Rock outcrops reduce temperature-induced stress for tropical conifer by decoupling regional climate in the semiarid environment

Abstract

We aimed to understand the effect of rock outcrops on the growth of Podocarpus lambertii within a microrefuge. Our hypothesis holds that the growth and survival of this species depend on the regional climate decoupling provided by rock outcrops. To test this hypothesis, we characterized the microclimate of (1) surrounding vegetation, (2) rock outcrop corridors, and (3) adjacencies. We assessed population structure by collecting data of specimen stem diameter and height. We also assessed differences between vegetation associated or not with outcrops using satellite imaging. For dendrochronological analyses, we sampled 42 individuals. Tree rings of 31 individuals were dated, and climate-growth relationships were tested. Rock outcrops produce a favorable microclimate by reducing average temperature by 4.9 °C and increasing average air humidity by 12 %. They also reduce the variability of atmospheric temperature by 42 % and air humidity by 20 % supporting a vegetation with higher leaf area index. Within this vegetation, specimen height was strongly constrained by the outcrop height. Although temperature and precipitation modulate this species growth, temperature-induced stress is the key limiting growth factor for this population of P. lambertii. We conclude that this species growth and survival depend on the presence of rock outcrops. These topography elements decouple regional climate in a favorable way for this species growth. However, these benefits are restricted to the areas sheltered by rock outcrops. Although this microrefuge supported P. lambertii growth so far, it is unclear whether this protection would be sufficient to withstand the stress of future climate changes.

Climatic microrefugia under anthropogenic climate change: implications for species redistribution

“…A slope value close to one corresponds to a high coupling whereas a slope value close to zero corresponds to a high decoupling. For instance, Locosselli et al () found a coupling of 0.6 between in‐situ (within concave topographic features) and free‐air maximum temperatures during summer 2013 in Brazil. Similarly, in a tropical forest in Brazil, Ewers and Banks‐Leite () used one year of data and found a coupling of 0.4, on average, between sub‐canopy and external air temperatures.…”

“…ravines) and habitat (e.g. trees) features (Xu et al 1997, Chen et al 1999, Grimmond et al 2000, Renaud and Rebetez 2009, Locosselli et al 2016. Although simplistic, this virtual species based on a uni-dimensional climatic niche, i.e.…”

“…sub-canopy) and macrohabitat (e.g. free-air) temperatures, being the response and explanatory variables, respectively, in an ordinary least-square (OLS) regression (Ewers and Banks-Leite 2013, Varner and Dearing 2014, Locosselli et al 2016. A slope value close to one corresponds to a high coupling whereas a slope value close to zero corresponds to a high decoupling.…”

The relationship between climate change and the endangered rainforest shrub Triunia robusta (Proteaceae) endemic to southeast Queensland, Australia

“…In SDM3, geological and geomorphological variables were coupled with climatic variables, and the model predicted much slower climate impacts. This demonstrates that local conditions could potentially buffer (decouple) regional climate46, although coupling fine-resolution substrates (25 × 25 m grid) and low-resolution climate (1 × 1 km grid) might result in models weighted toward substrate variables. Moreover, soil and vegetation dataset in future years that are consistent with IPCC scenarios were not available at the time of the study.…”

Towards establishing a new environmental archive – Annual growth periodicity, stable carbon isotope variability and reconstruction potential of 'Akoko (Euphorbia olowaluana), a native Hawaiian tree with C4 photosynthetic pathway

“…Many other studies have validated these findings and have raised important implications for the future of management decisions involving tropical forests (e.g. worBes 1989;1995;2002;sTahle 1999;dünisch and monTóia 2003;FichTler and clark 2003;Brienen and Zuidema 2005;Brienen and Zuidema 2006;Therrell et al 2006;Baker and BunyaveJchewin 2006;BräuninG et al 2009;lewis et al 2009;roZendaal et al 2009;Brienen et al 2010;héraulT et al 2010;roZendaal et al 2010;TroueT et al 2010;roZendaal and Zuidema 2011;wils et al 2011;donG et al 2012;Zuidema et al 2012;locosselli et al 2013;alTman et al 2013;de ridder et al 2013;Zuidema et al 2013;GroenendiJk et al 2014;locosselli et al 2016;colomBaroli et al 2016;volland et al 2016).…”