Lake evaporation estimates in tropical Africa (Lake Ziway, Ethiopia)

• Published in: Journal of hydrology (Amsterdam) 2001-05, Vol.245 (1/4), p.1-18
• Main Authors: Vallet-Coulomb, C, Legesse, D, Gasse, F, Travi, Y, Chernet, T
• Format: Article
• Language: English
• Subjects: air; air temperature; basins; Continental interfaces, environment; Earth sciences; Earth, ocean, space; energy balance; Environmental Sciences; Ethiopia, Ziway L; evaporation rate; Exact sciences and technology; Global Changes; groundwater; humidity; Hydrology; Hydrology. Hydrogeology; lakes; meteorological data; rivers; Sciences of the Universe; solar radiation; uncertainty; water balance
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/S0022-1694(01)00341-9

Estimates of evaporation from an open shallow lake in tropical Africa (Lake Ziway, Main Ethiopian Rift) are made by using monthly hydrometeorological data available for the past three decades. On the one hand, annual average estimates are inferred from three climatic approaches, which can be applied in areas with limited meteorological data. The lake energy balance yields an evaporation rate of 1780 mm yr-1, assuming a Bowen ratio of 0.15 (that of Lake Victoria). The Penman method gives an annual evaporation rate of 1870 mm. The complementary relationship lake evaporation model (CRLE) applied on monthly averaged values of air temperature, air humidity and sunshine duration gives 1730 mm yr-1. The sensitivity of each method to changes in input variables is analyzed in order to test the stability of the resulting estimates. This helps discuss uncertainties and possible inter-annual variations of the evaporation rate. On the other hand, the monthly lake level records together with precipitation and river discharge data between 1969 and 1990, allow us to estimate the water balance, providing an annual rate of 1937 mm for the combined evaporation and groundwater losses. The chloride budget is used to discriminate the groundwater from the evaporation loss. It gives us an annual evaporation rate of 1740 mm and a corresponding groundwater loss of 200 mm yr-1. The groundwater loss estimate is of the same order of magnitude as the surface outflow, but the associated error in the former is significant because the result is sensitive to the poorly known chloride content of river inflows. Our results can be used to forecast the impact of increased water consumption in the basin.