2 Minute read
This research project aims to:
- Critically evaluate the status, strengths, and weaknesses of existing data driven DSS and advisory tools used to guide water and fertiliser input management in smallholder farming systems in South Asia.
- Develop a computationally efficient and reliable approach combining crop modelling (APSIM), machine learning and optimisation techniques for determining optimal allocation of limited water and fertilizer inputs in small-scale production systems in target regions.
- Evaluate yield, income and environmental outcomes of optimised water and fertiliser management strategies in comparison with existing farmer heuristics and assess how these benefits are affected by differences in farm characteristics, production settings, model and input data uncertainty.
The details
Agriculture underpins rural livelihoods and economies across the Global South. However, crop yields achieved by smallholder farmers often fall far below agronomic potential. Limited or inefficient use of crop inputs – such as irrigation and fertilisers – is a key cause of crop yield gaps in smallholder farming systems.
A lack of reliable information about expected yield and income benefits, along with farmers’ aversion to investments in inputs like fertiliser or fuel for irrigation, is one of the main underlying drivers of low or inefficient input use in smallholder farming systems.
Development of data-driven advisory products and services that effectively empower smallholder farmers to make more informed and precise input use decisions therefore has the potential to be a game changer for rural development and climate change adaptation.
However, while digitally enabled tools for precision agriculture have scaled rapidly over recent years in high-productivity agricultural systems in regions such as Europe and Australia, their success in smallholder farming systems has been much more limited with most farmer-facing smart farming tools struggling to sustain outcomes at scale beyond the pilot phase of product development.
This PhD project will address these issues by developing new data-driven methods for precision input management in smallholder environments through integration of process-based crop modelling, advanced data analytics, and farm-level agronomic and socio-economic data stacks.
The graduate researcher on this project is: Pauline Njoki Kimani
Supervision team
- The University of Manchester: Doctor Tim Foster, Doctor Ben Parkes
- The University of Melbourne: Doctor Shu Kee Lam, Doctor Alexis Pang, Professor Pablo Zarco Tejada
First published on 29 March 2022.
Share this article
Other joint PhD projects
-
Identifying cellular and physiological factors associated with sub-clinical ocular inflammation and discomfort
This project will investigate the diurnal features of ocular surface immune cells.
-
The Sino-Japanese infrastructural rivalry in Kenyan port cities: a (geo)political ecology of competing regimes of production of spaces and mobilities
This project will conduct a comparative study of specific infrastructure projects in two Kenyan port cities.
-
Explainable Algorithm Selection and Configuration through Instance Space
This project will improve the explainability of optimisation algorithms in the context of AI.
-
Epigenomic monitoring of residual disease in melanoma using liquid biopsies
This research project will investigate the potential of circulating tumour DNA to monitor melanoma management and its potential for personalised treatments.