Exploring drivers of within-field crop yield variation using a national precision yield network.
Abstract
While abiotic drivers of yields represent important limiting factors to crop productivity, the role of biotic drivers that could be directly managed by farmers (e.g. agri-environment schemes supporting key ecosystem services) remains poorly understood. Precision yield mapping provides an opportunity to understand the factors that limit agricultural yield through the interpretation of high-resolution cropping data. This has the potential to inform future precision agricultural management, such as the targeted application of agrochemicals, promoting increased sustainability in modern agricultural systems. We used precision yield measurements from a network of 1174 fields in England (2006-2020) to identify drivers of within-field yield variation in winter wheat and oilseed rape. Potential drivers included climate, topography and landscape composition and configuration. We then explored relationships between in-field yield patterns and local landscape context, including the presence of features associated with ecosystem benefits. Proximity to the field edge was associated with reduced yields in 85% of wheat and 87% of oilseed fields. This translating to an approximate reduction of 10% in wheat and 18% in oilseed yields lost due to field edge effects. We found evidence that reduced yields at the field edges were associated with biotic features of the surrounding landscape, including the occurrence of semi-natural habitats. Specifically, agri-environment scheme (AES) presence increased the rate at which yields at field edges approach those of the field centres. This suggests that AES occurrence within a landscape (rather than field adjacent) may increase edge effects. However, these trends are unclear and suggest interactions between drivers and the spatial and temporal scale of investigation. Synthesis and applications. While we found evidence of landscape context mitigating against field edge effects, these were counterintuitive. For example, AES at a landscape scale appeared to increase the severity of edge effects. This study highlights a lack of environmental data at sufficiently high spatiotemporal resolution to match that of precision agriculture data. This mismatch is hindering the effective integration of precision agriculture data in an environmental policy and/or management context and potentially leading to unnecessarily poorly informed decisions related to AES deployment. This may limit environmental and economic benefits.