Open access journal: http://periodicos.uefs.br/ojs/index.php/sociobiology ISSN: 0361-6525 DOI: 10.13102/sociobiology.v63i4.1195Sociobiology 63(4): 1063-1068 (December, 2016) Ants Visible from Space Influence Soil Properties and Vegetation in Steppe Rangelands of Iran Introduction Information concerning Iranian ant’s effects on environment has tended to be sparse or is often published in Iran.This situation is likely to change due to the opening up of the country and removal of sanctions, although a further stimulus is the recent translation into Persian (farsi) and online availability of the book Ants: Standard Methods for Measuring and Monitoring Biodiversity (Agosti et al., 2000) by Mahsa Ghobadi and Mohammad Mahdavi (Agosti et al., 2015, down-loadable at https://zenodo.org/record/16183#. Vu-hzlKfSjg). A recent series of researches published in Iran by Ghobadi (2013) and Ghobadi et al.(2015; 2016) have quantified the influence of one of the most prominent Messor spp. on soil properties and vegetation in steppe rangelands of Iran. In view of the importance of this work, we here summarize and synthesise the findings of these two papers in order to bring them to the attention of the wider readership. Abstract Messor nests in Iranian steppe rangelands can be so large that they are visible from space. When compared with reference soils, nest soil is higher in nutrients and lower in pH. Ant nests also homogenise the nutrients throughout the upper soil profile, although this effect diminished when nests are abandoned. The denuded circles around nests are surrounded by rings of vegetation that differ in species composition from that of the surrounding vegetation, while abandoned nests are colonized by a different range of plant species. Data on the density and abundance of Messor cf. intermedius nests indicate that the soil in less than 1% of the area is impacted, although the cumulative effect of so many nests influences the plant species and vegetation structure of the region.The data indicate the importance of these ants in altering soil chemical composition and plant diversity, which could have flow-on effects to the diversity of animals. Sociobiology An international journal on social insects M Ghobadi1, D Agosti2, M Mahdavi1, MH Jouri1, J Majer3 Article History Edited by Evandro Nascimento Silva, UEFS, Brazil Received 22 September 2016 Initial acceptance 08 October 2016 Final acceptance 09 October 2016 Publication date 13 January 2017 Keywords Formicidae, ant nest, soil chemistry, plant composition, nest abandonment. Corresponding author Jonathan Majer School of Plant Biology, The University of Western Australia Perth, Australia E-Mail: jonathan.majer@uwa.edu.au The study Harvester ants from the genus Messor count among that selected group of invertebrates whose manifestations are clearly visible from space by satellite imagery (Fig 1). The study was performed in a 30 ha site near Roodshoor, Saveh, Iran, located at a point 60 km along the Tehran-Saveh highway (35.43802° N; 50.89633°E) in an area from which sheep and goats had been excluded for 40 years. The vegetation is a chenopod steppe shrubland whose dominant plant cover includes Artemisia sieberi and Stipa hohenackeriana, along with Salsola tomentosa, Brassica deflexa, and Poasinaica as co-dominant species (Mahdavi et al., 2009). Prominent ant nests were visually located, mapped and measured along ten 200 m long transects. Identification of ants was mainly to species group, since the taxonomy of Iranian ants has not been fully defined. Voucher specimens of these species are deposited in the Naturhistorisches Museum, Bern, (Code NMBE http://grbio.org/cool/5hwi-0wgz). 1 - Islamic Azad University, Noor Branch, Iran 2 - American Museum of Natural History, New York, United States of America 3 - The University of Western Australia, Perth, Australia SHORT NOTE M Ghobadi et al. – Iranian ants visible from space1064 The species recorded represented three genera (Messor (3 species), Cataglyphis (3 species) and Formica (1 species) (Table 1). Other ant species with more cryptic nests may also be present, but Messor cf. intermedius had by far the largest (mean diameter 2.3 m, modal diameter 3.5 m) and most abundant nests (Table 1), which were flat and circular in shape, as evidenced by the light circles in Figure 1. Some of the nests of this species were abandoned, but still visible (Fig 2). The study measured soil physical and chemical properties on 18 live and seven abandoned ant nests and also in un-nested control areas 3 m away from the live nests of Messor cf. intermedius. Chemical properties were measured at four depths and averaged across nests. Fig 1. Location of the study area at Roodshoor, Saveh, Iran, showing location of Messor cf. intermedius ant colonies (light circles like the one in the red square). (Photo courtesy of Google earth). Species Functional group Typical Habitat Color Size Mound shape Density (no./h) Mound Diameter mean (m) ±SE Material composition Messor cf. intermedius. Harvester Dry areas Black Small & Medium Flat 8.3 2.3± 0.03 Seed, Plant, Soil Messor cf. subgracilinodis Harvester Dry areas Black & Red Medium Dome 0.9 1.46± 0.10 Seed, Plant, Soil Messor cf. structor Harvester Dry areas Black Small Pore 0.1 - Soil Cataglyphis bellicosus Scavenger Steppe & Desert Black Large Pore 1.7 - Soil Cataglyphis cf. nodus. Scavenger Steppe & Desert Black & Red Medium Flat 0.5 0.95± 0.05 Soil Cataglyphis cf. lividus. Scavenger Steppe & Desert Orange Small Pore 0.2 - Soil Formica cf. epinotalis General forager Unknown Brown Small Pore 0.1 - Soil Table 1. Properties of ant species found in the study site. Active mound soil has a significantly lower pH than controls (Table 2) and significantly higher levels of nitrogen, phosphorus, potassium, magnesium, calcium, organic matter and carbon and electrical conductivity (Ec) (Table 2). Percentage sand was higher on active ant nests than on controls but percentage clay was lower. Soil temperatures were significantly higher on ant nests than control areas in both dry and humid seasons, as was soil moisture. Water infiltration, as measured by timing water in cylinders to percolate into soil, was significantly higher on ant nests in both dry and humid seasons (Table 2). None of the chemical or physical soil properties differed significantly between abandoned nests and control areas (Table 2), indicating that it had been the activity of ants that was responsible for the differences observed on active nests. Sociobiology 63(4): 1063-1068 (December, 2016) 1065 This was further illustrated when the organic matter and nutrient (N, P and K) levels throughout the soil profile were considered. Duncan’s multiple range test indicated that levels of all four measures were considerably higher in active nest soil compared to control soil at all four measured depths (0-10, 10.1-20, 20-1-30 and 30.1-40 cm), while levels soil of abandoned nests were virtually the same as in control soil (Fig 3). Furthermore, active ant nests had the effect of homogenizing the nutrient distribution throughout the profile, Fig 2. Mounds of (a) active and (b) abandoned Messor cf. intermedius nests at Roodshoor, Saveh, Iran. even though concentrations declined with depth in control area soils. Soil nutrient levels and profiles had regressed to control area levels in abandoned nests (Fig 3). Unlike the control areas, which were dominated by plants Stipa hohenackeriana, Brassica deflexa, and Artemisia sieberi, the active nest areas were dominated by Campanula stricta, Lepidium vesicarium, Achillea tenuifolia, Brassica deflexa, Papaver tenuifolium and Scabiosa oliveri (Table 3), the major part of which were concentrated in 1m wide rings Fig 3. Nutrient concentrations of soil at four depths in active (n = 18) and abandoned (n = 7) Messor cf. intermedius mounds and control areas (n = 18). M Ghobadi et al. – Iranian ants visible from space1066 Properties Mound Control site Abandoned mound pH 7.47 ± 0.01b* 8.30 ± 0.03 a 8.40 ± 0.04 a Ec 3.05 ± 0.03 a 1.43 ± 0.06 b 1.30 ± 0.06 b % OC 0.59 ± 0.04 a 0.18 ± 0.03 b 0.19 ± 0.02 b % OM 1.13 ± 0.08 a 0.33 ± 0.05 b 0.35 ± 0.06 b %N 0.05 ± 0.005 a 0.01 ± 0.03 b 0.01 ± 0.003 b P (ppm) 15.70 ± 0.7 a 4.01 ± 0.03 b 4.08 ± 0.1 b K (ppm) 551.90 ± 3.8 a 320.10 ± 1.02 b 322.50 ± 7.81 b Mg (mg g–1) 110.50 ± 0.2 a 90.20 ± 0.25 b 91.30 ± 0.01 b Ca (mg g–1) 78.20 ± 0.1 a 60.40 ± 0.01 b 61.20 ± 0.03 b % Sand 80.53 ± 0.19 a 77.93 ± 0.18 b 78.00 ± 0.15 b % Silt 8.4 0± 0.16 a 8.90 ± 0.2 a 8.93 ± 0.2 a % Clay 11.06 ± 0.2 b 13.13 ± 0.25 a 13.00 ± 0.21 a Temperate - Dry season 32.00± 0.1 a 28.10 ± 0.01 b 27.50 ± 0.1 b Temperate - Humid season 25.20 ± 0.01a 22.00 ± 0.2 b 23.40 ± 0.003 b Moisture - Dry season 6.50 ± 0.6 a 4.20 ± 0.01 b 4.90 ± 0.02 b Moisture - Humid season 9.30 ± 0.3 a 7.40 ± 0.05 b 7.80 ± 0.001 b Infiltration rate - Dry season (mm ) 34.74 ± 1.30 a 21.66 ± 0.56 b ** Infiltration rate - Humid season (mm ) 40.32 ± 1.88 a 22.94 ± 0.65 b ** Table 2. Line 1-16: Physical and chemical properties of soils (Mean from 0 to 40 cm) from Messor cf. intermedius mounds, dead mounds and control sites in the Roodshoor and significance of the comparison between sites by Duncan’s multiple range test (Mean±S.E; control and live nest each N=18, abandoned nest N= 7). Line 17-18: Results of soil infiltration measuring under conditions of high and low soil moisture contents by double rings method (t-test, Mean±S.E; control and colony each N=5). * The means of the rows with same letters were shown not significantly different by Duncan’s multiple range test ** Not measured around the periphery of the nest, which itself was largely devoid of plant cover (Fig 2). When these surrounding rings were included for comparison with control areas, the active nests had significantly higher vegetation cover, plant species richness and diversity, lower grass cover, higher forb cover and an absence of shrubs. The situation changed markedly on abandoned nests, which were totally dominated by the forb Campanula stricta, leading to significantly higher vegetation cover but lower plant species richness, diversity and evenness than on active nests (Table 3). Furthermore, the vegetation on abandoned nests was more evenly distributed, with no vegetated ring or bare centre being evident (Fig 2). Discussion These observations on the influence of Messor ants add to those from southern Europe (e.g., Cammeraat et al., 2002), Africa (e.g., Dean & Yeaton, 1993) and parts of Western Asia (e.g. Ginzberg et al., 2008; Brown et al., 2012), the other regions of the world where this genus of seed harvester ants is found. As in these other studies, the Iranian Messor nests have a profound influence on soil chemistry, although this effect diminishes once the nests are abandoned and the ants cease their harvesting and soil-moving activities. The time taken for this effect to diminish was not measured. Considering the nest diameter and density of Messor cf. intermedius nests, the area Family Species name IVI Control Active nest Abandoned nest Campanulaceae* Campanula stricta 1.03 18.78 45.03 Silene chaetodonta 1.11 1.1 - Chenopodiaceae Salsola laricina 2.07 - - Salsola tomentosa 2.54 - - Compositae Achillea tenuifolia 1.47 7.11 1.13 Anthemis gilanica 2.35 3.56 1.07 Artemisia sieberi 5.54 - - Centaurea behen 3.84 3.01 - Centaurea bruguierana 1.04 1.21 - Cousinia belangeri 1.07 1.25 - Echinops pungens 1.18 1.05 - Table 3. Above: Mean importance value index (IVI) for plants species and mean (± SE) percentage cover of different functional groups of plants. Below: Plant species richness (S), diversity (H'), and evenness (E) for active and abandoned Messor cf. intermedius nests and control sites (Mean ± SE). Sociobiology 63(4): 1063-1068 (December, 2016) 1067 under the influence of active nests represents less than 1% of the total area, even if the additional but unmeasured influence of Messor cf. subgracilinodis is included. However, the implications to biodiversity are profound. The active nests, and to a lesser extent, the abandoned nests, produce changes in the dominance of the vegetation and, as a result, increase the heterogeneity or patchiness of the environment. This is likely to enhance the diversity of invertebrate (Siemann, 1998), and possibly vertebrate animals (Murkin & Batt, 1987) that directly or indirectly depend on the plant species and vegetation structure of the region or the invertebrates on which they may feed. References Agosti, D., Majer, J., Alonso, E. & Schultz T, (eds.) (2000). Ants: Standard Methods for Measuring and Monitoring Biodiversity. - Biological Diversity Handbook Series. Smithsonian Institution Press, Washington D.C., 20+280 pp. doi: 10.5281/zenodo.11736 Agosti, D., Majer, J., Alonso, E. & Schultz T, (eds.) (2015). Ants: Standard Methods for Measuring and Monitoring Biodiversity. Nasr, Tehran, 533 pp. (Farsi version translated by M. Ghobadi And M. Mahdavi.). doi: 10.5281/zenodo.16183 Brown, G., Scherber, C., Ramos, Jr. P. & Ebrahim, E. K. (2012). The effects of harvester ant (Messor ebenius Forel) nests on vegetation and soil properties in a desert dwarf shrub community in northeastern Arabia. Flora, 207: 503-511. doi: 10.1016/j.flora.2012.06.009 Cammeraat, L. H., Willmott, S. J., Compton, S. G. & Incoll L. D.(2002). The effect of ants’ nests on the physical, chemical and hydrological properties of a rangeland soil in semi-arid Spain. Geoderma, 105: 1-20. doi:10.1016/S0016-7061(01)00085-4 Dean, W. R. J.&Yeaton, R. I.(1993). The effects of harvester ant Messor capensis nest-mounds on the physical and chemical properties of soils in the southern Karoo, South Africa. Journal of Arid Environments, 25: 249–260. doi: 10.1006/ jare.1993.1059 Ghobadi, M.(2013). The effects of harvester ant nest activities (Messor spp.) on some properties of soil structure and function of plants in steppe rangeland of Roodshoor. MSc thesis, Islamic Azad University of Noor. 156 pp. (In Farsi.) Ghobadi, M., Agosti, D., Mahdavi, M. & Jouri, M. H. (2015). Effects of harvester ant nest activity (Messor spp.) on structure and function of plant community in a steppe rangeland (Case Study: Roodshoor, Saveh, Iran). Journal of Rangeland Science, 5: 269-283. doi: 10.5281/zenodo.231998 Ghobadi, M., Agosti, D. & Mahdavi, M. (2016).Change in soil properties by harvester ant’s activity (Messor spp.) in Roodshoor steppe rangeland of Saveh, Iran). Journal of Rangeland Science, 6: 273-285. doi: 10.5281/zenodo.232012 Family Species name IVI Control Active nest Abandoned nest Compositae Senecio vernalis 1.06 1.12 - Taraxacum vulgare complex 1.32 1.61 - Cruciferae Alyssum marginatum 2.11 3.17 - Brassica deflexa 7.07 7 1.03 Lepidium vesicarium 1.21 15.07 - Sisymbrium officinale 1.05 1.06 - Gramineae Stipa hohenackeriana 9.14 1.12 - Aegilops columnaris 1.07 1.17 - Anisantha tectorum 1.21 2.09 - Hordeum murinum 2.11 1.2 - Poa sinaica 2.15 1.07 - Ephedraceae Ephedra strobilacea 2.15 - - Euphorbiaceae Euphorbia sororia 1.14 2.02 - Dipsacaceae Scabiosa oliveri 1.1 5.89 1.03 Geraniaceae Erodium oxyrrhynchum 2.2 3.4 - Labiatae Ziziphora tenuior 1.62 2.73 - Papaveraceae Papaver tenuifolium 1.09 6.07 1.05 Fabaceae Astragalus chaborasicus 2.08 1.1 - Apiaceae Ferula hirtella 2.12 1.11 - Valerianaceae Valerianella oxyrrhyncha 1.02 1.04 - Percentage plant cover ** Grass 14.83 ± 2.30 a 4.44 ± 0.01 b - Forb 21.83 ± 0.42 c 62.94 ± 0.94 b 82.57 ± 1.30 a Shrub 9.22 ± 0.01 - - Indices ** Vegetation Cover (%) 45.88 c 67.88 b 82.57a Richness (S) 6.44 b 12.33 a 2.85 c Diversity (H’) 1.73 b 2.26 a 0.39 c Evenness (E) 0.40 a 0.45 a 0.13 b * The underlined data has higher values of IVI for treatments ** The means of the rows with same letters were shown not significantly different by Duncan’s multiple range test. Table 3. Above: Mean importance value index (IVI) for plants species and mean (± SE) percentage cover of different functional groups of plants. Below: Plant species richness (S), diversity (H'), and evenness (E) for active and abandoned Messor cf. intermedius nests and control sites (Mean ± SE). (Continuation) M Ghobadi et al. – Iranian ants visible from space1068 Ginzburg, O., Whitford, W. G. & Steinberger, Y. (2008). Effects of harvester ant (Messor spp) activity on soil properties and microbial communities in a Negev Desert ecosystem). Biology and Fertility of Soils, 45: 165-173. doi: 10.1007/ s00374-008-0309-z Mahdavi, M., Arzani, H. & Jouri, M. H. (2009). Analysis of rangeland condition changes using a qualitative measure of rangeland health. Rangeland Journal, 3: 397-411. (In Persian). Murkin, R. H. & Batt, B. D. J. (1987). The interactions of vertebrates and invertebrates in peatlands and marshes. Memoirs of the Entomological Society of Canada 119: Supplement S140: 15-30. doi: 10.4039/entm119140015-1 Siemann, E. (1998). Experimental tests of effects of plant productivity and diversity on grassland arthropod diversity. Ecology, 79: 2057-2070. doi: 10.2307/176709 Authors contribution Mahsa Ghobardi conducted this work for a higher degree thesis under the supervision of Donat Agosti, Mohammad Mahdaviand Mohammad Hassan Jouri. Jonathan Majer assisted with the preparation of this review of the work. All data have been cleared for copyright with the journal in which the primary papers were published.