Modeling Database

Data from the subhumid agroclimatic zone (Lake Victoria crescent) were used to test the predictive performance of the model. The data set was compiled from 1998 to 2002, using runoff plots (15 x 10 m) established in four major agricultural land use types (annual crops, banana, coffee, and range lands), under the Lake Victoria Environmental Management Project (LVEMP). Each land use type was replicated three times, except banana, which was replicated four times. Plots under banana were mulched and not tilled, while those under coffee were neither tilled nor mulched. Soil chemical properties including soil organic carbon were determined every season from 0- to 30-cm depths. Input of organic matter (litter

Table 20.3 Modeling Parameters for Specific Sequestration Model

Historical Maximum Minimum Soil Agroclimatic Soil C (0-30 cm) Organic Carbon (Cr) Zones (Cmax) (%) (%)

Subhumid 4.7 <1.5

Source: Compiled from Radwanski, S.A., Ed. 1960. The Soils and Land-Use of Uganda. Memoirs of the Research Division, Series 1, no. 4, Soils. Uganda Protectorate, Department of Agriculture, Kampala, Uganda.

and mulch) was also monitored seasonally. The amount of soil organic carbon lost through water erosion for each season was determined by multiplication of the seasonal soil organic carbon concentration by seasonal soil loss. We assumed that the minimum/passive soil organic carbon, Cr = 1.5% (Paul and Clark, 1996), and the decomposition rate of organic matter to be 50% (Lal and Kimble, 2000). The maximum attainable rate of soil organic carbon loss was estimated for each of the four land use types (annual crops, banana, coffee, and range lands) as the difference between the historical maximum soil organic carbon and the minimum/passive soil organic carbon.

The model parameter Cs was estimated using regression method. Cs - Cr was the slope of C(ln(Lmax - L)/Lmax). Lmax was obtained as a product of (Cmax - Cr) and the mass of slice topsoil layer (0- to 30-cm depth). L was then taken to be the soil organic carbon content lost through erosion from the different land use types (annual crops, banana, coffee, and range lands). The relative growth rate of soil organic carbon (g) was the average of 4 years in g values computed from Equation 20.1. The parameters Cs and g are presented in Table 20.4. The change in soil organic carbon over a period of four years was then predicted and compared with observed data using regression analysis.

Table 20.4 Model Parameters Computed for Sub-humid Lake Victoria Crescent




(Soil Organic Carbon

Land Use


Relative Growth Rate)










Range land



Was this article helpful?

0 0

Post a comment