Digital Agriculture PhD Supplementary Scholarship projects



Exploring crop performance through optimisation in multiple dimensions

Supervisors: Quanxi Shao (CSIRO), Bangyou Zheng (CSIRO), Eric Stone (ANU) and Kerry Taylor (ANU).

The economic yield of a crop is determined by the interactions of genotype and environment. Characterisation of newly released cultivars in diverse environments is a challenge for breeders and growers. Currently, multiple environment trials (MET) are conducted where crop and environment observations are collected and analysed to represent crop performance.

Crop growth modelling is a robust simulation tool based on the interaction of crop physiology with environments. In the model, an assumption is made that environments are well observed, and a crop is characterised by optimisation of genotypic parameters. In reality, the model is unlikely to be provided with accurate records for environmental growth conditions (due to the geographical and temporal variations in MET), or they may be unavailable due to financial constraints. Consequently, the model will need to optimise multiple dimensions (i.e. multiple genotypic parameters and multiple environmental variables) to be useful to breeders and growers.

The PhD candidate will devise an analytical solution, using existing datasets from MET, to optimise genotype parameters and environment variables in a single pipeline, i.e. multiple dimension optimisation. The project will solve the underlying analytical challenge; deliver a solution to the crop industry; and may be extended into broader research domains.

The project will be supported by a multidisciplinary team of researchers with experience in farming systems modelling, agronomy, crop physiology, and data science as well as to CSIRO’s broader team working in Digital Agriculture in the pursuit of innovative solutions for resilient farming.

The successful candidate will be provided with:

1) Training in agronomy, simulation and associated data analysis and machine learning by CSIRO farming systems scientists to further explore model outputs. A background including coding in either R or Python would be desirable.

2) An in-depth understanding of crop breeding, and crop physiology.

3) Exposure to, and understanding of, the latest techniques in digital agriculture.

5) Transferable core scientific skills including data management, analysis, presentations, paper writing, and peer review, as well as working in a multi-disciplinary team.