odorata, except for TSA, which had no significant effect. Tree growth is constrained in years with strong El Niño and high values of AMO index during the rainy months (May, June and October). However, the WHWP showed a more pronounced negative effect in the beginning of the dry season (September). Our findings add valuable information on C. odorata responses to hydrological seasonality from SDTF and the fluctuations in oceanic teleconnections, which in turn, influence the rainfall dynamics in northeastern Brazil.