Abstract. The net primary productivity (NPP) of tropical forests is one of the most important and least quantified components of the global carbon cycle. Most relevant studies have focused particularly on the quantification of the above-ground coarse wood productivity, and little is known about the carbon fluxes involved in other elements of the NPP, the partitioning of total NPP between its above-and below-ground components and the main environmental drivers of these patterns. In this study we quantify the above-and below-ground NPP of ten Amazonian forests to address two questions: (1) How do Amazonian forests allocate productivity among its above-and below-ground components? (2) How do soil and leaf nutrient status and soil texture affect the productivity of Amazonian forests? Using a standardized methodology to meaCorrespondence to: L. E. O. C. Aragão (leocaragao@gmail.com) sure the major elements of productivity, we show that NPP varies between 9.3±1.3 Mg C ha −1 yr −1 (mean±standard error), at a white sand plot, and 17.0±1.4 Mg C ha −1 yr −1 at a very fertile Terra Preta site, with an overall average of 12.8±0.9 Mg C ha −1 yr −1 . The studied forests allocate on average 64±3% and 36±3% of the total NPP to the aboveand below-ground components, respectively.

Abstract. The net primary productivity (NPP) of tropical forests is one of the most important and least quantified components of the global carbon cycle. Most relevant studies have focused particularly on the quantification of the above-ground coarse wood productivity, and little is known about the carbon fluxes involved in other elements of the NPP, the partitioning of total NPP between its above- and below-ground components and the main environmental drivers of these patterns. In this study we quantify the above- and below-ground NPP of ten Amazonian forests to address two questions: (1) How do Amazonian forests allocate productivity among its above- and below-ground components? (2) How do soil and leaf nutrient status and soil texture affect the productivity of Amazonian forests? Using a standardized methodology to measure the major elements of productivity, we show that NPP varies between 9.3±1.3 Mg C ha−1 yr−1 (mean±standard error), at a white sand plot, and 17.0±1.4 Mg C ha−1 yr−1 at a very fertile Terra Preta site, with an overall average of 12.8±0.9 Mg C ha−1 yr−1. The studied forests allocate on average 64±3% and 36±3% of the total NPP to the above- and below-ground components, respectively. The ratio of above-ground and below-ground NPP is almost invariant with total NPP. Litterfall and fine root production both increase with total NPP, while stem production shows no overall trend. Total NPP tends to increase with soil phosphorus and leaf nitrogen status. However, allocation of NPP to below-ground shows no relationship to soil fertility, but appears to decrease with the increase of soil clay content.

Through their role as 'ecosystem engineers', termites provide a range of ecosystem services including decomposition, and carbon and nitrogen cycling. Although termite diversity levels differ between regions as a result of variation in regional species pool size, in general, termite diversity is thought to decline with elevation. This study (1) investigated how termite species density, abundance, functional group diversity and termite attack on dead wood vary with altitude along an Amazon-Andes altitudinal gradient in Peru; (2) identified likely environmental causes of this pattern; and (3) explored the implications of termite presence for ecosystem functioning (notably for decomposition). Termites were sampled with a standardized 100 Â 2 m straight-belt transect at five undisturbed forest sites along a gradient 190 to 3025 m, as were environmental variables and termite and fungus attack on dead wood. Termite diversity was similar to that found at comparable sites in South America, and there was little turnover of assemblage composition with elevation suggesting that montane specialists are not present. Termite diversity declined with increased elevation, though the upper distribution limit for termites was at a lower elevation than anticipated. We suggest that key drivers of this elevation pattern are reduced temperature with altitude and midelevation peaks in soil water content. Also, attack on dead wood diminished with decreasing termite indirect absolute abundance, while the depth of the soil humic layer increased. We hypothesize that termite abundance is a major accelerant of decomposition rates (and associated mineralization) in Amazonian forests.Abstract in Spanish is available at http://www.blackwell-synergy.com/loi/btp.

Abstract. Controversy exists over the cause and timing of the extinction of the Pleistocene megafauna. In the tropical Andes, deglaciation and associated rapid climate change began ~8,000 years before human arrival, providing an opportunity to separate the effects of climate change from human hunting on megafaunal extinction. We present a paleoecological record spanning the last 25,000 years from Lake Pacucha, Peru (3,100 m elevation). Fossil pollen, charcoal, diatoms, and the dung fungus Sporormiella, chronicle a two-stage megaherbivore population collapse. Sporormiella abundance, the proxy for megafaunal presence, fell sharply at ~21,000 years ago, but rebounded prior to a permanent decline between ~16,800 and 15,800 years ago. This two-stage decline in megaherbivores resulted in a functional extinction by ~15,800 years ago, 3,000 years earlier than known human occupation of the high Andes. Declining megaherbivore populations coincided with warm, wet intervals. Climatic instability and megafaunal population collapse triggered an ecological cascade that resulted in novel floral assemblages, and increases in woody species, fire frequency, and plant species that were sensitive to trampling. Our data revealed that Andean megafaunal populations collapsed due to positive feedbacks between habitat quality and climate change rather than human activity.

Palaeoecological records suggest that humans have been in the Andes since at least 14 000 years ago. Early human impacts on Andean ecosystems included an increase in fire activity and the extinction of the Pleistocene megafauna. These changes in Andean ecosystems coincided with rapid climate change as species were migrating upslope in response to deglacial warming. Microrefugia probably played a vital role in the speed and genetic composition of that migration. The period from ca 14 500 to 12 500 years ago was when novel combinations of plant species appeared to form no-analogue assemblages in the Andes. By 12 000 years ago most areas in what are today the Andean grasslands were being burned and modified by human activity. As the vegetation of these highland settings has been modified by human activity for the entirety of the Holocene, they should be regarded as long-term manufactutred landscapes. The sharp tree lines separating Andean forests from grasslands that we see today were probably also created by repeated burning and owe their position more to human-induced fire than climatic constraints. In areas that were readly penetrated by humans on the forested slopes of the Andes, substantial modification and settlement had occurred by the mid-Holocene. In hard-to-reach areas, however, the amount of human modification may always have been minimal, and these slopes can be considered as being close to natural in their vegetation. This article is part of the theme issue ‘Tropical forests in the deep human past’.