Using ecoinformatics to evaluate the impact of crop and herbicide rotations on herbicide intensity

Crop rotation is a central component of integrated weed management (IWM) under real-world conditions, yet its impact on herbicide use remains unclear. To address this challenge, we developed an ecoinformatics-driven analysis approach based on farmer-reported data from 11 farms, spanning approximately 400 fields and 3303 crop-year records across diverse arable zones. The workflow analysis was divided into three main stages: (i) trend-screening, examining temporal changes in herbicide intensity across major summer crops and rotational patterns of crops and herbicides; (ii) identification of key drivers associated with variability in herbicide intensity, and (iii) mechanistic clue finding, revealing explanations for patterns observed in the first two stages.

A significant positive association was found between herbicide intensity and years in maize (P < 0.001). The frequency of cotton in crop rotation was identified as the most consistent predictor of herbicide intensity in this crop (estimate = -0.25, P < 0.001). Maize fields in which cotton was prevalent in the rotation received approximately one fewer herbicide application than fields without cotton in the rotation. Principal component analysis (PCA) highlighted tillage as the primary driver of the differentiation between maize fields with and without a history of cotton. Temporal testing showed that the cotton effect persisted for 2 years.

Cotton-based rotations reduced herbicide inputs in maize, with cotton-specific tillage practices providing a mechanistic explanation for the weed-suppressive effect. These results highlight the advantage of IWM and demonstrate the utility of ecoinformatics for detecting rotation effects in large-scale farm data. © 2026 The Author(s). Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.