Global meta-analysis shows reduced quality of food crops under inadequate animal pollination

Animal pollination is vital for the reproduction of many flowering plants, including a large proportion of cultivated food crops. While some staple crops rely on wind or self-pollination, numerous animal-pollinated crops are significant sources of micronutrients contributing to diverse and healthy diets. However, recent decades have witnessed a decline in pollinator diversity and abundance due to various anthropogenic pressures. This decline, combined with the global expansion of pollination-dependent crops, raises concerns about potential pollination deficits worldwide. These deficits could negatively affect crop yields, agricultural production stability, and ultimately, human health. Much research has focused on quantifying the effect of pollinators on crop yields, but a comprehensive global assessment of their influence on food quality remains lacking. Food quality is a multifaceted concept encompassing sensory characteristics (appearance, taste) and nutritional profile. These traits influence food safety perception, supply chain standards, and consumer behavior. Therefore, quantifying the contribution of animal pollination to food quality and marketability is critical. This study aims to provide a global quantitative review of animal pollination's effects on various aspects of food quality, using a systematic literature review and meta-analyses to estimate the impact on 48 globally important crops.

Existing research has predominantly focused on the effects of pollination on yield-related measures such as fruit set and seed weight. Studies have demonstrated the crucial role of managed and wild pollinators in enhancing crop yield and its stability. However, a comprehensive global assessment of pollination's impact on various aspects of food quality has been lacking until now. This study fills this gap by synthesizing a large body of empirical research on the effects of pollination on multiple quality traits across various crops.

This study employed a systematic literature review and meta-analysis approach. Two separate literature searches were conducted in Scopus, ISI Web of Science, and Google Scholar, using keywords related to crops, pollinators, and quality. The searches encompassed peer-reviewed articles, conference proceedings, book chapters, and theses. Studies were included if they met specific criteria: (1) use of manipulative experiments; (2) measurement of at least one quality trait; (3) inclusion of at least two pollination treatments (open pollination, hand pollination, and pollinator exclusion); and (4) fruit production under all treatments. The pollination effect was quantified using two metrics: pollination service (difference between open pollination and pollinator exclusion) and pollination deficit (difference between open pollination and hand pollination). The natural logarithm of the response ratio (lnRR) was used as a measure of effect size. Multi-level meta-analyses were performed to account for the non-independence of observations (due to multiple trials within studies and multiple quality traits measured). Various moderators were considered, including quality trait, pollinator group, crop type, experimental scale, cropping environment, and climate. Publication bias was assessed using various methods, including funnel plots and Egger's regression. Sensitivity analyses were conducted to evaluate the robustness of results.

The meta-analysis included 1197 effect sizes from 153 publications on pollination service and 682 effect sizes from 86 publications on pollination deficit. Animal pollination significantly improved overall crop quality by 23% (95% CI = 16%-30%; p<0.001). This positive effect varied among quality traits, with larger improvements in fruit organoleptic traits (27%, CI=20%-34%) compared to nutritional traits (7%, CI=1%-14%). Pollinators mainly improved size, shape, commercial grade, and shelf life, with less pronounced effects on firmness and micronutrients. The positive effects were consistent across different pollinator groups, crop types (fruits and vegetables), experimental scales, cropping environments, and climatic regions. Analysis of pollination deficit (comparing open pollination with hand pollination) showed no overall deficit, but weak positive effects of hand pollination on organoleptic traits (size, p=0.054), and negative effects on nutritional traits (macronutrients, p=0.045). The analysis revealed considerable heterogeneity in effect sizes, and some key limitations were identified.

This global meta-analysis provides strong evidence for the significant contribution of animal pollination to the quality of many food crops, particularly in enhancing commercially important traits that affect market value and reduce post-harvest losses. The weak signals of pollination deficits for some traits, although statistically insignificant, suggest the potential for service decline in some agricultural landscapes. The lack of a significant overall deficit, despite pollinator declines in many areas, may be due to the common practice of supplementing pollinator communities with managed pollinators such as honeybees. The findings highlight the economic and environmental importance of maintaining optimal pollination services, especially considering the impact of suboptimal quality on market prices and food waste.

This study emphasizes the significant role of animal pollination in ensuring high-quality food production. While an overall pollination deficit wasn't observed, subtle negative effects on some traits and the observed benefits on marketable traits highlight the importance of pollinator conservation. Further research is needed to understand the effects on crops not sufficiently represented, like spices and stimulant crops, as well as the potential interaction between pollination and other biotic and abiotic factors. Integrating yield and quality benefits in economic valuations of pollination services is also crucial for effective conservation strategies.

Despite the large dataset, certain limitations exist. The study lacked sufficient data on some crop types (spices, stimulants). The analysis couldn't fully assess the impact of cultivar effects due to limited replication within studies. Furthermore, the data on vertebrate pollinators was insufficient to analyze their specific effects separately. Interactions between pollination and other biotic and abiotic factors are also largely unexplored. Finally, the subjective nature of some quality traits (e.g., taste) introduces uncertainty.