Tropical and temperate forests together encompass an estimated 38% of terrestrial carbon pools and 48% of terrestrial net primary production, and thus knowledge of their carbon budgets and of how these budgets respond to natural and anthropogenic global change is key to understanding the global carbon cycle today and in the future. Regrettably, there are still large gaps in our understanding of forest carbon pools, short- and longterm carbon fluxes, the mechanisms underlying these fluxes, and the likely impacts of global change – especially for tropical forests. The CTFS Global Forest Carbon Research Initiative aims to fill these gaps through research quantifying the sizes of forest carbon pools and fluxes, their spatial and temporal variation, and the drivers of this variation at multiple tropical and temperate forest sites around the globe.  Each site hosts a large (16-52 ha) forest dynamics plot in which all trees greater than 1 cm in diameter are being censused every 5 years.  Building on this work, we are now measuring carbon pools in soil, fine roots, woody debris, and lianas in these plots as well. We are also measuring tree growth, litterfall, and woody debris inputs and outputs annually in order to estimate interannual variation in associated carbon fluxes and link to interannual variation in climate.  Here, we use these data to investigate the correlates of spatial variation in aboveground tree carbon stocks within and among plots, and thereby gain insights into the causes and consequences of this variation.
Results/Conclusions

Spatial variation in aboveground tree carbon stocks largely parallels variation in coarse woody productivity (tree growth) within plots at the 20x20 m scale, but the relationship is much weaker among plots.  Variation in tree carbon stocks and productivity shows some relationship to topography within most plots, but the form of the relationship varies greatly among plots.  In all our plots, spatiotemporal variation in aboveground tree carbon stocks at the 20x20 m scale and above is due almost entirely to variation in tree size, with a small contribution from variation in wood density and essentially no contribution from variation in tree number.  Spatial variation in carbon stocks in woody debris aboveground also correlates with spatial variation in live tree carbon stocks within plots, but not among them.  We discuss the implications of the results for understanding the relative importance of tree growth and factors that limit it in driving standing tree biomass variation at different scales.