The assembly and importance of a novel ecosystem: The ant community of coffee farms in Puerto Rico

Abstract

Agricultural ecosystems are by their very nature novel and by definition the more general biodiversity associated with them must likewise constitute a novel community. Here, we examine the community of arboreally foraging ants in the coffee agroecosystem of Puerto Rico. We surveyed 20 coffee plants in 25 farms three times in a period of one year. We also conducted a more spatially explicit sampling in two of the farms and conducted a species interaction study between the two most abundant species, Wasmannia auropunctata and Solenopsis invicta, in the laboratory. We find that the majority of the most common species are well-known invasive ants and that there is a highly variable pattern of dominance that varies considerably over the main coffee producing region of Puerto Rico, suggesting an unusual modality of community structure. The distribution pattern of the two most common species, W. auropunctata and S. invicta, suggests strong competitive exclusion. However, they also have opposite relationships with the percent of shade cover, with W. auropunctata showing a positive relationship with shade, while S. invicta has a negative relationship. The spatial distribution of these two dominant species in the two more intensively studied farms suggests that young colonies of S. invicta can displace W. auropunctata. Laboratory experiments confirm this. In addition to the elaboration of the nature and extent of this novel ant community, we speculate on the possibilities of its active inclusion as part of a biological control system dealing with several coffee pests, including one of the ants itself, W. auropunctata.

Keywords: Solenopsis invicta; Wasmannia auropunctata; biological control; coexistence; community structure; dominance; invasive species.

Figure 1

Positions of the 25 farms in the sample. Image from Google Earth

Figure 2

Power function relationship between species abundance and species rank. Complete collection consists of all point with a subset of the intermediate ranks in red. The four most abundance species seem somewhat out of the general pattern for the intermediate ones (in red with the shallower slope)

Figure 3

Abundance at a site of the two most common species by visit (black closed circles) or by average of three visits at a site (open red circles). Note the strong tendency of one or the other being dominant, with only four visits exhibiting more than 20 (out of 100) bait occupancies of both species

Figure 4

Relationship between the two most common species and average percent of shade cover

Figure 5

A 12‐month record of the spatial dynamics of two of the most dominant species on farm UTUA 2 (grid is 5 × 5 m²). Size of the symbol proportional to activity at that site (largest = 5 baits occupied, smallest = 1 bait occupied, small dots are bushes that had no individuals of the three species on any of the five baits). Dashed outline square in January 2020(dry) frame indicates the position of area baited with ground baits and presented on the right. All ground baits with S. invicta, where apparently from young colonies

Figure 6

A 12‐month record of the spatial dynamics of two of the most dominant species on farm UTUA 20

Figure 7

Diagrammatic picture of the hypothetical spatio/temporal competitive process between S. invicta and W. auropunctata. Dashed connections indicate the negative effect that S. invicta colonies are hypothesized to have on the establishment and/or survival of W. auropunctata nests