Hydrologie : recherche actuelle

Modelling Phosphorus Losses from Tropical Agricultural Soils in Gilgel Gibe Watershed, Ethiopia

Abstract

Yalemsew Adela and Christian Behn

Phosphorus (P) is a vital nutrient for plants, however its excess loss from agricultural lands cause eutrophication on aquatic environment. The Gilgel gibe reservoir located in the southwest part of Ethiopia is exposed to this phenomenon whereby the water quality has been classified as mesotrophic with P concentration of 0.86 mg/l. The objectives of this study were to identify the operating P loss mechanisms from agricultural lands, quantify the amount of P exported and evaluate the factors for P loss using the best management practices (BMPs) concepts. Therefore, experimental data and the annual phosphorus loss estimation (APLE) model were used to study the underlying processes. Catenas surrounding the reservoir, used as arable and pastureland, were investigated. Topsoil samples were taken and analyzed at three slope positions. The soils are mainly nitisol and a smaller portion of vertisol where the parent materials are basalt and rhyolite. The APLE model was set using soil data from the arable lands with nitisol soil and hydrologic records. The phosphorus loss was simulated from 2001 to 2010. Besides, the experimental P sorption data were used to check the logical consistency of the model output. On average 12.66 ± 0.7 kg P ha-1 yr-1 is lost in the form of particulate and total dissolved P. Generally, 56% of P is lost in the form of particulate P due to erosion, and 44% as soils dissolved and direct fertilizer runoff P. A significant variation observed between the sediment and soil dissolved P loss (p-value= 0.000) which is attributed to the soil chemical and physical properties that control the phosphorus dynamics. Obviously the dominant P transfer from agricultural lands into the Gilgel Gibe River and reservoir is particulate P loss. An evaluation of causing factors using BMPs indicated that a reduction of sediment by 5-20% resulted to retain P from 2-9%. Similarly, a reduction of soil P content reduces the P loss from 2-8.5%. However, a reduction of fertilizer quantity applied on the fields within the same percent range is hardly reducing P loss relative to the earlier factors. Therefore, attention should be given to the application of precision agriculture to avoid such problems.