DEFORESTATION AND METHANE RELEASE FROM TERMITES IN AMAZONIA

To assess the role of termite populations in the change of global atmo spheric methane concentrations, we reevaluate the hypothesis that defo restation leads to higher populations of wood-feeding termites and to a significant increase of termite-emitted methane in areas of cleared and burned former primary rain forest. Calculations are based on a mod el that uses literature information on termite population size in prim ary forest and pasture 1 to 10 years after forest conversion, wood con sumption and methane emission rates of termites. We use two scenarios based on low- and high-end parameters based on data from rain forests in Brazilian Amazonia. In the low-end scenario, termite population bio mass is 25 kg . ha(-1) in primary forest; 4 kg . ha(-1) in year 1 afte r forest clearing, 51 kg . ha(-1) in a six-year-old pasture, and 4 kg . ha(-1) in a ten-year-old pasture. In the high-end scenario, all valu es are doubled and the initial breakdown in year 1 is omitted. Wood co nsumption rates are 49 and 270 mg wood . g termite . day(-1), and meth ane emission rates are 0.0023 and 0.0079 t of carbon released as metha ne per ton of carbon consumed, in the low- and the high-end scenario, respectively. In the low-end scenario no significant difference exists between the average termite population size in primary forest and pas ture modeled over a ten-year period. In the high-end scenario the aver age population size of years 1-10 after clearing is only 31% over that of primary forest. The population model data combined with the wood c onsumption rates allow for only 2.3-32.3% of the wood biomass left fro m forest bun to be consumed by termites within 10 years. The changes i n methane emissions from termite population change after deforestation were calculated using two approaches: ''Cumulative net emissions'' fo r the region, which measure the 10-year impact of a year's forest clea ring (e.g. 1.38 . 10(6) ha in 1990), increase by 0.0001 to 0.11 Tg CH4 in the 10 year-period in both scenarios, a negligible contribution to the increase of atmospheric methane concentrations of 45 Tg . yr(-1). The ''annual balance of net methane emissions'' from termites in all the different landscapes existing in the whole region in a single year (1990) increases by only 0.004 to 0.33 Tg CH4 (low- and high-end scen ario) because of the large proportion of old clearings >10 years old) with low methane emission rates: Termite populations do not tend to in crease as a function of the available wood mass only and therefore met hane emissions from termites in cleared areas of former rain forest do not make a significant contribution to the increase of the global met hane concentrations in the atmosphere. Copyright (C) 1996 Elsevier Sci ence Ltd