Bee Diversity (Hymenoptera: Apoidea) in a Tropical Rainforest Succession

• Main Authors: Allan Smith-Pardo, Victor H. Gonzalez
• Format: Article
• Language: English
• Published: Universidad Nacional de Colombia 2007-01-01
• Series: Acta Biológica Colombiana
• Subjects: Apoidea; Sucesión; Inventarios; Colombia
• Online Access: https://revistas.unal.edu.co/index.php/actabiol/article/view/27170
• ISSN: 0120-548X; 1900-1649

The purposes of this work were to study the bee diversity across successional stages in a Neotropical rainforest and to test four sampling methods. Bee diversity and richness were studied in northeastern Medellín (Antioquia, Colombia; 900-1100 m in altitude) over six months using a chronosequence with four successional stages (abandoned pastures, low shrubs, secondary forest, and mature forest). Bees were intensively and systematically sampled, using the following methods: entomological net, Malaise trap, Methyl Salicylate, and Van Somer trap with decomposed fresh water fish. A total of 15.356 specimens were collected, belonging to four families and 287 species, representing 62% of all bee species found in Colombia. About 50% of all individuals sampled were stingless social bees (Apidae, Meliponini). Trigona (Trigona) fulviventris was the most abundant species (~10%) in the survey. Augochlora and Megachile were the most specious genera. The pasture and secondary forest showed high values of diversity and richness and were significantly higher than those of the mature forest and low shrubs. In all successional stages, except in the mature forest, the number of new species collected in each sample period approached zero and the species accumulation curves tended to stabilize as time and sampling area increased. The net was the most efficient method in all successional stages, except in the forest, where most bee species and individuals were collected with the Van Somer trap. However, a higher percentage (50%) of rare species was collected with the Malaise trap. The number of new species collected in each sampled period and the species accumulation curves suggest that our survey was nearly sufficient to estimate the bee diversity in these early successional stages, but insufficient to study the mature forest apifauna. Due to the high efficiency of the Van Somer trap to attract bees in the forest, this trap should be used regularly in additional bee surveys in tropical rain forests. We also summarize the bee surveys in Colombia and highlight the importance of using other less common sampling methods to study bees from tropical ecosystems.