Mapping and assessment of pollination services on crops of Citrullus lanatus and Physalis ixocarpa in the coastal region of Jalisco, Mexico

Oliverio Delgado-Carrillo, Escuela Nacional de Estudios Superiores and Instituto de Investigaciones en Ecosistemas y Sustentabilidad, Universidad Nacional Autonoma de Mexico, Morelia, MH, Mexico, Jacob Cristobal-Perez, LABORATORIO NACIONAL DE ANALISIS Y SINTESIS ECOLOGICA, Universidad Nacional Autonoma de Mexico, MORELIA, MH, Mexico, Mauricio Quesada, Laboratorio Nacional de Analisis y Sıntesis Ecologica, Escuela Nacional de Estudios Superiores and Instituto de Investigaciones en Ecosistemas y Sustentabilidad, Universidad Nacional Autonoma de Mexico, Morelia, Mexico, Adrian Ghilardi, Universidad Nacional Autónoma de México, Laboratorio Nacional de Análisis y Síntesis Ecológica (LANASE), Morelia, Mexico; Universidad Nacional Autónoma de México, Centro de Investigaciones en Geografía Ambiental (CIGA), Morelia, Mexico and Diana Ramírez-Mejía, Universidad Nacional Autónoma de México, Laboratorio Nacional de Análisis y Síntesis Ecológica (LANASE), Morelia, MH, Mexico

Background/Question/Methods: Bees are one of the main groups of organisms responsible for carrying out the pollination service for crops used for human consumption. However, a decline in bee populations has been reported globally due to several biotic and abiotic factors. Habitat loss and fragmentation due to land-use change and land-use intensification are among the main factors driving bee declines. Here we evaluate the pollination services on crops of Citrullus lanatus (watermelon) and Physalis ixocarpa (green tomato) in the Tropical Dry Forest of Jalisco, Mexico. We assess how landscape configuration influences the abundance and richness of P. ixocarpa and C. lanatus pollinators. We selected C. lanatus and P. ixocarpa crops, as they are pollinator-dependent annual crops. We sampled several plots of C. lanatus and P. ixocarpa in 2008 and from 2014 to 2017 and recorded the floral visitors to each flower and the fruit set per sampled plant. To evaluate the influence of landscape configuration on the pollinator community, we mapped land use and land cover for 2007, 2012 and 2017. We generated circular buffer areas in a radius of up to 2000 km around each sampled plot and calculated the percentage of forest cover and distance to forests. We performed linear regressions to evaluate the relationships between the abundance of bees and forest cover and distance to forests. We use the INVEST pollinator model and compared it with the abundance obtained in the field. Results/Conclusions: For year 2008, we found a positive relationship between the abundance of bees and the cover and distance of the forest. We also found a positive relationship between the results of the INVEST model with the abundance we observed in the field. We obtained the per-pixel total abundance of 12 pollinators species for three years. We found a significant positive regression between the abundance predicted by the model and the real abundance for the year 2008 and with the observed abundance of pollinators and observed abundance of social bees. We did not find a strong regression with the predicted abundance and real abundance for 2014 and 2016. The results indicate that the conservation of natural areas is important for the maintenance of pollinators and the assurance of food safety.