Estimating Soil Carbon Pool Sizes from Texture and Profile Data.

Increasing threats of global warming has rekindled research in soil carbon (SOC) enhancement to offset atmospheric CO2 loading. Although SOC may change rapidly in the short-term, actual build occurs in the medium to long-term making modeling an appropriate tool for such studies. Because SOC is not physico-chemically homogenous, models must simulate the different pools. Information on pools sizes is often lacking, requiring indirect methods to initialize them. Our aim was to test the hypotheses that for soils having low activity clay, the stable pool (SOCs) size can be directly estimated from % clay + silt and that the soil type and field history provide an indication of SOC fractions. We developed relationships between SOC (in clay and silt) and % (clay +silt) using published data by fitting lower and upper envelope curves to the data. We assumed that the lower curve: SOCs (g/kg) = 0.15 x % (clay + silt) + 0.69 (R2 = 0.95), represented the stable pool size. This equation was tested on data from 135 and 56 farms at Wa and Kpeve, respectively, in the savanna and forest zones of Ghana. Additional profile data were obtained for 10 benchmark soils from India. A structured query method was developed for estimating the SOC fractions from soil type and field history. Our results showed good prediction for Wa (r = 0.85), where a large proportion of the SOC was in the stable pool. Despite the large carbon input at Kpeve, the prediction was good for degraded soils. The correlation between predicted SOCs and measured SOC (40-60 cm) for the Indian profiles was high (r = 0.81; P < 0.05). We conclude that SOC pool sizes could be estimated from readily available soil data, minimizing the need for indirect calculations.