Quantifying the Stock of Soil Organic Carbon using Multiple Regression Model in a Fallow Vegetation, Southern Nigeria
Identifying ecological variables that explain significant variation in the stock of carbon is indeed one way of sustaining its concentration in the soil. The stepwise multiple regression model was employed to identify ecological variables that explained significant variation of carbon in fallow soils. Using fallow genealogical cycles of 1st, 2nd, 3rd, 4th and 5th generations, soil and vegetation variables from 30 sampling plots were collected and subjected to linear regression analysis. The analysis generated three predictive models. The first and second models significantly (p<0.01) explained 78% (R2 = 0.78) and 87% (R2 = 0.87) of the variability in soil organic carbon (SOC), while the third (full) model significantly (p<0.01) explained 89% in the variability of carbon stock (R2 = 0.89) with vegetation cover, available phosphorus and total nitrogen being the most significant predictor variables. The full model was upheld because it identified three significant ecological variables that explained increased variability in the stock of SOC. The study suggested that management of the ecological variables identified in the full model which indeed were associated with the abundance of woody and herbaceous vegetation would not only increase the stock of SOC in the soil, but reduced its concentration in the atmosphere. For this to be feasible mostly in the present changing climate, healthy forest and land management practices, such as the creation of vegetal buffer zones around farmlands, zero-tillage practice, mulching, retaining of forest slash and crop residues, fertilizer application, elongation of fallow periods, and tree planting initiatives in degraded ecosystems were encouraged.
Keywords: Stepwise Regression, SOC, Healthy Forest and Management Practices Fallow Genealogy,
Woody Perennial Tree Species, Fallow Elongation