Responses of tree growth and biomass production to nutrient addition in a semi-deciduous tropical forest in Africa.

Experimental evidence of nutrient limitations on primary productivity in Afrotropical forests is rare and globally underrepresented yet are crucial for understanding constraints to terrestrial carbon uptake. In an ecosystem-scale nutrient manipulation experiment, we assessed the early responses of tree growth rates among different tree sizes, taxonomic species, and at a community level in a humid tropical forest in Uganda. Following a full factorial design, we established 32 (eight treatments × four replicates) experimental plots of 40 × 40 m each. We added nitrogen (N), phosphorus (P), potassium (K), their combinations (NP, NK, PK, and NPK), and control at the rates of 125 kg N ha-1 year-1 , 50 kg P ha-1 year-1 and 50 kg K ha-1 year-1 , split into four equal applications, and measured stem growth of more than 15,000 trees with diameter at breast height (dbh) ≥1 cm. After 2 years, the response of tree stem growth to nutrient additions was dependent on tree sizes, species and leaf habit but not community wide. First, tree stem growth increased under N additions, primarily among medium-sized trees (10-30 cm dbh), and in trees of Lasiodiscus mildbraedii in the second year of the experiment. Second, K limitation was evident in semi-deciduous trees, which increased stem growth by 46% in +K than -K treatments, following a strong, prolonged dry season during the first year of the experiment. This highlights the key role of K in stomatal regulation and maintenance of water balance in trees, particularly under water-stressed conditions. Third, the role of P in promoting tree growth and carbon accumulation rates in this forest on highly weathered soils was rather not pronounced; nonetheless, mortality among saplings (1-5 cm dbh) was reduced by 30% in +P than in -P treatments. Although stem growth responses to nutrient interaction effects were positive or negative (likely depending on nutrient combinations and climate variability), our results underscore the fact that, in a highly diverse forest ecosystem, multiple nutrients and not one single nutrient regulate tree growth and aboveground carbon uptake due to varying nutrient requirements and acquisition strategies of different tree sizes, species, and leaf habits.

Conversion of lowland tropical forests to tree cash crop plantations loses up to one-half of stored soil organic carbon.

Tropical deforestation for the establishment of tree cash crop plantations causes significant alterations to soil organic carbon (SOC) dynamics. Despite this recognition, the current Intergovernmental Panel on Climate Change (IPCC) tier 1 method has a SOC change factor of 1 (no SOC loss) for conversion of forests to perennial tree crops, because of scarcity of SOC data. In this pantropic study, conducted in active deforestation regions of Indonesia, Cameroon, and Peru, we quantified the impact of forest conversion to oil palm (Elaeis guineensis), rubber (Hevea brasiliensis), and cacao (Theobroma cacao) agroforestry plantations on SOC stocks within 3-m depth in deeply weathered mineral soils. We also investigated the underlying biophysical controls regulating SOC stock changes. Using a space-for-time substitution approach, we compared SOC stocks from paired forests (n = 32) and adjacent plantations (n = 54). Our study showed that deforestation for tree plantations decreased SOC stocks by up to 50%. The key variable that predicted SOC changes across plantations was the amount of SOC present in the forest before conversion--the higher the initial SOC, the higher the loss. Decreases in SOC stocks were most pronounced in the topsoil, although older plantations showed considerable SOC losses below 1-m depth. Our results suggest that (i) the IPCC tier 1 method should be revised from its current SOC change factor of 1 to 0.6 ± 0.1 for oil palm and cacao agroforestry plantations and 0.8 ± 0.3 for rubber plantations in the humid tropics; and (ii) land use management policies should protect natural forests on carbon-rich mineral soils to minimize SOC losses.