Government and international agencies have highlighted the need to focus global change research efforts on tropical ecosystems. However, no recent comprehensive review exists synthesizing humid tropical forest responses across global change factors, including warming, decreased precipitation, carbon dioxide fertilization, nitrogen deposition, and land use/land cover changes. This paper assesses research across spatial and temporal scales for the tropics, including modeling, field, and controlled laboratory studies. The review aims to (1) provide a broad understanding of how a suite of global change factors are altering humid tropical forest ecosystem properties and biogeochemical processes; (2) assess spatial variability in responses to global change factors among humid tropical regions; (3) synthesize results from across humid tropical regions to identify emergent trends in ecosystem responses; (4) identify research and management priorities for the humid tropics in the context of global change. Ecosystem responses covered here include plant growth, carbon storage, nutrient cycling, biodiversity, and disturbance regime shifts. The review demonstrates overall negative effects of global change on all ecosystem properties, with the greatest uncertainty and variability in nutrient cycling responses. Generally, all global change factors reviewed, except for carbon dioxide fertilization, demonstrate great potential to trigger positive feedbacks to global warming via greenhouse gas emissions and biogeophysical changes that cause regional warming. This assessment demonstrates that effects of decreased rainfall and deforestation on tropical forests are relatively well understood, whereas the potential effects of warming, carbon dioxide fertilization, nitrogen deposition, and plant species invasions require more cross-site, mechanistic research to predict tropical forest responses at regional and global scales.

Climate change and rising atmospheric carbon dioxide (CO2) concentrations are likely to alter tropical forest net primary productivity (NPP), potentially affecting soil C storage. We examined biochemical and physical changes in soil C fractions in a humid tropical forest where litter manipulation changed total soil C stocks. We hypothesized that: (1.) low-density soil organic C (SOC) fractions are more responsive to altered litter inputs than mineral-associated SOC, because they cycle relatively rapidly. (2.) Any accumulation of mineral-associated SOC with litter addition is relatively stable (i.e. low leaching potential). (3.) Certain biomolecules, such as waxes (alkyl) and proteins (N-alkyl), form more stable mineral-associations than other biomolecules in strongly weathered soils. A decade of litter addition and removal affected bulk soil C content in the upper 5 cm by +32% and -31% in total quantity, respectively. Most notably, C concentration in the mineral-associated SOC fraction was greater in litter addition plots relative to controls by 18% and 28% in the dry and wet seasons, respectively, accounting for the majority of increased bulk soil C stock. Radiocarbon and leaching analyses demonstrated that the elevated mineral-associated SOC consisted of new and relatively stable C, with only 3% of mineral-associated SOC leachable in salt solution. 13 C NMR spectroscopy indicated decadalscale stability of waxes (alkyl C) and microbial biomass compounds (O-alkyl and N-alkyl C) in mineral-associated SOC with litter removal, and losses plant-derived compounds (aromatic and phenolic C). We conclude that shifts in tropical forest NPP, as will likely occur with global change, are likely to alter the quantity, biochemistry, and stability of C stored in strongly weathered tropical soils.