In the tropics, shifting cultivation has long been attributed to large scale forest degradation, and remains a major source of uncertainty in forest carbon accounting. In the Philippines, shifting cultivation, locally known as kaingin, is a major land-use in upland areas. We measured the distribution and recovery of aboveground biomass carbon along a fallow gradient in post-kaingin secondary forests in an upland area in the Philippines. We found significantly higher carbon in the aboveground total biomass and living woody biomass in old-growth forest, while coarse dead wood biomass carbon was higher in the new fallow sites. For young through to the oldest fallow secondary forests, there was a progressive recovery of biomass carbon evident. Multivariate analysis indicates patch size as an influential factor in explaining the variation in biomass carbon recovery in secondary forests after shifting cultivation. Our study indicates secondary forests after shifting cultivation are substantial carbon sinks and that this capacity to store carbon increases with abandonment age. Large trees contribute most to aboveground biomass. A better understanding of the relative contribution of different biomass sources in aboveground total forest biomass, however, is necessary to fully capture the value of such landscapes from forest management, restoration and conservation perspectives.Secondary forests comprise more than half of the total forest area in tropical regions and are the dominant forest type 1 . In the tropics, secondary forests also represent a major global carbon sink that rapidly accumulates carbon in aboveground biomass [2][3][4] . Because they cover a large area in the tropics, accurate estimates of carbon in secondary forests are critical for quantifying the global carbon balance as well as for the successful implementation of climate change mitigation projects 5 . However, there remains a high level of uncertainty in tropical forest carbon accounting, firstly due to the unknown amount of deforestation and forest degradation 6 , and secondly due to a limited number of field studies that have estimated standing biomass in secondary forests 7