Intensive Eucalyptus plantation management in Brazil: Long-term effects on soil carbon dynamics across 300 sites

Intensively managed forest plantations now cover more than 6 million hectares in Brazil, and another 20 million hectares in other tropical regions. Although aboveground biomass, and therefore carbon, is well monitored due to commercial interest, the belowground carbon dynamics and site sustainability remain poorly understood. So, how does intensive silviculture change the storage of carbon in soils? Trends in soil organic carbon from land-use change indicate that conversion from pastures to Eucalyptus plantations should maintain soil carbon stocks. However, comprehensive, long-term studies are needed to understand the variability in these trends to better manage these systems for sustainable productivity across a highly variable landscape, as well as to understand the role that soils may play in sequestering carbon for climate change mitigation. In this unique, long-term soil study, soil samples were collected in the 1980s/90s, 2001, and 2010 across 300 intensively managed Eucalyptus plantation sites located in the states of Bahia, Espirito Santo, and Sao Paulo, Brazil. Natural ecosystems for these states include Savannah-Dry Forest, Atlantic Forest, and Savanna, respectively. The sampling covered at least three complete rotations of Eucalyptus at each site; climate, past land use, productivity, and soil characteristics vary across this geographic gradient. Across the two periods, both Espirito Santo (P<0.001) and Bahia (P=0.05) showed a decrease in soil carbon concentrations, while Sao Paulo saw no change over time. For the 0-30 cm layer, plantations in Espirito Santo state had the largest decrease in soil carbon concentration up to 2001, decreasing soil carbon stocks at an average rate of 1.3 Mg C ha^-1 year^-1. This, however, was followed by no significant change from 2001 to 2010 which may indicate stabilization of soil carbon stocks under the new land use. The Eucalyptus in Bahia created no change in the first sampling period, but saw a decline of 0.35 Mg C ha^-1year^-1 in soil carbon in the second sampling period from 2001-2010. Initial results show that, across the regions, sites that had higher soil carbon stocks tended to lose more soil carbon under intensive silviculture. In all three regions, clay content related strongly to soil carbon concentrations. However, the clay did not have a consistent relationship with the rates of change in soil carbon concentration, varying among regions, with negative, positive, and no relationship for the 2001-2010 time period. Further investigation will determine relationships with temperature and precipitation, past land use history, and Eucalyptus productivity. These results will be essential in evaluating the effects on soil organic carbon dynamics due to conversion to short-rotation Eucalyptus plantations in the tropics across a broad and variable landscape.