Global review of meta-analyses reveals key data gaps in agricultural impact studies on biodiversity in croplands

Aim: Agriculture depends heavily on biodiversity, yet unsustainable management practices continue to affect a wide range of organisms and ecosystems at unprecedented levels worldwide. Addressing the global challenge of biodiversity loss requires access to consolidated knowledge across management practices, spatial levels, and taxonomic groups. Location: Global Time period: 1994 to 2022 Major taxa studied: Animals, microorganisms, plants. Methods: We conducted a comprehensive literature review synthesising data from all meta-analyses about the impacts of agricultural management practices on biodiversity in croplands, covering field, farm, and landscape levels. From 200 retained meta-analyses, we extracted 1,885 mean effect sizes (from 69,850 comparisons between a control and treatment) assessing the impact of management practices on biodiversity, alongside characterising over 9,000 primary papers. Results: Seven high-income countries, notably the USA, China, and Brazil dominate agricultural impact studies with fertiliser use, phytosanitary interventions and crop diversification receiving widespread attention. The focus on individual practices overshadows research at the farm and landscape level. Taxonomically, Animalia, especially arthropods, are heavily studied while taxa such as annelids and plants receive comparatively less attention. Effect sizes are predominantly calculated from averaged abundance data. Significant gaps persist in terms of studies on the effects of agricultural interventions on specific taxonomic groups (e.g. annelids, mammals) and studies analysing functional traits. Main conclusions: Our study highlights the importance of analysing the effects of combined practices to accurately reflect real-world farming contexts. While abundance metrics are common, reflecting several biodiversity facets and adopting a more balanced research approach across taxa are crucial for understanding biodiversity responses to agricultural changes and informing conservation strategies. Given the unbalanced evidence on impacts of agricultural practices on biodiversity, caution is required when utilising meta-analytical findings for informing public policies or integrating them into global assessment models like life-cycle assessments or global flux models. ### Competing Interest Statement The authors have declared no competing interest.