Introduction to the Waikato Maize Mapping Project

Since October 2024, I have been developing a research project focused on mapping and monitoring maize cultivation across the Waikato region. This initiative uses remote sensing and drone-based photogrammetry to identify seasonal crop rotation patterns and to derive yearly statistics on planting times, field extent, and yield-related observations.

The project is anchored by a representative study site on a dairy farm near Ngātea in the Hauraki Plains. Figure 1 presents a drone-captured image from this area, highlighting the uniform planting patterns and row structure characteristic of large-scale maize cultivation. This base site provides a controlled environment for testing data capture methods and verifying remote sensing observations.

Figure 1. Harvest time. Photo S A Bardsley

Figure 2. Post harvest residue. Photo: S A Bardsley

Ground-truthing has been a critical part of the project. I have collected roadside photographs (Figure 2) of maize fields across various growth stages, from early emergence through to post-harvest residue. Figure 3 shows an early-stage image where maize seedlings are just breaking the surface, with distinct row spacing visible. These photos form a vital training and validation dataset for the machine learning classification component of the project, which will map maize extent and condition across multiple seasons.

Figure 3. Early growth maize. Photo: S A Bardsley

The remote sensing workflow is based on multi-temporal Sentinel-2 imagery, now further enhanced by the improved revisit frequency following the launch of Sentinel-2C. This allows for better temporal coverage during key stages of the crop lifecycle. To address cloud-related gaps in optical data, the project also explores the integration of Synthetic Aperture Radar (SAR) data.

A Python-based processing pipeline is being developed to automate classification and produce repeatable, robust outputs. The ultimate goal is to generate spatial insights that benefit both scientific understanding and practical applications, such as informing regional planning, farm-level decisions, and assessing the impact of climate variability on crop productivity.