Background Globally, there are estimated to be 25,000 -30,000 ant species, ~12,500 of which are described. Ants are ubiquitous in nearly every terrestrial habitat and can be found on every continent except Antarctica. At any one time, there might be 1 × 1016 living ants, whose collective biomass exceeds that of all humans (Hölldobler and Wilson 1990). They can be sampled accurately using rapid and efficient sampling techniques (Agosti et al. 2000, Gotelli et al. 2011). At the genus level, well-resolved phylogenies are available, and taxonomy is reasonably stable, but this is not the case for the majority of species clades (Moreau and Bell 2013). Additionally, the life histories and natural histories of most dominant ant species are understood (Bourke and Franks 1995, Hölldobler and Wilson 1990), and the behavioral responses of ants to resources can also be assessed efficiently (Kaspari et al. 2008, Lessard et al. 2011). Furthermore, there is an extensive literature on biotic and abiotic factors that influence ant community structure (Cole 1983; Fellers 1987, 1989; Hölldobler and Wilson 1990; Andersen 1992, 1997; Cerdá et al. 1997, 1998; Suarez et al. 1998; Holway 1999; Kaspari et al. 2000, Gotelli and Arnett 2000; Sanders and Gordon 2003, Sanders et al. 2003a, Sanders et al. 2003b) and spatial variation in the diversity of ants (Kusnezov 1957, Ward 2000, Kaspari et al. 2003, Sanders et al. 2007a, Sanders et al. 2007b, Dunn et al. 2009a, 2009b). Finally, ants function as important predators, competitors, seed dispersers, mutualists, and terrestrial ecosystem engineers. For example, 35% of all herbaceous species have their seeds dispersed by ants (Beattie 1985, Lengyel et al. 2009, Lengyel et al. 2010), and ants are responsible for as much as 99% of all seed dispersal events (Zelikova et al. 2009). Ants also modify ecosystem processes through soil turnover and litter processing, which affect decomposition and mineralization (e.g., Folgarait 1998, Terborgh et al. 2001, Wardle et al. 2013). Aims We intend to deploy 40 terrestrial mesocosms to investigate how variation in the density of ecosystem engineers (ants) influences the structure of soil food webs and the functioning of soil ecosystems. This experimental design also allows us to test for density-dependent population growth in ants, something that has been surprisingly difficult to test (Gordon et al. 2022). The first summer (summer 2023) will be used to set up the mesocosms. Measuring soil food web composition and decomposition rates require for the mesocosms to be colonized by soil communities, so we will conduct the first sampling round at the end of 2023 to establish baseline conditions. Therefore, we will need to wait until summer 2024 to gather all of the data. Materials and methods Establishment of mesocosms We will establish the experiment at UMBS along the east side of the Pine Point Trail, approximately 5-10 m from the trail. The density of ant nests will be manipulated in terrestrial mesocosms following inoculation with soil and leaf litter. Soil and leaf litter will be taken from beneath the mesoscosms as we stabilize them. The mesocosms are cattle tanks (186 cm in diameter, 61 cm in depth, 184.05 L in volume), which are typically used in aquatic ecology. To facilitate drainage, we will drill 20-40 holes in the bottom of each tank. Each hole will be a circle with a diameter of 2cm. To minimize organisms escaping the mesocosms, we will glue and staple a layer of mesh to cover holes at the bottom of the mesocosms. We will also apply a 7.5 cm wide and 0.15 cm thick coat of Tanglefoot® on the rim of the mesocosms to prevent ants from emigrating or colonizing the cattle tanks. The experiment will be set up as a block design, where each block comprises all 4 treatments (see manipulation of nest density below). Each block will be replicated 10 times (n = 10) for a total of 40 mesocosms. Within a block, center points of adjacent mesocosms will be 3 m apart, meaning the edges of adjacent mesocosm will be 114 cm apart. This will allow for the different mesocosms in each block to be under similar conditions, while being able to easily access them. Blocks will be at least 50 m apart to account for spatial heterogeneity in our study site (Dutilleul, 1993). Each block will be 15 m away from a dirt path, to minimize the trampling effect on soil communities (e.g., Grandchamp et al., 2000), and to allow easy access to the mesocosms. Mesocosms will be inoculated with soils during the month of June, as soil fauna activity is highest then (Wolda, 1988). Soils in the area of the experiment are well-drained, with a 1-3 cm deep AO horizon and a 10-15 cm E horizon on top of a B horizon (Nave et al., 2013). To mimic the soil natural physiological chemistry of the area, we will first add 3 cm of sand to improve drainage of the mesocosms. The sand will come from a local UMBS sand provider. On top, we will add 30 cm of a BE horizon. We will not add separate B and E horizons since it will be very hard to distinguish between both layers in the field. On top, we will add 3 cm of an O horizon. The soil for the BE and O horizons will come from the forest where the mesocosms will be placed. This will allow for the mesocosms to be inoculated with organisms, reproducing natural soil communities. We will add the soil in the mesocosms under the same meteorological conditions to ensure that we inoculate all the mesocosms with similar soil communities (Villani & Wright, 1990). We will set the mesocosms up under sunny conditions, when soil fauna activity is highest (Villani & Wright, 1990), allowing us to inoculate the mesocosms with more organisms. We will also allow for litter to fall on the mesocosms from the trees above the mesocosms, enhancing the inoculation of the mesocosms. Manipulation of ant nest density Artificial nests have been dispatched on the forest floor of pine wood forest since 2021. Most of those have been observed to be colonized by Aphaenogaster picea, a numerically dominant ant species in northern temperate forests. In our experiment, we will dispatch artificial nests until we have at least 100 colonized nests. We consider colonized nests once we observe a queen and workers inside the nest. Then, we will relocate the colonized nests to the mesocosms following the description below. The colonies will be collected from the area of the study; we are not altering gene flow between populations at UMBS. Aphaenogaster picea natural nest abundance ranges from 0.5 to 1.3 nests / m2 (Lubertazzi, 2012). As such, we will manipulate nest densities in mesocosms to range from below and to above natural nest densities. To standardize the experimental design, each mesocosm will contain a total of 6 artificial nest boxes, but the proportion of nests boxes occupied by a colony will vary. This will allow for A. picea colonies to migrate, as Aphaenogaster colonies are known to migrate multiple times during the summer season (Lubertazzi, 2012; Smallwood, 1982). We will inoculate mesocosms with nest boxes containing 0 colonies, 1 colony (0.17 occupancy, 0.37 nests / m2), 3 colonies (0.50 occupancy, 1.1 nests / m2) or 6 colonies (full occupancy, 2.2 nests / m2). We will relocate similar-sized colonies in mesocosms with multiple colonized nest boxes, to minimize inter-colony variation. Each occupancy level will be replicated 10 times for a total of 40 mesocosms (100 colonized artificial nests, 240 total artificial nests in total). Tracking soil food web changes We will assess changes in the soil food web of ground-dwelling invertebrates using two different techniques. We will use soil core extraction to sample the mesofaunal community. We will obtain 3 cores of soil (5 cm diameter, 9 cm deep) in each mesocosm. The corers will be made using a PVC soil corer. We will place each soil core inside a Berlese funnel. The Berlese funnel will consist of a 5 stainless steel mesh at the bottom of a plastic funnel (20 cm top diameter, 20 cm deep). We will place a 25-W incandescent light bulb 5 cm on top of each funnel as a heat source. Below the Berlese funnel, we will place a 500 mL container filled with 100 mL of ethanol 70% to collect the organisms for identification. The extraction will last 72h (Rochefort et al., 2006). Since the organisms need to be alive for the extraction (Rousseau et al., 2018), we will collect the soil cores and extract organisms one block at a time (9 cores), starting in July, to maximize soil fauna activity (Wolda, 1988). We will identify all organisms captured and visible to the lowest possible taxonomic resolution. We will focus on ant prey, such as Collembola, Carabidae, Braconidae, Hemiptera, Isoptera or Ixodida species, and predators, such as Araneae (Basset et al., 2022; Hawes et al., 2002; Sanders & van Veen, 2011). Some species will be classified by their morphospecies (Lessard et al., 2011), but the minimum resolution we will identify to is order, as it is commonly used for trophic web studies (e.g., Potapov, 2022; Potapov et al., 2022). We will be able to assess species abundance, composition, and evenness (Bestlemeyer et al., 2000). With species composition, we will reconstruct the food web assemblages and changes in stability according to different variations in abundance, based on feeding habits and interactions between the organisms (e.g., Potapov, 2022; Potapov et al., 2022). We will then let the mesocosms mature for a year. This will allow for the soil physiological-chemical properties and soil communities to stabilize. Through the litter fall during fall and the openings in the mesh at the bottom of the mesocosms, other organisms will also be able to colonize the mesocosms throughout the year. After the maturation year, we will analyze the soil food webs of the mesocosms using soil cores with Berlese funnels. This will allow us to know how the communities have stabilized, and what the communities are before the addition of ants. A month after the inoculation with ant colonies, we will analyze the soil food webs of the mesocosms using soil cores with Berlese funnels. This will give the soil communities time to readjust to different ant abundances and will allow us to analyze how variations in abundances have affected the soil communities. Ecosystem processes To estimate litter decomposition, we will follow the standard TeaBag Decomposition protocol (Keuskamp et al. 2013). We will measure the myrmecochory of each mesocosm by analyzing seed dispersal rate and seed dispersal distance a week after the inoculation of the mesocosms with ant colonies. We will first collect elaiosome bearing seeds from Sanguinaria canadensis in nearby populations where the mesocosms will be located (UMBS, 2022). The seeds will be stored in a freezer until further use, as freezing the seeds does not affect their attractiveness (Clark & King, 2012). In each mesocosm, we will place the same number seeds on top of a 10 cm in diameter Tupkee® white filter paper to facilitate the measurements (Prior et al., 2020). We will then assess the seed dispersal rate by counting how many seeds have been moved from the filter paper (Stuble et al., 2014). We will assess the distance of seed dispersal by measuring the distance to which the seeds are dispersed (Ness, 2004). We will note these measures every 30 minutes for 2.5 hours, or until all the seeds are dispersed (Heithaus et al., 2005). We will use a total of 1200 Sanguinaria canadensis seeds. Intraspecific competition To measure density dependence, we will record colony productivity (the production of new workers) as a function of nest density. Literature cited Basset, Y., Palacios-Vargas, J. G., Donoso, D. A., Castaño-Meneses, G., Decaëns, T., Lamarre, G. P., De León, L. F., Rivera, M., García-Gómez, A., Perez, F., Bobadilla, R., Lopez, Y., Ramirez, J. A., Cruz, M. M., Galván, A. A., Mejía-Recamier, B. E., & Barrios, H. (2020). Enemy-free space and the distribution of ants, springtails and termites in the soil of one tropical rainforest. European Journal of Soil Biology, 99, 103193. https://doi.org/10.1016/j.ejsobi.2020.103193 Bestlemeyer, B. T., Agosti, D., Alonso, L. E., Brandão, C. R. F., Brown Jr., W. L., Delabie, J. H. C., & Silvestre, R. (2000). Field techniques for the study of ground-dwelling ants. In D. Agosti, J. D. Majer, L. E. Alonso, & T. R. Schultz (Eds.), Ants: Standard methods for measuring and monitoring biodiversity (pp. 122–144). Smithsonian Institution Press. Bokhorst, S., & Wardle, D. A. (2013). Microclimate within litter bags of different mesh size: Implications for the ‘arthropod effect’ on litter decomposition. Soil Biology and Biochemistry, 58, 147–152. https://doi.org/10.1016/j.soilbio.2012.12.001 Bradford, M. A., Tordoff, G. M., Eggers, T., Jones, T. H., & Newington, J. E. (2002). Microbiota, fauna, and mesh size interactions in litter decomposition. Oikos, 99(2), 317–323. https://doi.org/10.1034/j.1600-0706.2002.990212.x Clark, R. E., & King, J. R. (2012). The ant, Aphaenogaster picea, benefits from plant elaiosomes when insect prey is scarce. Environmental Entomology, 41(6), 1405–1408. https://doi.org/10.1603/EN12131 Davidson, A., & McKerrow, A. (2016). GAP/LANDFIRE National Terrestrial Ecosystems 2011 [Data set]. U.S. Geological Survey. https://doi.org/10.5066/F7ZS2TM0 Dutilleul, P. (1993). Spatial Heterogeneity and the Design of Ecological Field Experiments. Ecology, 74(6), 1646–1658. https://doi.org/10.2307/1939923 Fellers, J. H. (1987). Interference and Exploitation in a Guild of Woodland Ants. Ecology, 68(5), 1466–1478. https://doi.org/10.2307/1939230 Grandchamp, A.-C., Niemelä, J., & Kotze, J. (2000). The effects of trampling on assemblages of ground beetles (Coleoptera, Carabidae) in urban forests in Helsinki, Finland. Urban Ecosystems, 4, 321–332. Greenslade, P. J. M. (1973). Sampling ants with pitfall traps: Digging-in effects. Insectes Sociaux, 20(4), 343–353. https://doi.org/10.1007/BF02226087 Hawes, C., Stewart, A., & Evans, H. (2002). The impact of wood ants (Formica rufa) on the distribution and abundance of ground beetles (Coleoptera: Carabidae) in a Scots pine plantation. Oecologia, 131(4), 612–619. https://doi.org/10.1007/s00442-002-0916-6 Heithaus, E. R., Heithaus, P. A., & Liu, S. Y. (2005). Satiation in collection of myrmecochorous diaspores by colonies of Aphaenogaster rudis (Formicidae: Myrmicinae) in Central Ohio, USA. Journal of Insect Behavior, 18(6), 827–846. https://doi.org/10.1007/s10905-005-8743-3 Kaspari, M., O’Donnell, S., & Kercher, J. R. (2000). Energy, density, and constraints to species richness: Ant assemblages along a productivity gradient. The American Naturalist, 155(2), 280–293. Lessard, J.-P., Dunn, R. R., & Sanders, N. J. (2009). Temperature-mediated coexistence in temperate forest ant communities. Insectes Sociaux, 56(2), 149–156. https://doi.org/10.1007/s00040-009-0006-4 Lessard, J.-P., Sackett, T. E., Reynolds, W. N., Fowler, D. A., & Sanders, N. J. (2011). Determinants of the detrital arthropod community structure: The effects of temperature and resources along an environmental gradient. Oikos, 120(3), 333–343. https://doi.org/10.1111/j.1600-0706.2010.18772.x Lubertazzi, D. (2012). The biology and natural history of Aphaenogaster rudis. Psyche: A Journal of Entomology, 2012, 1–11. https://doi.org/10.1155/2012/752815 Nave, L. E., Nadelhoffer, K. J., Le Moine, J. M., van Diepen, L. T. A., Cooch, J. K., & Van Dyke, N. J. (2013). Nitrogen uptake by trees and mycorrhizal fungi in a successional northern temperate forest: Insights from multiple isotopic methods. Ecosystems, 16(4), 590–603. https://doi.org/10.1007/s10021-012-9632-1 Ness, J. H. (2004). Forest edges and fire ants alter the seed shadow of an ant-dispersed plant. Oecologia, 138(3), 448–454. https://doi.org/10.1007/s00442-003-1440-z Potapov, A. M. (2022). Multifunctionality of belowground food webs: Resource, size and spatial energy channels. Biological Reviews, 97(4), 1691–1711. https://doi.org/10.1111/brv.12857 Potapov, Anton M., Frédéric Beaulieu, Klaus Birkhofer, Sarah L. Bluhm, Maxim I. Degtyarev, Miloslav Devetter, Anton A. Goncharov, et al. «Feeding Habits and Multifunctional Classification of Soil‐associated Consumers from Protists to Vertebrates». Biological Reviews 97, n.o 3 (junio de 2022): 1057-1117. https://doi.org/10.1111/brv.12832. Prior, K. M., Meadley‐Dunphy, S. A., & Frederickson, M. E. (2020). Interactions between seed‐dispersing ant species affect plant community composition in field mesocosms. Journal of Animal Ecology, 89(11), 2485–2495. https://doi.org/10.1111/1365-2656.13310 Rochefort, S., Therrien, F., Shetlar, D. J., & Brodeur, J. (2006). Species diversity and seasonal abundance of Collembola in turfgrass ecosystems of North America. Pedobiologia, 50(1), 61–68. https://doi.org/10.1016/j.pedobi.2005.10.007 Rousseau, L., Venier, L., Fleming, R., Hazlett, P., Morris, D., & Handa, I. T. (2018). Long-term effects of biomass removal on soil mesofaunal communities in northeastern Ontario (Canada) jack pine (Pinus banksiana) stands. Forest Ecology and Management, 421, 72–83. https://doi.org/10.1016/j.foreco.2018.02.017 Sanders, D., & van Veen, F. J. F. (2011). Ecosystem engineering and predation: The multi-trophic impact of two ant species. Journal of Animal Ecology, 80(3), 569–576. https://doi.org/10.1111/j.1365-2656.2010.01796.x Smallwood, J. (1982). The effect of shade and competition on emigration rate in the ant Aphaenogaster rudis. Ecology, 63(1), 124–134. https://doi.org/10.2307/1937038 Stuble, K. L., Patterson, C. M., Rodriguez-Cabal, M. A., Ribbons, R. R., Dunn, R. R., & Sanders, N. J. (2014). Ant-mediated seed dispersal in a warmed world. PeerJ, 2, e286. https://doi.org/10.7717/peerj.286 UMBS. (2022, July 18). Taxa of Northern Michigan (northern lower and eastern upper peninsulas). Mfield: Research and Data Hub, University of Michigan. https://mfield.umich.edu/ Villani, M. G., & Wright, R. J. (1990). Environmental Influences on Soil Macroarthropod Behavior in Agricultural Systems. Annual Review of Entomology, 35, 249–269. https://doi.org/10.1146/annurev.en.35.010190.001341 Wolda, H. (1988). Insect seasonality: Why? Annual Review of Ecology and Systematics, 19, 1–18.