Topic ID No.: 2026-G04
Title of research topic:
Integration of data-driven approaches into climate-resilient on-farm water management under semi-arid environments - field study part
Description:
The research addresses improving on-farm water management in semi-arid environments by integrating data science with agricultural, ecological and hydrological field studies. Combining high-resolution remote sensing imagery, sensors, and machine learning models with ecosystem field data as affected by management, will improve our system's understanding with the aim to optimize water use, enhance productivity, and build climate resilience.
Target region or country (if applicable):
Semi-arid Southern Africa (South Africa, Botswana, Namibia, Malawi, Zambia)
Topic background information / scientific relevance:
Agricultural production and food systems in semi-arid environments faces severe challenges due to limited and highly variable rainfall and frequent droughts. These conditions lead to water scarcity, soil degradation, and reduced productivity, threatening food security and farmers’ livelihoods. Improving on-farm water management is therefore critical to increase resilience of agricultural production in these regions.
Integrating data science and data driven approaches into on-farm water management opens a powerful research direction that combines agriculture, ecology, hydrology, and digital innovation. The improved availability of high-resolution satellite or drone based imagery, in-situ sensors, and climate datasets can provide real-time information on soil moisture, vegetation water(stress) feedbacks, and weather conditions. By integrating these data sources, machine learning and predictive analytics using e.g. digital twins of landscapes or farms can model complex soil–water–plant–climate interactions and generate forecasts of water demand and pathways. Decision support systems built on these models can help optimize farm water management, thus improving water use efficiency to maintain farm productivity and enhance climate resilience.
However, the integration of data-driven approaches into climate-resilient on-farm water management remains a challenge, particularly in semi-arid regions with fragmented data infrastructures and resource constraints. Consequently, there is a need to improve our system’s understanding based on evidence derived from integration of data science and agronomic, ecological, hydrological and socio-economical field studies. This will contribute to provide options for context-specific, scalable, and farmer-friendly management interventions to enhance the efficiency, sustainability, and resilience of farming systems under semi-arid environments.
Research objectives:
1. To improve our understanding of soil-water-plant-climate interactions as affected by management under semi-arid environments.
2. To improve on-farm water management and agricultural productivity in semi-arid environments through the application of data science techniques for real-time monitoring, prediction, and decision support.
3. To integrate multi-source environmental and agricultural data using data-driven and machine learning techniques to monitor and explain on-farm water dynamics in semi-arid environments as affected by management.
4. To develop and validate models for predicting plot, farm or landscape level water dynamics under variable climate and management.
5. To design a data-driven decision support framework that enhances farm-level climate resilience through optimized water management and evaluate the performance and scalability.
Required skills and qualifications of the applicant:
The ideal candidate addressing the field study part of the research topic should hold a Master’s degree in Agriculture, Ecology, Biology, Environmental Sciences, or related fields. A solid academic background in ecosystem functioning and ecological processes as well as water related environmental sciences is required. Advanced experience and knowledge in ecological field methods and/or water related field studies should be present. Experience in data science, crop or water household modelling and statistical methods will be added advantage. The candidate should demonstrate strong problem-solving and technology skills, willing to spend considerable time in carrying out field work under difficult environmental conditions. Experiences in farm management and communication with farmers and agriculture-based communities will be an asset. We require the ability to work independently and collaboratively in an interdisciplinary and international research environment.
Contact person and institute in charge:
Dr. Marcus Giese, Dr. Marc Cotter. University of Hohenheim. Institute for Agricultural Sciences in the Tropics (Hans-Ruthenberg-Institute)