Measuring competition coefficients in an ant community: Implications for intraspecific adaptation load
Data files
Oct 13, 2025 version files 175.80 KB
-
File_1-6_in_Uematsu_et_al..zip
169.65 KB
-
README.md
6.15 KB
Abstract
Understanding the stable coexistence of species despite resource competition has been a central topic in ecology. Ant communities are particularly enigmatic as various species coexist despite resource overlap. Community ecology theory predicts stable species coexistence when intraspecific competition is stronger than interspecific competition, but due to their perennial and underground life, competition coefficients of ants have never been rigorously measured in the field. We tackled this problem by studying Diacamma cf. indicum, which allows for non-invasive mark–recapture of whole colonies. Several ant species coexisted at the study site where Diacamma was most abundant, and baiting experiments and stable isotope analyses suggested overlapping food niches. Consistently, per-worker brood production of Diacamma colonies was significantly negatively correlated with con- and heterospecific worker densities within the foraging area, suggesting exploitative competition among the ants. In terms of net population growth, however, the estimated intraspecific competition coefficient was about five times larger than the interspecific competition coefficient. This is possibly because exploitative competition for food occurs both intra- and interspecifically, whereas interference competition occurs mostly among conspecifics. Indeed, for Diacamma worker survival, there was a significant (non-linear) negative correlation only with the density of conspecific colonies within the foraging area. This is consistent with the observation that Diacamma rarely fought with other species, although it violently attacked conspecific aliens encountered in their nest vicinity. We interpreted these results in light of the recent theory of intraspecific adaptation load. This theory predicts that density-dependent adaptation to intraspecific conflict can intensify intraspecific competition and act to suppress per-capita population growth in dominant species, thereby leading to species coexistence with overlapping resources. Our inclusive fitness model suggests that the intraspecific territorial aggression in Diacamma may be a counter-adaptation to intraspecific conflict, i.e., brood abduction between conspecific colonies. This aggression pattern can cause the observed density-dependent worker mortality. Our population dynamic model indicates that such density-dependent excess mortality acting on dominant competitors can promote stable coexistence with subordinate competitors. Overall, our results support the intraspecific adaptation load theory that aims at integrating behavioral and community ecology to understand how adaptation interacts with population and community dynamics.
https://doi.org/10.5061/dryad.8pk0p2nwk
File: File_1-6_in_Uematsu_et_al..zip
Here, you can access all the data used in this study.
To measure the intra- and interspecific competition coefficients of Diacamma cf. indicum in the field, we examined the relationship between the intra-colony demography of Diacamma and the local ant community structure using the following methods.
Intact colonies of Diacamma were collected using the bamboo-traps. After carefully examining their colony composition (number of workers, brood weight, etc.), all workers were individually marked, and the colonies (nested in traps) were released intact at random locations in the study site. After one week, those colonies were recaptured, and any changes in colony composition were examined. In addition, local ant communities within a 5-m radius of the release point (the usual foraging distance of Diacamma workers) were examined by pitfall trapping and visual observation.
The data were collected in Excel files and analyzed with R scripts.
Description of the data and file structure
The data consist of the following six Excel files and one PDF file (for Zenodo).
File 1. Worker survival of Diacamma in relation to the local ant communities.
Each row indicates the data for each individual worker.
Column A indicates the ID of the focal worker.
Column B indicates the colony to which each worker belongs.
Column C: colony size: W1 indicates the number of workers in the focal colony before release into the field.
Columns D and E indicate the year and month of the experiment.
Column F indicates the presence or absence (death) of the focal individual.
Column G indicates the local density of Diacamma colony at the release site of the focal colony.
Columns H to O indicate the local density of each ant species at the release location (within a 5-m radius of the release point) of the focal colony (Each species density is calculated by multiplying the number of individuals of each species collected in pitfall traps by the mean per worker fresh weight of each species. See File 5).
File 2. Colony demographic parameters of Diacamma (brood production, colony growth and etc.) in relation to the local ant communities.
Each row indicates the data for each colony.
Column A indicates the ID of the focal colony.
Column B indicates the month of the experiment.
Column C: W2 indicates the number of marked workers after recapture (workers that emerged after release were not included).
Column D: W3 indicates the total number of workers after recapture (including workers that emerged after release).
Column E: B1 indicates the total brood weight (the sum of fresh weight of eggs and larvae; we excluded cocoons from B1 because they are destined to emerge) before release.
Column F: B2 indicates the total brood weight after recapture (the sum of the fresh weights of eggs, larvae, and cocoons; we included newly pupated cocoons here, because they reflect larval growth).
Column G: B3 indicates the total brood weight before release (the sum of eggs, larvae, and cocoons).
Column H: B4 indicates the total brood weight after recapture (the sum of eggs, larvae, and all cocoons).
Column I indicates the brood production (B2 - B1) of the focal colony.
Column J indicates the absolute colony growth, where M in the formula means the average fresh weight of an individual adult of the focal colony.
Column K indicates the net colony growth per worker.
Column L: colony size: W1 indicates the number of workers in the focal colony before release into the field.
Column M indicates the local density of Diacamma colony at the release location (within a 5-m radius of the release point) of the focal colony.
Columns N to AB indicate the local density of each ant species at the release location of the focal colony (Each species density is calculated by multiplying the number of individuals of each species collected in pitfall traps by the mean per worker fresh weight of each species. See File 5).
File 3. Results of the stable Carbon isotope analysis of seven dominant ants in the study site.
Columns indicate ant species and rows indicate colony-based data in each species. Measurements are reported in delta notation (δ): δ13C = [(Rsample/Rstandard) − 1] ∙ 1000, where R is the ratio of the heavy/light isotope content (13C/12C). Isotope ratios are expressed in per mil (‰) relative to international reference standards VPDB (Vienna PeeDee Belemnite) for carbon.
File 4. Results of the stable Nitrogen isotope analysis of seven dominant ants in the study site.
Columns indicate ant species and rows indicate colony-based data in each species. Measurements are reported in delta notation (δ): δ15N = [(Rsample/Rstandard) − 1] ∙ 1000, where R is the ratio of the heavy/light isotope content (15N/14N). Isotope ratios are expressed in per mil (‰) relative to international reference standards VPDB (Vienna PeeDee Belemnite) for atmospheric nitrogen for nitrogen.
File 5. Mean per worker wet weight of the seven dominant ant species in the study site.
Column A indicates an ant species.
Column B indicates the average fresh weight for each species. 20 workers per species were collected from two to four colonies. They were weighed to the nearest 0.001 mg, and the average wet weight was calculated.
File 6. The number of individuals of each ant species collected in pitfall traps in each capture-release-recapture trial.
Column A indicates an ant species.
Columns B to Y indicate the number of individuals of each ant species collected in pitfall traps for each experiment (N=24).
File 7. Mathematical model of population dynamics.
We found that the worker mortality of Diacamma cf. indicum ants increased as the local density of conspecific colonies increased. We analyzed the effect of this density-dependent mortality on species coexistence using this Mathematica code (for Zenodo).
- Uematsu, Jumpei; Yamamichi, Masato; Tsuji, Kazuki (2025). Measuring competition coefficients in an ant community: Implications for intraspecific adaptation load. Ecology. https://doi.org/10.1002/ecy.70274
