ESTIMATE OF CO2 EFFLUX OF SOIL, OF A TRANSITION FOREST IN NORTHWEST OF MATO GROSSO STATE, USING MULTIPLE REGRESSION

Abstract

Many research groups have being studying the contribution of tropical forests to the global carbon cycle, and the climatic consequences of substituting the forests for pastures. Considering that soil CO2 efflux is the greater component of the carbon cycle of the biosphere, this work found an equation for estimating the soil CO2 efflux of an area of the Transition Forest, using a model of multiple regression for time series data of temperature and soil moisture. The study was carried out in the northwest of Mato Grosso, Brazil (11°24.75’S; 55°19.50’W), in a transition forest between cerrado and AmazonForest, 50 km far from Sinop county. Each month, throughout one year, it was measured soil CO2 efflux, temperature and soil moisture. The annual average of soil CO2 efflux was 7.5 ± 0.6 (mean ± SE) ì mol m-2 s-1, the annual mean soil temperature was 25,06 ± 0.12 (mean ± SE) ºC. The study indicated that the humidity had high influence on soil CO2 efflux; however the results were more significant using a multiple regression model that estimated the logarithm of soil CO2 efflux, considering time, soil moisture and the interaction between time duration and the inverse of soil temperature.