Abstract
An evaluation of crop management factor (C) for coffee using USLE method was conducted in hilly humid tropical area of Lampung, Indonesia. The treatments were as follows : coffee without cover crop (clean-weeded plot) ; coffee with Paspalum conjugatum as cover crop (Paspalum plot); and coffee with natural weeds (natural weeds plot). Weed management was done every two weeks by clearing all the weeds in clean-weeded plot, and cutting the weeds around the coffee trees at a diameter of 1 m for the weedy plots (Paspalum and natural weeds plots). Two methods of estimating C-factor for coffee were used :(1)using similar condition with other crops (Ct), and (2) using equivalent method based on the existing value of coffee-C factor (Ce). The results showed that the use of Ct gave soil loss prediction 9-24 times higher than that measured, while the use of Ce gave 10-81 times higher. The predicted values of soil loss using Ct were 7.7 t/ha/year and 14.1 t/ha/year for Paspalum and natural weeds plots respectively. These values were still acceptable and reasonable to the soil loss tolerance, and very low compared to the other Indonesian studies which could be hundreds of ton/ha/y. This experiment showed that the measured coffee C-factor was 0.045 for clean-weeded plot, 0.006 for Paspalum plot and 0.004 for natural weeds plot, which were lower than the common value (0.2) usually used in Indonesia. By introducing the effect of weeds as the weeds C-subfactor (Cs) and coffee C-factor (Cb) obtained from this experiment measurement, the coffee C-factor (Cc) with various weeds coverage could be estimated by the equation Cc=Cb Cs.
