Global Biogeochemical Cycles

Effects of global change on carbon storage in tropical forests of South America


  • Robert B. McKane,

  • Edward B. Rastetter,

  • Jerry M. Melillo,

  • Gaius R. Shaver,

  • Charles S. Hopkinson,

  • David N. Fernandes,

  • David L. Skole,

  • Walter H. Chomentowski


We used a process-based model of ecosystem biogeochemistry (MBL-GEM) to evaluate the effects of global change on carbon (C) storage in mature tropical forest ecosystems in the Amazon Basin of Brazil. We first derived a single parameterization of the model that was consistent with all the C stock and turnover data from three intensively studied sites within the Amazon Basin that differed in temperature, rainfall, and cloudiness. The range in temperature, soil moisture, and photosynthetically active radiation (PAR) among these sites is about as large as the anticipated changes in these variables in the tropics under CO2-induced climate change. We then tested the parameterized model by predicting C stocks along a 2400-km transect in the Amazon Basin. Comparison of predicted and measured vegetation and soil C stocks along this transect suggests that the model provides a reasonable approximation of how climatic and hydrologic factors regulate present-day C stocks within the Amazon Basin. Finally, we used the model to predict and analyze changes in ecosystem C stocks under projected changes in atmospheric CO2 and climate. The central hypothesis of this exercise is that changes in ecosystem C storage in response to climate and CO2 will interact strongly with changes in other element cycles, particularly the nitrogen (N) and phosphorus (P) cycles. We conclude that C storage will increase in Amazonian forests as a result of (1) redistribution of nutrients from soil (with low C:nutrient ratios) to vegetation (with high C:nutrient ratios), (2) increases in the C:nutrient ratio of vegetation and soil, and (3) increased sequestration of external nutrient inputs by the ecosystem. Our analyses suggest that C:nutrient interactions will constrain increases in C storage to a maximum of 63 Mg/ha during the next 200 years, or about 16% above present-day stocks. However, it is impossible to predict how much smaller the actual increase in C storage will be until more is known about the controls on soil P availability. On the basis of these analyses, we identify several topics for further research in the moist tropics that must be addressed to resolve these uncertainties.