ANT AND SPIDER INTERACTIONS IN COFFEE AGROECOSYSTEMS: PATTERNS AND MECHANISMS

Abstract

Both abiotic and biotic factors affect the abundance and distribution of organisms and therefore impact ecosystems and agroecosystems. Indeed, some organisms have strong effects on the distribution of others because they act as keystone species and/or ecosystem engineers whose effects cascade to other trophic levels. Ants can affect the distribution and abundance of a variety of taxonomic groups. In coffee agroecosystems from Southern Mexico Azteca sericeasur ants actively patrol the tree where their nest is built and the nearby coffee plants influencing the ant community and biological control. Here we report the relationship that A.sericeasur has with spiders, an abundant predatory group found in coffee farms. We present evidence of the relationships between A. sericeasur and spider communities found at the coffee layer and at the tree trunk layer; and we suggest potential mechanisms driving the observed patterns. During the summers of 2008 and 2009 we sampled spiders in the coffee layer and we found that spiders per coffee bush were richer and more abundant in the presence of A. sericeasur that in its absence. Indeed, in the presence of these ants, spider richness increased by 27% whereas spider abundance increased by 67 %. Although, spider richness and abundance per coffee plant increased in the presence of A. sericeasur, spider species composition did not change between plants with and without the ants. In addition, we found that insect abundance per coffee plant increased in the presence of the ants. Based on this evidence we propose resource availability and enemy free space as driving mechanisms In regard to the spider community found at the trunk of shade trees in 2011 and 2012 we found that A. sericeasur was positively correlated with spider abundance and biomass. Indeed, spider abundance increased by 57.5% in the presence of the ants whereas spider biomass was also strongly correlated with their presence. However, we also identified that excluding the two most abundant spiders (Ischnothele digitata and Azilia guatemalensis), strongly changed the observed patterns. Fieldwork observations supported that both I. digitata and A. guatemalensis are predators of A. sericeasur. Likewise the coffee layer, spider species composition in the tree trunk layer was not strongly affected by the presence of the ant. Finally, we report the association between Falconina sp and A. sericeasur; a swift spider that travels in and out of A. sericeasur nests. Our data suggest that Falconina is a scavenger that takes advantage of dead A. sericeasur individuals. Overall we show that through resource availability, enemy free space and predation, A. sericeasur ants positively drive spider communities found at the coffee and at the tree trunk layers of coffee agroecosystems. (CONACYT, University of Michigan-Rackham School and School of Natural Resources and Environment, COCYTECH and NSF-DEB-1262086 granted to S. Philpott).