Current research concentrates on ways to investigate and improve water productivity (WP), as agriculture is today’s predominant freshwater consumer, averaging at 70% and reaching up to 93% in some regions. A growing world population will require more food and thus more water for cultivation. Regions that are already affected by physical water scarcity and which depend on irrigation for growing crops will face even greater challenges regarding their water supply. Other problems in such regions are a variable water supply, inefficient irrigation practices, and over-pumping of available groundwater resources with other adverse effects on the ecosystem. To face those challenges, strategies are needed that use the available water resources more efficiently and allow farming in a more sustainable way. This work focused on the management of sensor-based deficit irrigation (DI) systems and improvements of WP through a combined approach of simulation-optimization and irrigation experiments. In order to improve irrigation control, a new sensor called pF-meter was employed, which extended the measurement range of the commonly used tensiometers from pF 2.9 to pF 7. The following research questions were raised: (i) Is this approach a suitable strategy to improve WP; (ii) Is the sensor for irrigation control suitable; (iii) Which crop growth models are suitable to be part of that approach; and (iv) Can the combined application with experiments prove an increase of WP? The stochastic simulation-optimization approach allowed deriving parameter values for an optimal irrigation control for sensor-based full and deficit irrigation strategies. Objective was to achieve high WP with high reliability. Parameters for irrigation control included irrigation thresholds of soil-water potentials because of the working principle behind plant transpiration where pressure gradients are transmitted from the air through the plant and into the root zone. Optimal parameter values for full and deficit irrigation strategies were tested in irrigation experiments in containers in a vegetation hall with drip irrigated maize and compared to schedule-based irrigation strategies with regard to WP and water consumption. Observation data from one of the treatments was used afterwards in a simulation study to systematically investigate the parameters for implementing effective setups of DI systems. The combination of simulation-optimization and irrigation experiments proved to be a suitable approach for investigating and improving WP, as well as for deriving optimal parameter values of different irrigation strategies. This was verified in the irrigation experiment and shown through overall high WP, equally high WP between deficit and full irrigation strategies, and achieved water savings. Irrigation thresholds beyond the measurement range of tensiometers are feasible and applicable. The pF-meter performed satisfactorily and is a promising candidate for irrigation control. Suitable crop models for being part of this approach were found and their properties formulated. Factors that define the behavior of DI systems regarding WP and water consumption were investigated and assessed. This research allowed for drawing the first conclusions about the potential range of operations of sensor-based DI systems for achieving high WP with high reliability through its systematical investigation of such systems. However, this study needs validation and is therefore limited with regard to exact values of derived thresholds.

West Africa is subject to frequent yield losses due to erratic rainfall and degraded soils. At the same time, its population is expected to double by 2050. This situation is alarming in northern Togo, a West African dry savannah area, where rainfed maize is a staple food. Thus, it is necessary to improve agricultural productivity, e.g., by evaluating and introducing alternative irrigation management strategies, which may be implemented in this region. For this purpose, the present investigation focused on evaluating the potential of deficit and supplemental irrigation, as well as assessing the impact of climate and soil variability on maize yield under irrigated agriculture using irrigation optimisation strategies in northern Togo. The Optimal Climate Change Adaption Strategies in Irrigation (OCCASION) framework was adapted and employed to address the research objectives. It involves: (i) a weather generator for simulating long-term climate time series; (ii) the AquaCrop model, which was utilised to simulate the irrigation during the growing periods and the maize yield response to given irrigation management strategies; and (iii) a problem-specific algorithm for optimal irrigation scheduling with limited water supply. Five irrigation management strategies viz. T1: no irrigation (NI), T2: controlled deficit irrigation (CDI) and T3: full irrigation (FI) in the wet season, T4: controlled deficit irrigation (CDI) and T5: full irrigation (FI) in the dry season were assessed regarding their impact on maize yield in northern Togo. The results showed high variability in rainfall during the wet season, which led to substantial variability in the expected yield for NI. This variability was significantly lessened when optimised supplemental irrigation management strategies (CDI or FI) were applied. This also holds for the irrigation scenarios under the dry season. Finally, these findings were validated by an irrigation field experiment conducted at an agricultural research institute in northern Togo. Under a moderate level of deficit irrigation during the vegetative and reproductive growth stages, the above-ground biomass and the maize grain yield were reduced. However, a moderate level of deficit irrigation during the vegetative growth stage could result in similar values of water productivity to that of fully irrigated treatment. It was found that, based on the values of the statistical indicators, AquaCrop has accurately simulated the maize grain yield for all the irrigation strategies evaluated. The results of this study revealed that climate variability might engender a higher variability in the maize yields of northern Togo than soil variability does. Large- and smallscale water harvesting, access to groundwater, and irrigation infrastructures would be required for implementing the irrigation management strategies assessed in this study.:Declaration iii Declaration of Conformity v Dedication vii Acknowledgements ix Abstract xi Table of Contents xv List of Figures xvii List of Tables xix List of Acronyms and Abbreviations xxi 1. Introduction 1 1.1 Background and Problem Statement 1 1.1.1 Global Fresh and Agricultural Water Use 1 1.1.2 Erratic Rainfall, Rising Temperatures, and Soil fertility depletion in West Africa 2 1.1.3 Transboundary Water Issues in West Africa 3 1.1.4 Agriculture and Water Use in Togo 3 1.2 Objectives of the Study 4 2. State of the Art 6 2.1 Relevant Agroecosystems, Farming Systems and Irrigation Management in West Africa 6 2.2 Key Performance Indicators: Water productivity and Food Security 8 2.3 Common Approaches Used to Evaluate Crop Water Productivity 9 2.4 Key production Factors: Climate, Soil and Management 9 2.5 Crop Yield Modelling 12 2.6 Integrated Modelling 13 3. Novel Framework for Optimising Irrigation Systems in West Africa 15 3.1 Model-based Sensitivity Analysis of Climate and Management Impact on Crop Water Productivity, Water Demand and Food Security 15 3.2 Experimental Validation of the Farm Model and Management Strategies, Soil Data Analysis and Modelling 17 3.3 Joint Stochastic Analysis of the Impact of Climate and Soil Variability on Crop Water Productivity and Food Security 19 4. Overview of Publications 21 4.1 Potential of Deficit and Supplemental Irrigation under Climate Variability in Northern Togo, West Africa 21 4.2 Impact of Irrigation Strategies on Maize (Zea mays L.) Production in the Savannah Region of Northern Togo (West Africa) 22 4.3 Impact of climate and soil variability on maize (Zea mays L.) yield under full and deficit irrigation in the savannah region of northern Togo, West Africa 23 5. Conclusion and Outlook 26 References 28 A. Selected Publications of the Author 37 A.1 Potential of Deficit and Supplemental Irrigation under Climate Variability in Northern Togo, West Africa 39 A.2 Impact of Irrigation Strategies on Maize (Zea mays L.) Production in the Savannah Region of Northern Togo (West Africa) 61 A.3 Impact of Climate and Soil Variability on Maize (Zea mays L.) Yield under Full and Deficit Irrigation in the Savannah Region of Northern Togo, West Africa 81 B. Histograms of distributions of the expected maize yield in northern Togo (scenarios in the third paper) 121