Conference Diary
From Perth to Wheat Production: Notes after IPC2025
Presenting a geospatial machine learning study in Perth was a useful test of how clearly spatial evidence can speak to agricultural questions.
Context
IPC2025 offered a compact setting to discuss how geospatial machine learning can describe spatial disparities in wheat production. My oral report focused on why climate, environmental, and soil factors should be interpreted with their spatial structure rather than treated as isolated variables.
The Presentation
The core message was simple: a production map is not just a map of values. It is also a record of where determinants interact, where regional differences matter, and where the same management strategy may not transfer well from one place to another.
Conversations
The most useful conversations were about translation. How can a method that identifies spatial determinants become useful for crop management, adaptive variety selection, and regional planning? These questions are exactly where method development needs to meet domain knowledge.
Notes to Keep
A conference report is often less about the talk itself and more about the questions it leaves behind. For me, the next step is to keep refining how spatial heterogeneity can be communicated to people who work with land, climate, and food systems every day.