Linking Hydrology, Ecosystem Function, and Livelihood Outcomes in African Papyrus Wetlands Using a Bayesian Network Model

Relationships between hydrology, ecosystem func- tion and livelihood outcomes in Nyando papyrus wetland in western Kenya were integrated in a Bayesian Network model with 34 nodes which was populated with results of ecohydrological and socio-economicresearchand stakeholder consultations. Scenarios for dry and wet seasons were evalu- ated. For a current iaveragei year in Nyando wetland the probabilities of iEcosystem functioni and iLivelihoods out- comesi being igoodi were 62 and 19 %, respectively. Under dry conditions, these values changed to 33 and 37 %; and under wet conditions to 85 and 6 %, respectively, indicating that wet conditions had a positive effect on the ecosystem but a negative effect on livelihoods. Ecosystem function and livelihood outcomes were most sensitive to flooding, conver- sion to agriculture, livestock grazing, and papyrus harvest. Flooded conditions limit cropping, livestock herding and veg- etation harvesting but have a positive effect on ecosystem function. The advantage of this interdisciplinary and partici- patory modeling approach is that it allows incorporation of formal and informal knowledge, allows evaluation of policy scenarios and trade-offs between ecosystem services, and recognizes uncertainty in system outcomes. Further work will incorporate the impact of management policies and institu- tions on resource use. This approach will be useful in decision support for wise use of wetlands.