Developing an Impact-Based Combined Drought Index for Monitoring Crop Yield Anomalies in the Upper Blue Nile Basin, Ethiopia

Having a robust drought monitoring system for Ethiopia is crucial to mitigate the adverse impacts of droughts. Yet, such monitoring system still lacks in Ethiopia, and in the Upper Blue Nile (UBN) basin in particular. Several drought indices exist to monitor drought, however, these indices are unable, individually, to provide concise information on the occurrence of meteorological, agricultural and hydrological droughts. A combined drought index (CDI) using several meteorological, agricultural and hydrological drought indices can indicate the occurrence of all drought types, and can provide information that facilitates the drought management decision-making process. This thesis proposes an impact-based combined drought index (CDI) and a regression prediction model of crop yield anomalies for the UBN basin. The impact-based CDI is defined as a drought index that optimally combines the information embedded in other drought indices for monitoring a certain impact of drought, i.e. crop yield for the UBN. The developed CDI and the regression model have shown to be effective in indicating historic drought events in UBN basin. The impact-based CDI could potentially be used in the future development of drought monitoring in the UBN basin and support decision making in order to mitigate adverse drought impacts.

1. Introduction 1.1 Background 1.2 Drought monitoring 1.3 Problem statement 1.4 Research objectives 1.5 Main steps in research methodology1.6 Research significance and innovation 1.7 Description of the study area 1.8 Dissertation structure 2. Spatio-temporal assessment of meteorological drought under the influence of varying record length 2.1 Introduction 2.2 Stations selection and data analysis 2.3 Selection of the Probability Distribution Function (PDF) for the Standardized Precipitation Index (SPI)2.4 Methodology of experiments 2.5 Results and discussion 2.6 Conclusion 3. Comparison of the performance of six drought indices in assessing and characterising historic drought events 3.1 Introduction 3.2 Data 3.3 Drought indicators 3.4 Methods3.5 Results and discussion 3.6 Conclusion 4. Developing a combined drought index and prediction model to monitor drought-related crop yield reduction 4.1 Introduction 4.2 Data 4.3 Methods4.4 Results and discussion 4.5 Conclusions 5. Application of Earth observation data for developing a combined drought index and crop yield prediction model 5.1 Introduction 5.2 Data 5.3 Methods5.4 Results and discussion 5.5 Prediction models of crop yield anomalies 5.6 Conclusions 6. Summary, conclusions and recommendations 6.1 Summary 6.2 Conclusions 6.3 Recommendations References Appendix A: Gamma distribution based SPI calculation Appendix B: Time series of drought indices Appendix C: Spider web plots of drought indicator results for selected stations Appendix D: Scatter plots of drought indices versus crop yield anomalies Appendix E: The regression equations developed for the selected eight zones and for the four crops

Yared was born in Ambo town in Ethiopia in 1979.He studied his Bachelor of Science (BSc) degree in Irrigation Engineering at the former Arbaminch Water Technology Institute (AWTI) the later Arbaminch University where he graduated in 2002.He joined Addis Ababa Institute of Technology (AAiT), Ethiopia for Maters studies in 2007, and graduated with a MSc degree in Civil Engineering, with specialization in Hydraulic Engineering in 2009. Yared pursued a PhD in the "sandwich programme" between IHE Delft Institute for Water Education/TU-Delft, The Netherlands, and Ambo University, Ethiopia and graduated in 2018. Since 2016,he has beenworking in the National Drought Mitigation Centre (NDMC) under the school of Natural Resources of the University of Lincoln-Nebraska, USA, as a visiting scientist.