Agroforestry systems have been recognised as a possible refuge for biodiversity especially when bordering intact landscapes. The intensification of crop management to increase yields is usually associated with a reduction of shade trees and heavy use of chemicals, typically correlated with a decrease in biodiversity. The relationship between intensity of crop management and biodiversity, however, is not clear-cut and is dependent on environmental and geographical differences. We assessed the influence of different shade cover, shade tree richness, richness of other crops, distance from the forest, and use of chemicals on the diversity, richness and abundance of butterflies, a bioindicator in coffee home gardens. We collected data in 42 coffee home gardens in West Java, Indonesia, via Pollard transects, totalling 15.1 km (July-August 2019 and July-August 2020). We found 54 species of butterflies in the gardens. Via Generalised Additive Mixed Models, we found that the use of chemicals negatively influenced the abundance (p = 0.001) and richness (p = 0.039) of butterflies, while shade tree richness positively influenced the abundance (p < 0.001), diversity (p = 0.046) and richness (p < 0.001) of butterflies. The other predictors did not have a significant effect. The high diversity of butterflies in the study area suggests that the agroforestry environment is now resilient, but the relationship between butterfly abundance, diversity, and richness with shade tree richness indicates an urgency to maintain and improve current ecosystem complexity. The negative relationship between butterfly abundance and richness and the use of chemicals further indicates that organic farming should be promoted to preserve ecosystem services provided by pollinators. Coffee production in Indonesia has dramatically increased in the last 10 years and producers are keener to use more intensive farming techniques with a consequent reduction of ecosystem complexity. This process can break the resilience of agroforestry habitats if actions are not taken immediately.
Hedger, KatherineAhmad, NabilAdinda, EstherNijman, Vincent
Budiadi, BudiadiImron, Muhammad AliNekaris, K.A.I.
Department of Social Sciences
Year of publication: 2021Date of RADAR deposit: 2021-10-12