Honey bee (Apis mellifera) foraging ecology in coffee landscapes and description of “coffee garden honey”

Abstract

When coffee is grown under shade trees, as in Mexico and Guatemala, coffee landscapes are complex and offer a high diversity of plants to the bees. This study focuses on the manner in which bees adapt their foraging behavior as a function of variation in tree coverage on a landscape scale. Study sites were characterized by satellite images and categorized according to percentage of tree cover: low (50–62%), medium (63–77%), and high (77–83%). Honey samples were collected from several apiaries in each landscape. In order to identify resources used by Apis mellifera, samples were analyzed according to melissopalynology. A random forest model analysis performed to compare different honeys did not show significant differences between honeys in any given landscape. Nevertheless, percentage of tree cover was positively correlated, in a significant manner, with the relative proportion of pollen from trees and from shrubs present in honey samples. These analyses, along with physical–chemical and sensory analyses were used to develop a first standard of reference for coffee garden honey.

Cuando el café crece bajo árboles de sombra, como en México y Guatemala, estos paisajes cafetaleros son complejos y ofrecen una alta diversidad de plantas para las abejas. Este estudio se enfocó en la manera en la cual las abejas adaptan su forrajeo en función de la variación en la cobertura arbórea a escala de paisaje. Los sitios de estudio fueron caracterizados por imágenes de satélite y categorizados de acuerdo al porcentaje de cobertura arbórea: baja (50–62%), media (63–77%) y alta (77–83%). Las muestras de miel se colectaron de varios apiarios en cada paisaje. Para identificar los recursos utilizados por Apis mellifera, las muestras fueron analizadas por medio de melisopalinología. La comparación estadística entre las mieles se hizo por medio del modelo Random Forest, en el cual no se encontraron diferencias significativas entre las mieles en ninguno de los paisajes. Sin embargo, el porcentaje de cobertura arbórea fue correlacionado positivamente con la proporción de polen de árboles y arbustos presente en las muestras de miel. Estos análisis, junto con los físico-químicos y sensoriales fueron usados para desarrollar el primer estándar de referencia para la miel de cafetal.

Keywords:

• Apis mellifera
• Coffea arabica
• coffee garden
• agroecosystem
• honey characterization

Acknowledgements

We thank those responsible for these projects and the beekeepers’ organizations of Mexico and Guatemala that actively participated in this study (Mieles del Sur and Guaya’b).

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This study was made possible by three grants: the Belgian Technical Cooperation project “Characterization of honey or the generation of new niches for fair trade beekeepers”, the Mexico-European Union project Foncicyt 94293 “Mutual” – “Mutualisms and bees in tropical landscapes: risks and recovery for biodiversity and agriculture”, and the project Sep-Conacyt 106043 “Land use effect on bee’s biodiversity conservation”.