Informing pro-poor water management in Ethiopia

Decades of water resource management have convinced many about the relevance of dealing with resources from a basin-wide perspective. Coherently, EScGD researchers have been working toward devising a set of tools to improving water management and decision making in the Central Rift Valley basin. Their work is premised upon the notion that in water stressed contexts it is crucial to conduct equitable, effective and efficient water allocation, which, in turn, requires rigorous information as well as the local appropriation of participatory decision making process.

Ethiopia’s Central Rift Valley encompasses a chain of four large lakes – i.e. Ziway, Langano, Abyata and Shala. Though highly productive in edible fish and a wide variety of aquatic and wild life, the four lakes altogether are increasingly under degradation as a result of human activities. And albeit dwellers in the valley inhabitants seem aware of the impacts of diminishing water resources, they seem reluctant to accept that upstream water abstraction is detrimental for downstream users and ecosystems. In fact, it is not uncommon to hear lack of water stems from a reduction in rainfall – despite the available empirical evidence showing that precipitation has remained stable over the last 50 years. To further compound the problem, water resources may suffer from increased depletion in the mid-term, as the population doubles, reaching an estimate of four to five million people by 2035.

To address such challenges, between July and September 2009 we asked local stakeholders a number of questions: which practices underpin water management across the basin, how they perceive the policies that are being implemented, or in which ways they participate in the elaboration of public efforts to tackle water-related environmental problems. In a second stage, we gathered quantitative data and set out to characterize the basin. We did so by employing the SWAT model developed for the USDA Agricultural Research Service. SWAT model is well suited to assess the impacts over long periods of time of land management practices over large complex watersheds with varying soils, land use and management conditions. Along the process we had to overcome a set of challenges, though. To start with, the calibration of the model was not straightforward, since we were compelled to combine a variety of sources. Hence, and in order to reduce data uncertainty and model complexity, a previous hydrologic assessment of the basin based in HEC-HMS simulation was deemed crucial. HEC-HMS was developed by the United States Army Corps of Engineers (USACoE) Hydrologic Engineering Center (HEC) to run precipitation-runoff simulations for a variety of applications in dendritic watershed systems. As a first approach HEC-HMS was implemented for basin modeling in order to get physical parameters of interest; the results we obtained from HEC-HMS calibration were used to setup the accuracy of the ArcSWAT numerical modelling.

Fortunately, our efforts have yielded some preliminary results. Thus, for instance, our analysis suggests that basin management would be improved if Regional Basin Organizations prioritize agreements amongst stakeholders on how to allocate water and rationalize withdrawals, and based upon criteria of accountability, equity and efficiency. The latter will only be possible, however, if at least three key aspects are thoroughly understood and assessed: (1) how water dynamics are perceived at the basin level, including the different impacts of water use; (2) how environmental services related to water and land management are valued; and (3) which are the socio-economic impacts of managing such resources.