Food production is challenged by changes in climate and land use and expanding human population growth. Proper pollination can increase the production and quality of fruit, nut, oil, and fiber crops. We reviewed crop dependence on pollinators and estimated the economic value of pollination per year for each crop in Brazil. We analyzed 141 crops and found that 85 depend on pollinators. Almost one-third of these crops have an essential or great dependence on pollinators. The economic contribution of pollinators totals ∼30% (∼US$12 billion) of the total annual agricultural income of the dependent crops (totalizing almost US$45 billion). However, half of these figures includes soybean crop (US$5.7 billion of pollinators' contribution and US$22 billion of annual income). Soybean was considered as having a modest dependence on pollinators, but this remains controversial because pollinator dependence might vary among different varieties cultivated in Brazil. Moreover, we have no information about pollinator dependence regarding some important crops, showing the urgent need for basic research into reproductive biology and pollination ecology.

Biotic interactions have been considered as an important feature in species distribution modeling, but little is known about the effect of including different types of interactions or performing different strategies of integrating biotic interactions in species distribution modelling. This study compares different methods for including interspecific interactions in species distribution models for bees, their cleptoparasites, and the plants they pollinate. Hostparasite interactions among bumble bees (genus Bombus: generalist pollinators and brood parasites) and specialist plant-pollinator interactions between Centris bees and Krameria flowers were used as case studies. We used 7 different modelling algorithms available in the BIOMOD R package. Adding biotic information to present day predictions of potential occurrence significantly improved the cross-validated area under the receiver operating characteristic curve (AUC), a measure often applied to estimate model accuracy. Different species and types of interaction showed different AUC results in line with data quality, level of biological linkage and interdependence of each interaction. Furthermore, the species that presented the best improvement of AUC was projected under future climate scenarios. The results showed marked differences when using abiotic data only, and when including biotic interactions of various types. The results show that choosing the correct biotic information, based on biological and ecological knowledge, is critical to improve the accuracy of species distribution models, and also to forecast distribution change.

No abstract