Rooting characteristics and inorganic nitrogen distribution in three landuse systems on an oxisol in western Kenya

In western Kenya intensive cropping and leaching losses of nutrients have reduced soil fertility and, consequently, led to low crop yield. As a remedial measure, farmers in the area leave portions of their land fallow. Weeds and sesbania sesban are prominent in the fallow. Planted tree fallows have been speculated to reduce leaching losses and pump up nutrients from the subsoil, but little has been studied on these processes. Since understanding of rooting characteristics was essential in this context, a field study was conducted on a Kandiudalfic Eutrudox soil to compare relationships between rooting patterns and soil nitrogen in planted sesbania fallow, natural fallow, and maize system. Two root measurement techniques (profile wall and soil monolith) were used. For the profile wall, a trench was dug in each replicate of the three landuse systems; the surface of the wall was then smoothed, sprayed with water and forked to expose roots. A transparent polythene sheet marked with 15 cm x 15 cm grids was placed against the profile wall to map the total number of exposed roots in each grid. For the monolith technique, soil from the profile wall, to the depth of rooting, was collected from 15 cm x 15 cm grids with a metal sampler (volume = 2250 cm cubic). From each soil sample, 100-200 g was subsampled for gravimetric water content, ammonium and nitrate determination. The remaining soil of each sample was soaked overnight, washed and then roots were separated from debris and sorted by plant species. Root length, root diameter and dry weight of maize, weeds and sesbania were determined. Square root transformation for root data and logarithm transformation for inorganic-N data were used and both data statistically analysed as SAS (Version 6.10). Root tip counts indicated a rooting depth of 120 cm in maize system, 240 cm in weed fallow, and greater than 405 cm in sesbania fallow. Root length (cm cm minus squared) at 0-120 cm was greater (P less than or equal two 0.05) for weed fallow than maize and sesbania. On the other hand, root length at 120-240 cm was higher for sesbania (8.0 cm cm minus squared) than weed fallow (5.5 cm cm minus squared). High root length in the top soil of weed fallow was attributed to greater percentage of fine roots. In turn, fine roots in weed fallow were greater due to the abundance of grasses such as Digitaria abyssinica and Paspalum scrobiculatum. Root dry weight in sesbania and weed fallows and maize system was high (greater than 60 percent of total) in the 0-30 cm soil depth as compared to the lower soil profile, indicating the concentration of roots in the top soil. Total plant biomass was 50.4 t ha-1 for sesbania, 9.6 t ha-1 for weeds and 4.m t ha-1 for maize. At 0-120 cm, nitrate-N in weed and sesbania fallows was lower (p less than or equal two 0.05) than the nitrate-N content in maize system, suggesting lower uptake due to less root length density and/or less demand for N by maize than sesbania and weeds. At 120-240 c