Soil Nitrate and Water Dynamics in Sesbania Fallows, Weed Fallows, and Maize

We hypothesized that the integration of trees into agricultural land-use systems can reduce NO., leaching and increase subsoil N utilization. A field study was conducted on a Kandiudalfic Eutrudox (Ochinga site) and a Kandic Paleustalf (Muange site) in the subhumid highlands of Kenya to measure changes in soil NO, and water to 200-cm depth for one rainy season in four land-use systems (LUS): (i) planted tree fallow using Sesbania sesban (L.) Merr., (ii) unfertilized maize (Zea mays L.), (iii) weed fallow, and (iv) bare fallow. Subsoil (50-200 cm) NOj-N at the start of the season ranged from 58 to 87 kg ha" for the four LUS and two sites. In maize, subsoil NOj-N differed by <5 kg ha~' between planting and harvest at both sites. In sesbania, subsoil NOj-N decreased by 22 kg ha~' at both sites, whereas in weed fallow subsoil NO..-N decreased by 26 and 38 kg ha~' at Ochinga and Muange, respectively. At both sites, subsoil water contents at the start of the season were similar in the four LUS; but at the end of the season, soil water at 100 to 200 cm was significantly lower for sesbania than for maize. Adsorption of NO] increased with soil depth. Sorbed NO3 at 100 to 200 cm was about 60% in the Kandiudalfic Eutrudox and about 15% in the Kandic Paleustalf. Rotation of maize with either a sesbania fallow or a weed fallow can result in more effective subsoil NO] and water utilization than maize monoculture. N FREQUENTLY ACCUMULATES in tropical Soils during the onset of rains following a dry season (Birch, 1958; Semb and Robinson, 1969). As the rains continue, the accumulation of soil NO3 is usually followed by a rapid decrease in topsoil NO3 due to a combination of plant uptake, denitrification, immobilization, and leaching (Greenland, 1958). Leaching can result in appreciable loss of topsoil NO3 (Poss and Saragoni, 1992) and accumulation of NO3 in the subsoil (Leutenegger, 1956; Jones, 1976). Nitrate leached down the profile of tropical soils can be adsorbed on positively charged surfaces (Wild, 1972; Cahn et al., 1992). Sorption of NO3 typically increases with depth, decreased pH, decreased organic matter, increased kaolinite, and increased Fe and Al oxides (Black and Waring, 1979). Sorption of NO3 can retard downward movement of NO3 (Wong et al., 1987) and results in accumulation of NO3 below the rooting depth of crops. Michori (1993) observed 2200 kg NO3-N ha-' at 1to 5-m depth under fertilized coffee in Kenya. The peak in subsoil NO3 corresponded to a soil layer with low pH, high positive surface charge, and 1:1 clay minerals. Natural fallows have long been a way to overcome soil fertility depletion that results from continuous cropA.E. Hartemink, Univ. of Technology, Dep. of Agriculture, Private Mail Bag, Lae, Papua New Guinea; B.H. Janssen, Dep. of Soil Science and Plant Nutrition, Wageningen Agricultural Univ., P.O. Box 8005, 6700 EC Wageningen, the Netherlands; and R.J. Buresh and B. Jama, International Centre for Research in Agroforestry (ICRAF), P.O. Box 30677, Nairobi, Kenya. Received 1 Mar. 1995. *Corresponding author (r.buresh® cgnet.com). Published in Soil Sci. Soc. Am. J. 60:568-574 (1996). ping with no nutrient inputs (Nye and Greenland, 1960). As land pressure increases due to increasing population and other competing land-use demands, long-duration natural fallows are no longer a viable option. As a result, a shift is required to more permanent food-production systems that optimize nutrient cycling, enhance soil biological activity, and maximize the use efficiency of minimal external nutrient inputs (Sanchez, 1994). One such possible system is a short-duration, planted tree fallow. In planted tree fallows, a preferred tree is grown as the fallow species in rotation with cultivated crops. The ideal tree species is typically fast growing, N fixing, and efficient at nutrient capture and cycling. One promising tree species is sesbania sesban. Kwesiga and Coe (1994) showed that 1to 3-yr planted sesbania fallows increased yield of subsequent maize crops on a N-responsive soil in Zambia, but little is known about the ability of sesbania fallows, compared with natural fallows and cereal crops, to reduce the loss of soil NO3 and to utilize subsoil NO3. The objective of this study was to compare the effects of a sesbania planted fallow, a weed fallow, and maize on seasonal changes in soil water and NO3 to the 200-cm depth. A bare fallow was included as a control in order to assess soil water and NO3 status in the absence of plant growth. MATERIALS AND METHODS