Soil and Carbon Losses under Rainfall Simulation from Two Contrasting Soils under Maize-Improved Fallows Rotation in Eastern Zambia

Abstract

Fallows have been reported to improve soi1 organic carbon (OC), but information on how much OC is lost through erosion is not readily available. The aim of this study was to determine potential sediment OC losses during the cropping phase in a maize-improved fallow rotation, The investigation was carried out on two contrasting soils under improved fallow treatments using simulated rainfall at 35 mm h-’ on l-m2 microplots, at Kalunga, which had 7% clay, and Msekera with 15% clay (O-20 cm). The simulations involved dry runs and wet r-uns. Four treatments studied were, Sesbania sesban, Tephrosia vogelii, natural fallow (NF) and continuous maize (Zea mays). At Kalunga time to runoff and steady state infiltration rate generally decreased, whereas runoff, soi1 loss, sediment enrichment in OC, and OC loss generally increased from October 2000 to October 2001 (dry runs), and fi-om dry to wet runs in October 2001. Runoff and erosion were significantly greater under continuous maize than in fallow treatments, but the differences decreased from 2000 to 2001 and from dry to wet runs. In 2001, loss of OC during dry r-uns was twice greater under continuous maize than in fallow treatments, but differences between treatments were smaller during wet runs. Differences between fallow treatments were small in general, thus as compared with natural fallow, improved fallow did not seem any better for soi1 and water conservation at Kalunga. At Msekera, similar overall tendencies were observed in the comparisons between 2000 and 2001, between dry and wet runs, and between treatments. The main differences with Kalunga were the greater soi1 loss in natural than in improved fallow treatments and the tendency to greater OC loss in NF than in S. sesban treatment (in 2001), thus the greater interest of planted fallows. Additionally, sediment enrichment in OC was smaller in continuous maize than in fallow treatments at Msekera, but not at Kalunga. On the whole, soi1 susceptibility to runoff and OC losses was generally greater at Msekera (15% clay, 4% slope) than at Kalunga (7% clay, 2% slope), but differences between sites were less clear for soi1 loss, and sediment enrichment in OC was generally smaller at Msekera. We concluded that fallowing reduced soi1 and OC losses, and that improved fallows were not superior to natural fallow except for soi1 loss at Msekera site.