Wild bees are key providers of pollination services in agroecosystems. The abundance of these pollinators, and the service they provide, relies on the availability of supporting resources in the landscape. Because of this, spatially explicit models have been developed to quantify wild bee abundance and pollination services in food crops, while accounting for the influence of locally available foraging and nesting resources. However, model implementation is limited by the availability of land cover maps and experts on pollinators capable of establishing the quality of local habitats for pollinators. In this study, we present how remote sensing data can be linked to the estimation of wild bee abundance, and act as an alternative to the conventional use of land cover maps and local expertise in spatially explicit models estimating pollination services. For this, we used landscape characteristics derived from remote sensors to qualify nesting resources in the landscape and thereafter estimate the delivery of pollination services by mining bees (Andrena spp.) in 30 fruit orchards located in the Flemish region of Belgium. Mining bees were selected for this study for their major role as local pollinators and underground nesting habits. Estimated pollination services were compared with those derived from conventional qualifications of nesting resources and showed no significant differences (P=0.68) in the amount of explained variation in activity of mining bees on the studied orchards. Estimates derived from remote sensing data and conventional inputs explaining 69% and 72% of the total variation, respectively. These results confirmed that remote sensing data can deliver nesting suitability characterizations suitable for the estimation of pollination services, This research also illustrates the relevance of nesting resources and highlights the importance of considering soil resources in the estimation of pollination services provided by pollinators like mining bees. Our results support the development of holistic agro-environmental policies that rely on the use of modern tools like remote sensors and promote pollinators by considering nesting resources.

Data was collected in 30 fruit orchards (Malus domestica, n=15 and Prunus avium, n=15) in the Flemish region of Belgium. Pollinators were surveyed on each orchard during peak crop bloom (April 10th-20th) in 2019, for 1h by two samplers with nets. Each sampler followed a between-row transect that started on opposite sides of the orchard. Surveys were done only from 11h00 until 17h00 with environmental temperatures ranging between 12-24°C and clear skies. Sampled pollinators and collected data were processed in the laboratory of agrozoology at the faculty of Bioscience Engineering in Ghent University.

For privacy reasons, the coordinates of each surveyed field have not been provided.