Data from: Differences in carbon stocks along an elevational gradient in tropical mountain forests of Colombia

Tropical mountain forests provide an exceptional opportunity to evaluate the patterns of variation of carbon stocks along elevational gradients that correspond to well-defined temperature gradients. We predicted that carbon stored in live aboveground biomass, aboveground necromass, and soil components of forests on the eastern flank of the Colombian Andes would change with elevation along this gradient extending from 750 to 2800 m above sea level. The rationale was that the corresponding change in temperature (14°C to 26°C) would influence tree growth and decomposition of organic matter. To address this hypothesis, we examined the carbon stored in these three components using data from 20 0.25-ha plots located along this elevational gradient. The mean total carbon stock found in the study region was 241.3±37.5 Mg C/ha. Aboveground carbon stocks decreased with elevation (p =0.001), as did necromass carbon stocks (p =0.016). Although soil organic carbon stocks did not differ significantly along the gradient (p =0.153), they contributed proportionately more at higher than at lower elevations, counterbalancing the opposite trends in aboveground carbon and necromass carbon stocks. As such, total carbon stocks did not vary significantly along the elevational gradient (p =0.576).Carbon Components in Mountain Forests in Colombia20 permanent plots of 0.25-ha (50 m x 50 m) were established along the elevational gradient following Feeley et al. (2011) (Table S1; Fig. S1). The plots were distributed across ten elevational sections, each with a range of 250 m asl. The average horizontal distance between plots was 6.48 km (median: 5.72 km; range: 0.29-17.4 km). Within each plot, we measured, mapped and tagged all trees and palms ≥10 cm diameter at breast height (DBH) following standardized methods for long-term tropical forest dynamics plots (Condit 1998). In the case of tree species with buttresses, diameter was measured 50 cm above them. In all cases, the DBH point was painted and measured with diameter tape calibrated to one millimeter. We excluded arborescent ferns and lianas. Total height (H, in m) was measured on 40 percent of the individuals within each plot, using either a telescopic Hastings fiberglass rod (H