| 英文摘要: | Analysis of data from 92 forested sites across the globe indicates that nutrient availability is the dominant driver of carbon retention in forests. Global carbon budgets indicate that approximately 27% of anthropogenic CO2 emissions are stored in terrestrial ecosystems with a similar percentage stored in the oceans1. Of the terrestrial ecosystems, forests are by far the most important carbon sink, due to the long storage time of carbon in stem wood2. Declining global forest sinks could potentially increase the growth rate of atmospheric CO2 concentration by 50%. Unfortunately, model predictions of the fate of carbon in forests over long timescales (the coming decades) are highly uncertain because of the many interacting drivers that affect forest carbon cycling. Apart from forest management, many field studies and model approaches suggest that atmospheric CO2 concentration3 and climate variables, including temperature changes and precipitation4, play a key role in carbon cycling. However, in most cases data were collected in a restricted area or the datasets/models did not include all influencing drivers, such as nitrogen deposition, site nutrient availability and ozone exposure. Consequently, we do not yet know which factor(s) most strongly govern forest carbon storage. This leads to an uncertainty in any future prediction of atmospheric CO2 increase and related temperature rise. In a study published in Nature Climate Change, Marcos Fernández-Martínez and colleagues5 propose that the availability of important plant nutrients (such as nitrogen, phosphorus and potassium) is the chief determinant of the amount of carbon sequestration in forests (trees and soils) on a global scale. The team created a dataset of 92 forest stands — across a range of boreal, temperate, Mediterranean and tropical forests — consisting of observations of gross primary production (GPP), net ecosystem production (NEP) and ecosystem respiration (Re), combined with information on forest management and stand age. In this context, NEP (equal to the difference between GPP and Re) quantifies the forest carbon sink. Estimates of climate variables, including mean annual temperature, precipitation and water deficit, were assigned on the basis of climate interpolations and satellite-based observations. Finally, they assigned a nutrient availability status to each site using, among other features, soil characteristics and information on nitrogen and phosphorus concentrations in soil and foliage. Sites were classified as either nutrient-rich (with no apparent nutrient limitation) or nutrient-poor (with an apparent nutrient limitation). The dataset was then used to determine which variables best explain the variation in NEP and carbon-use efficiency at the ecosystem level (CUEe), defined as the ratio of NEP to GPP. Statistical models were applied to disentangle the effects of the explanatory variables (including GPP, management, stand age, climatic factors and nutrient availability) from their interactions. Model results showed that commonly assumed controls, such as water availability and management, had an insignificant effect on NEP. The most important variable was nutrient availability, which alone explained 19% of the variance in NEP, whereas temperature alone explained only 9%. Most strikingly, the analysis suggests that nutrient-rich forests retain a much larger proportion (33%) of carbon, which is exchanged by photosynthesis GPP, than nutrient-poor forests (6–17%). In line with previous hypotheses6, the authors propose various mechanisms that explain why increasing nutrient availability increases the NEP; the most important are summarized in Fig. 1.
Affiliations
-
Wim de Vries is at the Environmental Systems Analysis Group, Wageningen University, 6700 AA Wageningen, The Netherlands
|
