Response of an old‐growth tropical rainforest to transient high temperature and drought

Abstract

Tropical rainforests have experienced episodes of severe heat and drought in recent decades, and climate models project a warmer and potentially drier tropical climate over this century. However, likely responses of tropical rainforests are poorly understood due to a lack of frequent long-term measurements of forest structure and dynamics. We analyzed a 12-year record (1999-2010) of 47 817 annual measurements of canopy height to characterize the response of an old-growth Neotropical rainforest to the severe heat and drought associated with the 1997-1998 El Niño. Well-drained soils on slopes and plateaus experienced a threefold increase in the fraction of the landscape in gaps (≤2 m) and a reduction in the fraction in high canopy (>15 m) causing distributions of canopy height to depart from equilibrium for a period of 2-3 years. In contrast, forests on low-lying alluvial terraces remained in equilibrium and were nearly half as likely to experience upper canopy (>15 m) disturbance over the 12 years of observation. Variation in forest response across topographic positions suggests that tropical rainforests are more sensitive to moisture deficits than high temperature and that topography likely structures landscape-level variation in the severity of drought impacts.

Ecosystem carbon storage does not vary with mean annual temperature in Hawaiian tropical montane wet forests

“…Although MAT most likely has a strong influence on maximum attainable forest biomass in the absence of water limitation and disturbance (Larjavaara & Muller‐Landau, ), realized biomass most often falls below this ‘carrying capacity’ (Stegen et al ., ). We suggest that gap‐phase dynamics, weather‐related events, or other localized disturbance events may be more important drivers of realized forest biomass (Kellner & Asner, ; Stegen et al ., ; Silva et al ., ). Although forest productivity often varies positively with MAT, net primary productivity is a poor predictor of aboveground biomass, especially in tropical forest ecosystems (Keeling & Phillips, ).…”

Chronosequence predictions are robust in a Neotropical secondary forest, but plots miss the mark

“…Secondary forests are particularly prone to such variation because patches of differently aged forests are non‐randomly situated with respect to these variables. In the gradient examined here, previous work in old‐growth forests has shown that soil nutrient status varies with topography (Clark, Clark, Brown, Oberbauer, & Veldkamp, ; Porder, Clark, & Vitousek, ) and that lowland and upland soils support forests that respond differently to drought stress (Silva et al., ). However, despite these caveats, the space‐for‐time substitution at the landscape‐scale resulted in predictions that were close to the observations after 11.5 years.…”

“…Non‐random variation in secondary succession could undermine the space‐for‐time substitution due to violations of the spatial component, or because conditions before and after initial measurement are not the same. The 1998 ENSO resulted in elevated frequencies of canopy damage in old‐growth forest at this site that are anomalous during the subsequent 12 years (Silva, Kellner, Clark, & Clark, ). Our analysis of the temperature, precipitation, and solar radiation records during the 11.5 years before and after the first measurement indicates that conditions were not significantly different, and that the temporal component of the space‐for‐time substitution is unlikely to have been violated by changes in these variables.…”

Fragmentation, forest structure, and topography modulate impacts of drought in a tropical forest landscape

“…, Potts , Silva et al. ). Moisture varies across slope position (ridges vs. slopes vs. valleys) because of differences in drainage and runoff (Burt and Butcher , Western et al.…”

“…Topography can generate significant heterogeneity in forest drought response (Daws et al 2002, Potts 2003, Silva et al 2013. Moisture varies across slope position (ridges vs. slopes vs. valleys) because of differences in drainage and runoff (Burt and Butcher 1985, Western et al 1998, Daws et al 2002 and with slope aspect because of solar radiation (Stephenson 1990).…”