Navigation – Plan du site

AccueilNuméros32-1Anthills as dynamic landforms: a ...

Anthills as dynamic landforms: a 5-year monitoring study of red wood ant mounds in a temperate forest landscape of NW France

Etude zoogéomorphologique de microreliefs biogéniques (dômes de fourmis rousses) dans une forêt tempérée du nord-ouest de la France
François Bétard

Résumés

Les fourmis rousses des bois (groupe Formica rufa) sont des espèces clés jouant un rôle géomorphologique important mais sous-estimé dans les écosystèmes de forêt boréale et tempérée d’Europe. En tant qu’espèces ingénieures, elles creusent des galeries dans des matériaux meubles formant la partie hypogée de leur nid, tandis qu’elles créent en surface des monticules ou dômes de dimension métrique correspondant à la partie épigée du nid. Composés principalement de matière organique mais contenant aussi une fraction minérale, les dômes de fourmis rousses constituent des microhabitats pour de nombreuses espèces myrmécophiles, ce qui accroît la nécessité de protéger et conserver ces microreliefs d’origine biologique. L’objectif de cette étude zoogéomorphologique est d’examiner la distribution spatiale et l’évolution temporelle des dômes de Formica rufa dans un massif forestier de basse altitude et de milieu tempéré océanique, à travers un suivi réalisé sur cinq ans (2020-2024) sur un Espace Naturel Sensible situé dans le nord-ouest de la France (Vendée). Le protocole méthodologique repose sur un relevé GPS exhaustif des dômes de fourmis rousses sur la totalité du massif forestier (62 ha) et sur des mesures morphométriques des dômes le long de 13 transects par bande sélectionnés pour le suivi pluriannuel. Les principales conclusions de l’étude montrent que (i) la densité des dômes varie fortement dans l’espace, principalement influencée par le type de peuplement forestier (avec une préférence pour les peuplements mixtes) et par la déclivité du relief (avec une préférence pour les pentes faibles ou nulles) ; (ii) les dômes de fourmis rousses apparaissent comme des microreliefs dynamiques dont la localisation, la forme, la taille et le volume évoluent rapidement, principalement en réponse à la variabilité climatique interannuelle et aux perturbations anthropiques. Les recherches futures tenteront de déchiffrer les relations possibles entre la distribution des dômes, les émissions de gaz géogéniques et le réseau de fractures armoricaines, et de conduire un inventaire systématique des organismes myrmécophiles dans et autour des plus gros nids, susceptible de révéler une biodiversité extraordinaire associée aux dômes de fourmis rousses.

Haut de page

Notes de la rédaction

Reçu le 11 août 2025, définitivement accepté le 20 octobre 2025

Texte intégral

This work was made possible by the material support of the laboratory Médiations of Sorbonne University, the technical platform Géotéca and the laboratory of physical geography of Paris-Cité University. I am particularly thankful to Matthieu Gastinois (Vendée Department) and Gautier Duthoit (ONF) who authorize and facilitate the monitoring study and participate in the GPS survey and inventory of RWA mounds on the site. Finally, I would like to sincerely thank the two reviewers for their thoughtful and constructive comments, which have helped me to improve the clarity and quality of the manuscript.

1. Introduction

1Ants are ubiquitous and abundant organisms in most terrestrial environments of the Earth surface, from equatorial to subarctic regions, with numerous ant genera known to have geomorphological impacts, such as Lasius and Formica in Eurasia, Pogonomyrmex in Africa, and Atta in central and South America (Whitford and Eldridge, 2013; Viles et al., 2021). As social insects – a biological trait shared with termites –, ants are organized in colonies whose size vary from around twenty to several million workers, with the ability to build important biogenic structures. Despite their high biomass and numerical abundance, the geomorphological role of ants has often been ignored or underestimated (Viles et al., 2021). Their role as “geologic agents” has been recognized for more than a century (Branner and Reid, 1900), but zoogeomorphological research on insects and invertebrates is still underdeveloped compared to studies on vertebrates (Butler, 1995; Bétard, 2021). Among the various ant communities that inhabit the Earth surface, non-arboreal, soil-dwelling ants play many geomorphological roles in terrestrial ecosystems. They create small-scale landforms such as subterranean galleries and surface mounds and, by this way, exert an influence on hydrology, soil erosion and sediment transfer at a wider landscape scale (Bétard, 2021). They also have a pedoturbational impact and play important roles in biogeochemical cycling, rock weathering and soil chemical processes (Whitford and Eldridge, 2013; Viles et al., 2021).

2In boreal and temperate forests of Eurasia, red wood ants (RWA) of the Formica rufa group are considered keystone species and ecosystem engineers as they change their habitat by burrowing subterranean galleries, modifying soil physical and chemical properties, and constructing prominent mounds – up to 3 m high and 20 m wide – that can last many years (Lenoir, 2001; Frouz et al., 2016; Sorvari, 2022) (fig. 1). Within the Formica rufa group, eight mound-building RWA species are known to occur in Europe: Formica rufa Linnaeus, 1761, Formica polyctena Förster, 1850, Formica pratensis Retzius, 1783, Formica aquilonia Yarrow, 1955, Formica lugubris Zetterstedt, 1838, Formica paralugubris Seifert, 1996, Formica truncorum Fabricius, 1804, and Formica uralensis Ruzsky, 1895, with frequent cases of hybridization between species. All RWA species are known to play many important ecological roles in forest ecosystems. They help to disperse the seeds of many herbaceous and woody plants around their nests – a process called myrmecochory – and therefore influence vegetation dynamics and floristic biodiversity. They also have multi-trophic interactions with other animal communities as they are predators of many invertebrates, are mutualists with aphids and scale insects, and are a food resource for many arthropods, birds and other vertebrates (Robinson et al., 2016; Sorvari, 2022). As generalist predators and sentinel species, they participate in the regulation of forest pests (processionary caterpillars, ticks, etc.) and have a decisive role in the dynamic equilibrium of the forest ecosystem (Torossian and Humbert, 1982; Gouraud, 2021; Fürjes-Mikó et al., 2025). In addition to their interest for biological pest control, RWA are considered as good indicators of the state of conservation of forest ecosystems (Torossian, 1979; Marage et al., 2017). Finally, RWA are essential umbrella species since their nests serve as microhabitats for a high number of associated species (i.e., myrmecophiles), so their conservation is relevant for a wide range of other taxa (Sorvari, 2022; Balzani et al., 2022).

Fig. 1 – Red wood ants and related biogenic landforms. A: worker of Formica rufa (photo: Michal Kukla, CC BY-SA 4.0); B: anthills created by Formica rufa in the study site (Bois des Jarries, Vendée, NW France); the mound in the foreground is  1 m high and is aligned with two other smallest mounds.
Fig. 1 – Les fourmis rousses des bois et leurs constructions biogéomorphologiques. A : ouvrière de Formica rufa (photo : Michal Kukla, CC BY-SA 4.0) ; B: dômes construits par Formica rufa sur le site d’étude (Bois des Jarries, Vendée, France) ; le dôme au premier plan fait  1 m de hauteur et est aligné avec deux autres dômes plus petits.

Fig. 1 – Red wood ants and related biogenic landforms. A: worker of Formica rufa (photo: Michal Kukla, CC BY-SA 4.0); B: anthills created by Formica rufa in the study site (Bois des Jarries, Vendée, NW France); the mound in the foreground is  1 m high and is aligned with two other smallest mounds.Fig. 1 – Les fourmis rousses des bois et leurs constructions biogéomorphologiques. A : ouvrière de Formica rufa (photo : Michal Kukla, CC BY-SA 4.0) ; B: dômes construits par Formica rufa sur le site d’étude (Bois des Jarries, Vendée, France) ; le dôme au premier plan fait  1 m de hauteur et est aligné avec deux autres dômes plus petits.

(photo: François Bétard).
(photo : François Bétard).

3While the role of RWA as geomorphic agents has rarely been highlighted (Dupuis and Verger, 1964; Bétard, 2021), they undoubtedly modify the geomorphic environment by creating a rugged topography of prominent mounds, or “moundscape”, under the forest canopy, and by excavating significant volumes of sediment from the subterranean galleries forming the hypogean nest. Thus, the epigean nest (i.e., mound), mainly composed of organic material collected by the ants in the forest environment (pine needles, twigs, etc.), also contains a proportion of mineral grains originated in the loose sediment or weathering materials excavated from the burrowing galleries. RWA anthills can therefore be considered as constructional biogenic landforms which participate in the geomorphological heterogeneity of the forest land systems and contribute to the soil bioturbation and sediment redistribution on the hillslopes. Mound distributions and densities vary from place to place and are dependent from a range of environmental variables (altitude, geology, forest stands, human pressure, etc.). Additionally, RWA mounds rapidly change over time – in terms of shape, size and volume – in response to worker activity itself dependent from external factors such as climate variability and anthropogenic disturbances.

4In this study, I examine the spatial distribution and temporal evolution of RWA mounds in a temperate forest landscape of Northwestern France through a 5-year monitoring study (2020-2024). Whereas most studies on RWA focus on northern (boreal) and central (continental) Europe, this case study offers new data on RWA mound densities and temporal changes in a lowland (< 300 m a.s.l.) and temperate oceanic environment (< 100 km from the Atlantic coast). By combining an exhaustive GPS mapping and a monitoring study of anthills involving annual morphometric measurements along 13 strip transects, it provides a qualitative and quantitative assessment of how these dynamic landforms arrange and evolve over space and time. All collected data are interpreted by examining the main environmental variables controlling the densities and evolution of mounds during the survey period. The discussion opens on (i) the possible relationships between RWA mound distribution and the tectonic framework, (ii) the influence of climate change and interannual variability on RWA mound locations and volumes, and finally (iii) the main conservation issues and ecological implications related to the presence of ant mounds which can be considered as a biogeomorphological heritage to be protected and conserved.

2. Materials and methods

2.1. Study site

5The study was carried out in a forest area of 62 ha located in Northwestern France (Bois des Jarries; 46°51'27"N, 0°54'12"W; fig. 2). This is an old-growth forest, whose existence is attested to by Cassini’s map (the forest massif already existed in the 18th century). Today the study site is a managed public forest, protected as a Sensitive Natural Area (ENS), with an integrative sustainable forest management by the French National Forest Office (ONF) and the Nature Service of the Vendée Department. The objective of the 2019-2038 management plan is to promote biodiversity and ecological diversity of the woodland, to maintain the quality of the landscape while facilitating public access, and to ensure the long-term survival of the forest stands, through a multifunctional view combining the protective, recreational and productive functions of the forest (ONF, 2019). The 12 forest stands (tab. 1) can be grouped into two main types of vegetation structure and woodland management: (i) simple coppice of oak (Quercus robur and Q. petraea) and chestnut (Castanea sativa), sometimes mixed with isolated conifer trees (Pinus sylvestris and P. pinaster), and (ii) high forest, stump-planted (deciduous trees) or regularly-planted (conifers). In this landscape matrix, three different age classes can be distinguished: (i) < 10 year-old stands after clear-cutting (chestnut coppices), (ii) 10-35 year-old stands (oak and chestnut coppices, chestnut high stands), and (iii) > 35 year-old stands (mature oak and pine plantations). Although clear-cutting was the norm until recently (two forest plots were clear-cut in 2017 and 2021), the stands are now managed according to sustainable practices including thinning at appropriate times. Moreover, an area of 12.27 ha of “long time unmanaged forest” (HSN: excluding forestry – free-evolving area) has been defined in the summit part of the forest massif, in order to maintain old conifer and oak stands and to favour biodiversity in general. Finally, selective ecological engineering operations are regularly conducted to restore relict dry Atlantic heathlands by removing the stunted oak coppice, thus favouring the edge effects and the maintenance of low shrubs and woody plants (Erica cinerea, Ulex europaeus) that are favourable to a rich biodiversity (flora, avifauna, reptiles, entomofauna).

Tab. 1 – Synthetic description of the 12 forest stands of the study site (after ONF, 2019).
Tab. 1 – Description synthétique des 12 types de peuplements forestiers sur le site d’étude (d’après ONF, 2019).

Tab. 1 – Synthetic description of the 12 forest stands of the study site (after ONF, 2019).Tab. 1 – Description synthétique des 12 types de peuplements forestiers sur le site d’étude (d’après ONF, 2019).

6Geologically, the study site is composed of medium-grained to porphyritic Variscan granites with a complex geodynamic history (fig. 2C). Late Variscan strike-slip movements created NW-SE trending dextral faults throughout the granitic massif, orthogonally intersected by NE-SW trending secondary faults (Rolin et al., 2000). After a long phase of continental weathering and erosion during the Mesozoic and the Palaeogene, post-Eocene transtensional faulting reactivated the Variscan faults and accompanied a post-Eocene regional uplift (100 to 150 m) with subsequent denudation, in response to the later phase of the Alpine lithospheric buckling (Bétard, 2010). Today, the studied forest is located on two granitic hills culminating at 235 and 262 m a.s.l., overlooking to the north a wide erosion corridor interpreted as an Ypresian fluvial paleo-valley (Godard et al., 1994). Under the forest cover, sandy-skeletal and brown soils develop on a shallow granitic grus, covered by thin colluvial deposits on the hillsides. Locally, minor granite bedrock landforms (<1 m high) emerge from the arenaceous weathering mantle. A wide Mesolithic site has been discovered at this place in the 1940s, with a lithic industry buried into the soil or displaced on the hillsides by colluviation (Rouillon and Daniel, 1991). Since the early Holocene, the climate has been temperate oceanic. Over the period 1991-2021, the annual mean air temperature in the area is 12.1 °C (average monthly temperatures of 19.4 °C in July and 5.5 °C in January) and the mean annual precipitation is 814 mm.

Fig. 2 – Location of the study site. A: hypsometric map from the RGEAlti 1 m DTM (IGN/Géoservices); equidistance of contour lines 5 m; B: aerial photograph showing the studied forest massif isolated in an agricultural-dominated hedgerow landscape (IGN/Géoportail); the photograph was acquired on 10 May 2022; C: geological map at 1:50,000 scale showing the main lithologies (BRGM/Infoterre).
Fig. 2 – Localisation du site d’étude. A : carte hyspométrique dérivée du MNT RGE Alti 1 m (IGN/Géoservices) ; équidistance des courbes de niveau 5 m ; B : photographie aérienne montrant le massif forestier étudié, isolé dans un paysage de bocage majoritairement agricole (IGN/Géoportail) ; la photographie a été acquise le 10 mai 2022 ; C : carte géologique à l’échelle 1/50 000 indiquant les principales lithologies rencontrées dans la zone d’étude (BRGM/Infoterre).

Fig. 2 – Location of the study site. A: hypsometric map from the RGEAlti 1 m DTM (IGN/Géoservices); equidistance of contour lines 5 m; B: aerial photograph showing the studied forest massif isolated in an agricultural-dominated hedgerow landscape (IGN/Géoportail); the photograph was acquired on 10 May 2022; C: geological map at 1:50,000 scale showing the main lithologies (BRGM/Infoterre).Fig. 2 – Localisation du site d’étude. A : carte hyspométrique dérivée du MNT RGE Alti 1 m (IGN/Géoservices) ; équidistance des courbes de niveau 5 m ; B : photographie aérienne montrant le massif forestier étudié, isolé dans un paysage de bocage majoritairement agricole (IGN/Géoportail) ; la photographie a été acquise le 10 mai 2022 ; C : carte géologique à l’échelle 1/50 000 indiquant les principales lithologies rencontrées dans la zone d’étude (BRGM/Infoterre).

2.2. Inventory and mapping of RWA mounds

7The zoogeomorphological study was conducted over a 5-year period (2020-2024). During the summer of 2020, a field survey enabled a first inventory of RWA mounds along the forest paths, the collection of ant specimens for determination, and the selection of the forest plots for application of the sampling method (see below, 2.3). Between February and April 2024, all the ant mounds were inventoried in the 12 stands of the forest massif, covering a total area of 62 ha. RWA mounds were inventoried and mapped using a GNSS receiver and antenna connected to the CentipedeRTK network, with centimetric accuracy. Mounds were recorded in the field using mobile GIS mapping software SW Maps 2.10.1.0. Ant nests were classified according to approximate mound size (see 3.2 for the classification of mound sizes used during the survey sampling). GPS track recording provided good coverage of the study area, with a maximum spacing of 20 m between each track, that is sufficient to locate ant mounds in the undergrowth in late winter or early spring. GIS processing, geostatistical analyses and cartographic production were carried out using ArcGIS Pro 3.2.0 software.

8To investigate the influence of environmental variables on RWA mound distribution and densities, several geospatial data were added to the GIS database. Forest information was taken from the 2019-2038 forest management plan provided by the French National Forest Office (ONF, 2019). Geological variables (lithological and geochemical composition of the bedrock, tectonic framework) were extracted from the BDCharm-50 database which compiles the BRGM’s geological maps of each mainland France Department, vectorised and harmonised to the scale of 1:50,000. The influence of geomorphology on RWA mound distribution was assessed through the calculation of morphometric parameters (terrain slope, exposure) from a 1 m-resolution digital terrain model (RGEAlti DTM, IGN). Terrain slope was calculated by applying the slope gradient algorithm of the Spatial Analyst Toolbox (ArcGIS Pro 3.2.0). Six slope classes were determined: (i) flat terrain (< 1°), (ii) very gentle slopes (1-5°), (iii) gentle slopes (6-10°), (iv) moderate slopes (11-15°), (v) steep slopes (16-20°), and (vi) very steep slopes (> 20°). Terrain exposures were calculated using the aspect algorithm of the Spatial Analyst Toolbox (ArcGIS Pro 3.2.0). Descriptive statistics (numbers and densities of mapped mounds per unit/class) were calculated for all the environmental variables. Finally, expected numbers of RWA nests were calculated proportional to the investigated forest area.

2.3. Monitoring of RWA mounds: sampling protocol and morphometric parameters

9One of the objectives of this study was to carry out a monitoring of Formica rufa mounds on the forest massif in order to assess the status and evolution of ant populations during five years (2020-2024) and to analyse the morphological changes of mounds over time. The sampling method, adapted from Torossian (1979), was based on a qualitative and quantitative study of anthills along strip transects representative of the different forest stands. 13 strip transects (50 m long x 20 m wide) were defined according to the 12 main stand types (dry heaths and stunted coppices of native oak, grouped in the same stand type in the forest management plan, were here separated) (fig. 3). This transect method makes it possible to calculate mean and total mound volumes based on surface areas of 1000 m² and to compare them every year in the same location and at the same period (end of July). The survey only counts mounds located within the transect right-of-way (10 m on either side of the transect line) and requires measuring the perpendicular distance of each mound from the line. For all observed mounds, the following parameters were recorded in the field:

  • mound morphometric parameters: large diameter (D), small diameter (d), height (H);

  • nest activity: many workers on the nest (N), moderate activity (M), few workers on the nest (R), abandoned nests (A);

  • slope: flat, gentle, moderate, steep;

  • aspect: N, NE, NW, S, SE, SW, E, W;

  • sun exposure: weak, medium, strong;

  • nest materials: needles, twigs, quartz grains...;

  • presence of dead wood: standing (P) or on the ground (S), with measurement of average diameter of observed woods;

  • vegetation structure and floristic composition of the different vegetation strata.

Fig. 3 – Forest stands and sampling plan locating the 13 strip transects at the study site (map base from ONF, 2019).
Fig. 3 – Types de peuplements forestiers et plan d’échantillonnage localisant les 13 transects par bande sur le site d’étude (fond cartographique d’après ONF, 2019).

Fig. 3 – Forest stands and sampling plan locating the 13 strip transects at the study site (map base from ONF, 2019).Fig. 3 – Types de peuplements forestiers et plan d’échantillonnage localisant les 13 transects par bande sur le site d’étude (fond cartographique d’après ONF, 2019).

1: Scots pines with chestnuts; 2: Atlas cedars; 3: native oaks; 4: oaks with chestnuts; 5: chestnuts; 6: chestnuts with oaks; 7: native oaks; 8: oaks and chestnuts; 9: chestnuts; 10: chestnuts and maritime pines; 11: chestnuts and Scots pines; 12: dry heaths and stunted oak coppice on skeletal soils; 13: forest paths; 14: plot limits; 15: monitored transect. Numbers with white outline are plot numbers; numbers in italics are subplot numbers.
1 : pin sylvestre sur châtaignier ; 2 : cèdre de l’Atlas ; 3 : chênes indigènes ; 4 : chênes et châtaignier ; 5 : châtaignier ; 6 : châtaignier et chênes ; 7 : chênes indigènes ; 8 : chênes et châtaignier ; 9 : châtaignier ; 10 : châtaignier et pin maritime ; 11 : châtaignier et pin sylvestre ; 12 : lande sèche et taillis rabougri sur sol rocheux superficiel ; 13 : sentiers forestiers ; 14 : limite de parcelle forestière ; 15 : transects étudiés. Les numéros sur fond blanc sont les numéros de parcelles forestières ; les numéros en italique sont ceux des sous-parcelles.

10Based on field surveys and measurements, several indices, defined by Torossian (1979) and Boudjema et al. (2006), were calculated for each transect:

  • the pseudobiovolume, characterizing all nest materials (needles, twigs...) forming the epigeous dome: mound volume was calculated according to the average obtained from the mathematical formulas of a semi-ellipsoid [1] and of a conical dome [2]:

  • the necrovolume, corresponding to the volume of abandoned nests (A) or nests being abandoned (R);

  • the total biovolume of active nests, obtained by subtracting necrovolume from biovolume;

  • the mean biovolume, corresponding to the total volume of active nests divided by the number of mounds.

11Finally, a classification of mound size into four classes was applied based on calculated nest volumes. The used classification follows that proposed by Torossian (1979) for RWA mounds in montane conifer forests but can also apply to lowland mixed forests (Nageleisen, 1999):

  • small mounds: < 32 dm3;

  • medium mounds: from 32 dm3 to 256 dm3;

  • large mounds: from 256 dm3 to 2,048 dm3;

  • very large mounds: > 2,048 dm3.

3. Results

3.1. RWA mound distribution, densities and variations over space

12A total of 767 RWA mounds constructed by the species Formica rufa were recorded within the studied forest massif, for an average density of 12.4 mounds/ha. The number of mounds per hectare differs significantly among the forest stand types (fig. 4). The highest values are found in mixed-conifer forest stands, where densities can reach 20 mounds/ha (tab. 2). In contrast, monospecific chestnut coppices show the lowest values, with average densities less than 10 mounds/ha. On forest plot n°2, the very dense recent coppice resulting from a clear-cut in 2017 shows particularly low densities (5.5 nests/ha), with most of the mapped mounds located at the edge of the plot (fig. 3 and 4). In plot n°6, a more recent clear-cut in 2021 (sparing only a few isolated Scots pines) resulted in high nest densities in July 2024 (30.7 nests/ha), but this is misleading: 3 years after the clear-cut, the coppice density was still relatively low (still allowing plenty of light to pass through) and the vast majority of nests inventoried were small (only 13 % of large mounds). This recent chestnut coppice will gradually evolve into a dense, closed coppice comparable to that of plot n°2, with the inevitable abandonment of the vast majority of newly-formed mounds.

Fig. 4 – Spatial distribution of RWA mounds according to the main forest stand types at the study site.
Fig. 4 – Distribution spatiale des dômes de Formica rufa selon les principaux types de peuplements forestiers sur le site d’étude.

Fig. 4 – Spatial distribution of RWA mounds according to the main forest stand types at the study site.Fig. 4 – Distribution spatiale des dômes de Formica rufa selon les principaux types de peuplements forestiers sur le site d’étude.

1: stumped-planted high forest of oaks and chestnut; 2: mixed conifer regularly planted high forest; 3: monospecific coppice of chestnut; 4: coppice of oaks and/or chestnut; 5: long-term unmanaged oak coppice; 6: clear-cutting in 2017; 7: clear-cutting in 2021; 8: very large mound; 9: large mound; 10: medium mound; 11: small mound.
1 : futaie sur souche de chênes et châtaignier ; 2 : peuplement mixte résineux-châtaignier ; 3 : taillis de châtaignier monospécifique ; 4 : taillis de chênes et/ou châtaignier ; 5 : taillis de chênes laissé en libre évolution ; 6 : taillis récent (coupe rase en 2017) ; 7 : taillis récent (coupe rase en 2021) ; 8 : très gros dôme ; 9 : gros dôme ; 10 : dôme moyen ; 11 : petit dôme.

Tab. 2Descriptive statistics of expected numbers of RWA nests, mapped nests and mound densities according to the main forest stand types.
Tab. 2 – Statistiques descriptives sur les nombres de nids attendus, les nids inventoriés et les densités de nids selon le type de peuplement forestier.

Tab. 2 – Descriptive statistics of expected numbers of RWA nests, mapped nests and mound densities according to the main forest stand types. Tab. 2 – Statistiques descriptives sur les nombres de nids attendus, les nids inventoriés et les densités de nids selon le type de peuplement forestier.

Expected nest values are obtained by multiplying the total number of nests (n = 767) by the percentage of the area of each class.
Les valeurs de nids attendues sont obtenues en multipliant le nombre total de nids (n = 767) par le pourcentage de la superficie de chaque classe.

13Native oak coppices classified as HSN (“long time unmanaged forest”) in the forest management plan are proving to favour the establishment and maintenance of RWA mounds, with an average density of 14.9 nests/ha. These environments are associated with the largest and probably oldest nests, undisturbed by silvicultural interventions. Stands treated as regular high forest or ‘futaie’ (dominated by sessile or pedunculate oak, often with a proportion of coppiced chestnuts) are also destined to last longer in the forest landscape, without major silvicultural intervention: here too, high anthill densities are maintained (13.9 nests/ha), with a high proportion of large nests (> 50 %). As a rule, simple coppicing, particularly as applied to pure chestnut stands (i.e., monospecific), is very unfavourable to Formica rufa settlement, compared with a plot managed as a diversified high forest or coppice with a mixture of species and undergrowth strata (fig. 5).

Fig. 5 – Two adjacent forest stands in the northern part of the study site, which are very different in terms of their suitability for RWA nests. A: coppice of native oak trees with diverse undergrowth vegetation, conducive to the establishment of numerous ant nests; B: coppice of monospecific chestnut trees, with very poor undergrowth vegetation, an environment unfavorable to the establishment of Formica rufa.
Fig. 5 – Deux peuplements forestiers mitoyens dans la partie nord du boisement, très contrastés en termes d’accueil de dômes de fourmis rousses. A : taillis de chênes indigènes avec une végétation diversifiée de sous-bois, propice à l’implantation de nombreuses fourmilières ; B : taillis de châtaigniers monospécifiques, très pauvre au niveau du sous-bois, un milieu défavorable à l’installation de Formica rufa.

Fig. 5 – Two adjacent forest stands in the northern part of the study site, which are very different in terms of their suitability for RWA nests. A: coppice of native oak trees with diverse undergrowth vegetation, conducive to the establishment of numerous ant nests; B: coppice of monospecific chestnut trees, with very poor undergrowth vegetation, an environment unfavorable to the establishment of Formica rufa.Fig. 5 – Deux peuplements forestiers mitoyens dans la partie nord du boisement, très contrastés en termes d’accueil de dômes de fourmis rousses. A : taillis de chênes indigènes avec une végétation diversifiée de sous-bois, propice à l’implantation de nombreuses fourmilières ; B : taillis de châtaigniers monospécifiques, très pauvre au niveau du sous-bois, un milieu défavorable à l’installation de Formica rufa.

Photos: François Bétard, 15 April 2024.
Photos : François Bétard, 15 avril 2024.

14Among the other environmental variables, geomorphology is expected to have a strong influence on spatial variations of RWA mounds, given the high geomorphological heterogeneity of the study site. An overlay of mound distribution on a slope map (derived from 1 m DTM) shows a preference for Formica rufa on low-gradient (15.1 nests/ha), very low-gradient (17.5 nests/ha) or flat topography (14.6 nests/ha), while steep (7.9 nests/ha) to very steep slopes (3.7 nests/ha) are clearly less favourable (fig. 6; tab. 3). Surprisingly, slope aspect is not a relevant factor for RWA mounds: densities are fairly comparable whatever the exposures of the slope (not shown). At the scale of the granite massif, the lithological nature of the substratum does not appear as a discriminant factor, between leucogranites and leuco-monzogranites which remain very similar in terms of petrographic and geochemical composition (fig. 7). However, the decomposition of the granite rocks into a sandy grus appears clearly favourable for digging galleries and building nests, compared to adjacent lithologies of schists and hornfels where RWA mounds are absent. Other parameters linked to geology, such as the fault pattern, might have an important role in the distribution and densities of RWA mounds in the study area, as suggested by the frequent coincidence between nest alignments and known tectonic directions in the region, in particular NW-SE (N120°) and NE-SW (N40°) trending fractures (fig. 7). Additional data and complementary analyses would be needed to demonstrate and explain such a relationship (see discussion below, 4.1).

Fig. 6 – Spatial distribution of RWA mounds according to the slope gradient at the study site.
Fig. 6 – Distribution spatiale des dômes de Formica rufa selon la pente (en °) sur le site d’étude.

Fig. 6 – Spatial distribution of RWA mounds according to the slope gradient at the study site.Fig. 6 – Distribution spatiale des dômes de Formica rufa selon la pente (en °) sur le site d’étude.

1: very large mound; 2: large mound; 3: medium mound; 4: small mound; 5: ENS perimeter.
1 : très gros dôme ; 2 : gros dôme ; 3 : dôme moyen ; 4 : petit dôme ; 5 : périmètre ENS.

Tab. 3 - Descriptive statistics of expected numbers of RWA nests, mapped nests and mound densities according to the slope gradient.
Tab. 3 – Statistiques descriptives sur les nombres de nids attendus, les nids inventoriés et les densités de nids selon la classe de pente.

Tab. 3 - Descriptive statistics of expected numbers of RWA nests, mapped nests and mound densities according to the slope gradient. Tab. 3 – Statistiques descriptives sur les nombres de nids attendus, les nids inventoriés et les densités de nids selon la classe de pente.

Expected nest values are obtained by multiplying the total number of nests (n = 767) by the percentage of the area of each class.
Les valeurs de nids attendues sont obtenues en multipliant le nombre total de nids (n = 767) par le pourcentage de la superficie de chaque classe.

Fig. 7 – Spatial distribution of RWA mounds according to the main lithologies and possible relationships with tectonic faults at the study site.
Fig. 7 – Distribution spatiale des dômes de Formica rufa selon la lithologie et relations possibles avec la trame tectonique sur le site d’étude.

Fig. 7 – Spatial distribution of RWA mounds according to the main lithologies and possible relationships with tectonic faults at the study site.Fig. 7 – Distribution spatiale des dômes de Formica rufa selon la lithologie et relations possibles avec la trame tectonique sur le site d’étude.

1: leucogranite; 2: leuco-monzogranite; 3: hornfels and spotted schists; 4: schists; 5: sandstones; 6: mapped fault (BRGM); 7: inferred fault (this study); 8: very large mound; 9: large mound; 10: medium mound; 11: small mound; 12: ENS perimeter.
1 : leucogranite ; 2 : leuco-monzogranite ; 3 : cornéennes et schistes tachetés ; 4 : schistes ; 5 : grès ; 6 : faille cartographiée (BRGM) ; 7 : faille interprétée (cette étude) ; 8 : très gros dôme ; 9 : gros dôme ; 10 : dôme moyen ; 11 : petit dôme ; 12 : périmètre ENS.

3.2. RWA mound morphometry, volumes and changes over time

15Results of the 5-year monitoring study along the 13 transects allowed to collect information on the types of mounds (shape, size class), their activity (active vs. abandoned mounds) and the nest materials, as well as indications on their evolution and morphological change over time. Various shapes were observed in the field, from simple domes and cone-shaped mounds – with or without wood armature – to more complex mounds (nested cones with two generations of mound, half domes leaning against a trunk, dissymmetric domes on talus or hillslopes, etc.). Active and abandoned mounds were recorded along each transect based on the abundance of workers on the nests, but they are also discriminable based on their morphometry (dome- or cone-shaped active mounds vs. truncated or flattened abandoned mounds). The preferred nest materials are pine needles (from Pinus sylvestris and/or P. pinaster, and locally from Cedrus atlantica) and the volumes are typically 90 % of organic material and 10 % of mineral grains (mainly quartz). Based on measurements on the 13 transects during the 5-year period, mean biovolumes are significantly higher in mixed-conifer forest stands than in deciduous stands (fig. 8). This observation is similar to that made by other authors in lowland and mountain forests in eastern France (e.g., Nageleisen, 1999; Lempérière et al., 2002), pointing out a strong link between RWA populations and mixed or coniferous stands. It is mainly explained by the materials available and adapted to the construction of mounds (pine needles), the lighting (more open forest stands favouring partial sunlight) and the available food resource (honeydew excreted by needle aphids and other sap-sucking insects).

Fig. 8 – Variations of size and activity of RWA mounds along two strip transects (50 x 20 m) during five consecutive years (between 2020 and 2024).
Fig. 8 – Variations de la taille et de l’activité des dômes de fourmis rousses le long de deux transects par bande (50 x 20 m) durant cinq années consécutives (entre 2020 et 2024).

Fig. 8 – Variations of size and activity of RWA mounds along two strip transects (50 x 20 m) during five consecutive years (between 2020 and 2024). Fig. 8 – Variations de la taille et de l’activité des dômes de fourmis rousses le long de deux transects par bande (50 x 20 m) durant cinq années consécutives (entre 2020 et 2024).

Diameters are in cubic root of volume. Dark: active mounds; light gray: abandoned mounds. Note the aligned mounds along known tectonic directions in the region.
Les diamètres sont exprimés en racine cubique du volume. Noir : dômes actifs ; gris clair : dômes abandonnés. Noter les alignements de nids selon des directions tectoniques connues dans la région.

16Between 2020 and 2024, the monitoring survey shows relative stability of biovolumes in the mixed-conifer forest stands, which provide the best habitat for the species. The largest mounds of the monitored transects are found in these habitats: some of them have grown rapidly, reaching up to 1 m in height and over 3 cubic meters in less than 5 years, involving high bioturbation and bioconstruction rates (fig. 9). Transects in deciduous stands, with overall lower biovolumes, also show greater interannual variability (fig. 8). Level of nest activity could partly reflect the climate variations and meteorological conditions during the five years of the survey (fig. 10). Indeed, the year 2022 corresponds to the lowest nest activity, strongly impacted by the spring drought and summer heatwave that affected Northwestern France at that time. The resumption of nest activity in 2023 and 2024 could be attributed to cooler and wetter weather conditions, particularly during the spring preceding the surveys (fig. 10). These observations confirm the idea that excessively hot and dry climatic conditions are unfavourable to Formica rufa, but they also indicate some resilience of ant colonies to interannual climate variability. In many places, nests that appeared to have been abandoned are sometimes reactivated in more favourable weather years, while dome relocations are common immediately after a nest has been damaged or abandoned (fig. 9). Finally, between 2020 and 2024, the evolution of necrovolumes has been relatively stable, and there has even been an increase in total biovolumes across the 13 studied transects. All these results therefore indicate the overall good health of RWA populations on the study site.

Fig. 9 – Comparative evolution of biovolumes of RWA mounds according to the two main forest stand types, between 2020 and 2024, on the 13 monitored transects.
Fig. 9 – Évolution comparée du biovolume des dômes de fourmis rousses selon les deux principaux types de peuplement forestier, entre 2020 et 2024, sur les 13 transects étudiés.

Fig. 9 – Comparative evolution of biovolumes of RWA mounds according to the two main forest stand types, between 2020 and 2024, on the 13 monitored transects.Fig. 9 – Évolution comparée du biovolume des dômes de fourmis rousses selon les deux principaux types de peuplement forestier, entre 2020 et 2024, sur les 13 transects étudiés.

1. Mixed-conifer forest stands; 2. Deciduous forest stands.
1. Peuplements mixtes résineux-feuillus ; 2. Peuplements feuillus.

Fig. 10 – Relationships between interannual climate variability and nest activity of RWA mounds during the 5-year monitoring study on the 13 strip transects. A: Average temperatures (°C) and total precipitation (mm) at the meteorological station of Les Herbiers (located at 8 km from the study site) between March and July corresponding to the period of activity of red wood ants preceding the surveys (source of climatic data: https://www.mellifere.com/​climat/​85500-les-herbiers.php); B: Level of nest activity during the surveys: N: many workers on the nest; M: moderate activity; R: few workers on the nest; A: abandoned nests. The dotted line represents the percentage of nests abandoned (A) and being abandoned (R).
Fig. 10 – Relations entre variabilité climatique interannuelle et niveau d’activité des nids de Formica rufa selon les années sur l’ensemble des 13 transects étudiés. A : Températures moyennes (°C) et précipitations totales (mm) à la station des Herbiers (située à 8 km du site d’étude) entre les mois de mars et juillet correspondant à la période d’activité des fourmis rousses précédant les relevés (source des données climatiques : https://www.mellifere.com/​climat/​85500-les-herbiers.php) ; B : Niveau d’activité des nids lors des relevés : N : nombreuses ouvrières sur le nid ; M : activité moyenne ; R : quelques rares ouvrières sur le dôme ; A : nid abandonné. La ligne pointillée représente la part de nids abandonnés (A) et en cours d’abandon (R).

Fig. 10 – Relationships between interannual climate variability and nest activity of RWA mounds during the 5-year monitoring study on the 13 strip transects. A: Average temperatures (°C) and total precipitation (mm) at the meteorological station of Les Herbiers (located at 8 km from the study site) between March and July corresponding to the period of activity of red wood ants preceding the surveys (source of climatic data: https://www.mellifere.com/​climat/​85500-les-herbiers.php); B: Level of nest activity during the surveys: N: many workers on the nest; M: moderate activity; R: few workers on the nest; A: abandoned nests. The dotted line represents the percentage of nests abandoned (A) and being abandoned (R).Fig. 10 – Relations entre variabilité climatique interannuelle et niveau d’activité des nids de Formica rufa selon les années sur l’ensemble des 13 transects étudiés. A : Températures moyennes (°C) et précipitations totales (mm) à la station des Herbiers (située à 8 km du site d’étude) entre les mois de mars et juillet correspondant à la période d’activité des fourmis rousses précédant les relevés (source des données climatiques : https://www.mellifere.com/​climat/​85500-les-herbiers.php) ; B : Niveau d’activité des nids lors des relevés : N : nombreuses ouvrières sur le nid ; M : activité moyenne ; R : quelques rares ouvrières sur le dôme ; A : nid abandonné. La ligne pointillée représente la part de nids abandonnés (A) et en cours d’abandon (R).

4. Discussion

4.1. Tectonic framework, geogenic gases and distribution of RWA mounds

17Recent scientific research conducted in several regions of central and eastern Europe revealed that the spatial distribution of RWA mounds and gas-bearing fault zones are strongly correlated (Schreiber et al., 2009; Berberich et Schreiber, 2013; Berberich et al., 2016; Del Toro et al., 2017; Berberich et al., 2019). In all the areas investigated, RWA nests are not distributed randomly, but exhibit a spatial pattern as revealed by geostatistical analyses: most RWA mounds are located on or near active gas permeable faults. More specifically, results of these studies showed that RWA mounds were eight times more likely to be found within 60 m of known tectonic faults (Berberich et al. 2016; Del Toro et al. 2017; Berberich et al., 2019). In addition, it has been demonstrated that concentrations and/or fault-related emissions of geogenic gases – especially CO2, CH4, He, Rn and H2S – play a decisive role in the settlement of RWA nests (Berberich et al., 2013, 2016, 2018, 2022). One of the main hypotheses explaining the association between RWA nests and fault zones is that the geophysical conditions would favour thermoregulation of the wood ants in relation to temperature anomalies induced by ascending geogases of deep reaching fracture zones (Schreiber et al., 2009).

18At the scale of the study site, there is no clear spatial pattern between observed nests and mapped faults (fig. 7). However, at the intersection zone between the NW-SE (N120°) and NE-SW (N40°) trending faults mapped in the northern part of the study site, a cluster of RWA mounds occurs, and a linear alignment of nests can be observed along the NE-SW fracture zone running in the eastern part of the forest (fig. 7). Other nest alignments following NW-SE directions at the local scale may also reflect secondary faults connected with the South Armorican Shear Zone (Rolin et al., 2000). In addition, the granite massif covering the study site is mapped by IRSN (Institute for Radiation Protection and Nuclear Safety) as having a high natural Radon potential (https://www.irsn.fr/​savoir-comprendre/​environnement/​cartographie-potentiel-radon-formations-geologiques). Recent studies investigating the interactions between ant biology and geological parameters showed that the presence of RWA nests is strongly correlated with Radon emission from Variscan granites, as part of the radioactive decay of unstable Uranium isotopes (Berberich et al., 2016, 2018, 2022). The heat generated by the radioactive decay of U and Th could contribute to the thermoregulation of nests during winter hibernation: the continuous emission of heat from Rn-rich granites, acting as an internal heat source, may promote the survival of the colony, especially for nascent and small mounds that are unable to warm themselves and regulate their own temperature (Berberich et al., 2022). Dense fracturing of the granite massif probably acts as an additional factor favouring Rn degassing and nest clustering on or near fracture zones. This hypothesis could be further tested against more in-depth geostatistical analyses combined with gas analyses from soil air in this site and on additional study sites of the seismically-active Armorican Massif.

4.2. Climate change and evolution of RWA mound locations and volumes

19Ants of the Formica rufa group are considered as highly vulnerable species to climate change (Gouraud, 2021; Sorvari et al., 2021; Fürjes-Mikó et al., 2025). Changes in temperature and precipitation are known to cause stress in RWA colonies, and prolonged drought are harmful for the species. Increased winter precipitation could affect the humidity of nests and, consequently, their health by promoting the development of pathogens associated with increased moisture. Shorter and milder winters could disrupt the phenology of colonies, e.g., by desynchronising the resumption of worker activity with the food availability offered by the forest environment (presence of prey, aphids) (Gouraud, 2021). Higher winter temperatures have also been shown to increase worker mortality during winter, and this effect was almost twice as significant for colonies that had recently experienced forest clear-cutting (Sorvari et al., 2021). Furthermore, climate change is responsible for an increase in natural hazards and disturbances such as the frequency of storms and forest fires, as well as the development of pathogens that are harmful to the survival of colonies (Gouraud, 2021). Habitat loss or deterioration due to climate change is considered an additional factor threatening RWA populations, with increased risk of local or regional extinction (Fürjes-Mikó et al., 2025).

20In northwest France, RWA populations at the study site reach the southern distribution range limit of the Formica rufa species, increasing their susceptibility to climate change. However, results of the 5-year monitoring study showed the relative stability and good health of RWA colonies, as reflected by the evolution of biovolumes and necrovolumes, despite the severe drought of the year 2022. One hypothesis to explain the resilience of ant colonies in this context could be found in the geomorphological heterogeneity of the forest massif. During the year 2022, new mounds were formed on the cooler north-facing slopes of the massif (especially on transect no. 11, where no mound was observed before 2022), but they were largely deserted during the next year marked by cooler and wetter weather conditions. The north-facing slopes could thus act as microclimatic refuges for wood ants during very hot and dry years, suggesting that the geomorphological heterogeneity of the site could be an essential factor that would favour the adaptation of Formica rufa to climate change. A further aspect could be that fault and fracture zones provide higher humidity near the surface, which ants can take advantage of during very dry years (Schreiber and Berberich, 2011).

21Finally, the combination of mixed-conifer forest stands, fractured Variscan granites with high Rn potential and a high geomorphological heterogeneity appears as the main environmental factors controlling the spatial distribution of RWA nests in the study area as well as their resilience over time. These interpretations largely coincide with the conclusions of a recent study conducted by Berberich et al. (2022) in a wide area of NE Bavaria (Germany). However, new data are needed to verify our hypotheses at the study site, for example by installing microclimate sensors near monitored mounds in different topoclimatic positions to test the sensitivity of RWA to interannual variability.

4.3. Conservation issues and ecological implications

22RWA species are increasingly threatened due to several reasons (climate change, habitat loss and degradation, inadequate forest management, intensive agriculture on the edges of woodlands, lack of protection status), with growing evidence of local declines and extinctions in France and Europe (e.g., Mabelis and Korczyńska, 2016; Marage et al., 2017; Balzani et al., 2022). As keystone species and ecosystem engineers, they fulfil many ecological functions and provide a range of ecosystem services (regulation of forest pests, RWA nests as habitats for many other species, ants as a food resource for vertebrates, soil nutrient cycling, seed dispersal, etc.), so conservation issues are of major concern (Gouraud, 2021; Balzani et al., 2022; Fürjes-Mikó et al., 2025). In Northwest France, RWA colonies of the Bois-des-Jarries site appear as the most important population of Formica rufa in the Pays de la Loire region (after the ERFA continued inventory; Gouraud, 2020), with 767 mounds distributed over a forest area of 62 ha, and mean densities of 12.4 nests/ha. The ancientness of the forest, attested at least since the 18th century, could indicate the relict status of the population, today very isolated and threatened by the landscape fragmentation. The Vendée Department, which owns and co-manages the site with the ONF, therefore has a high biological responsibility for the conservation of this population, especially since the species does not benefit from any regulatory protection status at regional, national or European level.

23To assist site managers and provide direction for forestry management by better addressing the issue of RWA mounds, a method for assessing sensitivity at the forest subplot level (i.e., the forest management unit) has been developed. The sensitivity map obtained (fig. 11) is the result of cross-referencing two parameters: (i) nest densities and (ii) the percentage of large nests. Thus, the most sensitive subplots are those with both the highest nest densities and the highest proportion of large mounds out of the total number of mapped nests. Sensitivity levels are globally higher in the northern part of the forest massif, with three sub-plots showing very high sensitivity (8.1, 12.1 and 13.3). The challenges of conserving ant populations at the study site are not limited to the Formica rufa colonies. A myrmecological inventory carried out during the years 2023 and 2024 allowed us to list 20 ant species, including the discovery of Temnothorax pardoi, a very rare small yellow ant found in the dry heaths and considered endemic to the European Atlantic façade (Gouraud et al., 2021). The presence of this vulnerable species therefore represents an additional conservation challenge for the site.

Fig. 11 – Sensitivity map to the “red wood ant” issue at the scale of the forest management unit on the study site.
Fig. 11 – Carte de sensibilité à l’enjeu « fourmis rousses des bois » à l’échelle de l’unité de gestion sur le site d’étude.

Fig. 11 – Sensitivity map to the “red wood ant” issue at the scale of the forest management unit on the study site.Fig. 11 – Carte de sensibilité à l’enjeu « fourmis rousses des bois » à l’échelle de l’unité de gestion sur le site d’étude.

24One key ecological function of mound-building RWA is the role of their nests in supporting a highly diverse community of myrmecophiles (i.e., “guests of ants”), some of which are obligate mutualists that cannot survive outside RWA nests (Robinson et al., 2016). By increasing local geodiversity and soil nutrient content inside the nest, RWA mounds provide species-rich microhabitats hosting many myrmecophilous, commensal, mutualistic or parasitic organisms, including some rare and vulnerable species listed on the IUCN Red List (Balzani et al., 2022). This reinforces the importance of conserving RWA mounds which can be viewed as a biogeomorphological heritage with numerous scientific and added values (Bétard, 2022). In the study site, the observed presence of the rare ladybird Coccinella magnifica illustrates this close association between Formica rufa mounds and myrmecophiles. In their review, Parmentier et al. (2014) listed 125 obligate myrmecophilous species, including 52 beetles, 15 hymenopterans, 10 dipterans, 7 hemipterans, and 36 arachnids, to which numerous facultative guests must be added. Therefore, there are still a large number of myrmecophiles (especially in the Coleoptera Order) to be recorded on the site, particularly inside large mounds or in their immediate vicinity. Following the protocol defined by Parmentier and Claus (2019), a specific inventory of coleopteran myrmecophiles could be carried out by collecting nest material at appropriate times and depths in RWA mounds, likely to reveal a high number of regionally-novel or rare species.

5. Conclusion

25This 5-year monitoring study of Formica rufa mounds in NW France provides a better understanding of the distribution and dynamics of RWA populations and related biogenic landforms in a context of lowland forest landscape and temperate oceanic climate. It adds to the rare research studies on the Formica rufa group in French lowland forests, where their often sparse and scattered populations are considered vestigial (Nageleisen, 1999). The proposed method of monitoring based on strip transects, adapted from Torossian (1979), proved to be relatively simple to implement, low time-consuming and non-destructive. This type of zoogeomorphological survey makes RWA mounds a valuable bioindicator of the health and functioning of forest ecosystems, while also providing a relevant insight into the evolution of dynamic landforms constructed by ant communities. The results of this study indicate both the relative stability and good health of RWA populations over the five years of monitoring, but the status of the species remains precarious and vulnerable in the context of climate change and forest fragmentation that characterizes the Vendée region. The geomorphological heterogeneity of the forest massif is one of the environmental factors currently enabling Formica rufa colonies to adapt to interannual climate variability, particularly thanks to the presence of north-facing slopes that act as microclimatic refuges during hot and dry years. In addition, the exhaustive inventory and mapping of RWA mounds provided a more accurate picture of the density and distribution of nests across the forest massif and highlighted a number of key environmental variables (types and ages of forest stands, slope gradient, geology). The indicators obtained, in terms of mound density and percentage of large nests, were used to produce a sensitivity map at the forest subplot scale, which enabled the ‘red wood ant’ issue to be better considered in the silvicultural and ecological management of the site. Avenues for future research include (i) the study of relationships between tectonic faults, geogenic gases and the distribution of RWA mounds at the scale of the study site and, possibly, at the scale of the Armorican Massif based on citizen science data, (ii) the use of Sfm (Structure from motion) photogrammetry to improve the accuracy for estimating mound volumes and their morphological changes over time, and (iii) the systematic inventory of myrmecophilous organisms in and around large nests, likely to reveal an unexpected biodiversity associated with RWA mounds.

*Correspondence: francois.betard@sorbonne-universite.fr

Haut de page

Bibliographie

Balzani P., Dekoninck W., Feldhaar H., Freitag A., Frizzi F., Frouz J., Masoni A., Robinson E., Sorvari J., Santini G. (2022) – Challenges and a call to action for protecting European red wood ants. Conservation Biology, 36(6), e13959. https://conbio.onlinelibrary.wiley.com/doi/10.1111/cobi.13959

Berberich G.M., Berberich M.B., Gibhardt M. (2022) – Red wood Ants (Formica rufa-group) prefer mature pine forests in Variscan granite environments (Hymenoptera: Formicidae). Fragmenta entomologica, 54(1), 1–18. https://rosa.uniroma1.it/rosa02/fragmenta_entomologica/article/view/474

Berberich G.M., Ellison A.M., Berberich M.B., Grumpe A., Becker A., Wöhler C. (2018) – Can a red wood-ant nest be a trap for fault-related CH4 micro-seepage? A case study from continuous short-term in-situ sampling. Animals, 8, 46. DOI : 10.3390/ani8040046

Berberich G.M, Grumpe A., Berberich M., Klimetzek D., Wöhler C. (2016) – Are red wood ants (Formica rufa-group) tectonic indicators? A statistical approach. Ecological Indicators, 61, 968–979. DOI : 10.1016/j.ecolind.2015.10.055

Berberich G.M., Klimetzek D., Paraschiv M., Stancioiu P.T., Grumpe A. (2019) – Biogeostatistics confirm: Even a low total number of red wood ant nests provide new information on tectonics in the East Carpathian Orogen (Romania). Ecological Indicators, 101, 486–500. DOI : 10.1016/j.ecolind.2019.01.005

Berberich G.M., Schreiber U. (2013) – GeoBioScience: Red wood ants as bioindicators for active tectonic fault systems in the West Eifel (Germany). Animals, 3(2), 475–498. DOI : 10.3390/ani3020475

Bétard F. (2010) – Uplift and denudation history at low-elevation passive margins: Insights from morphostratigraphic analysis in the SE Armorican Massif along the French Atlantic margin. Comptes Rendus Geoscience, 342, 215–222. DOI : 10.1016/j.crte.2010.02.004

Bétard F. (2021) – Insects as zoogeomorphic agents: An extended review. Earth Surface Processes and Landforms, 46(1), 89–109. DOI : 10.1002/esp.4944

Bétard F. (2022) – Anthills and termite mounds as a biogeomorphological heritage. 10th International Conference on Geomorphology, Coimbra, Portugal, 12-16 September 2022, ICG2022-5. https://meetingorganizer.copernicus.org/ICG2022/ICG2022-5.html

Boudjema G., Lempérière G., Deschamps-Cottin M., Molland D.G. (2006) – Analysis and nonlinear modeling of the mound-building ant Formica lugubris spatial multi-scale dynamic in a larch-tree stand of the southern French Alps. Ecological Modelling, 190(1-2), 147–158. DOI : 10.1016/j.ecolmodel.2005.03.021

Branner J.C., Reid H.F. (1900) – Ants as geologic agents in the tropics. The Journal of Geology, 8(2), 151–153.

Butler D.R. (1995) – Zoogeomorphology: animals as geomorphic agents. Cambridge University Press, 240 p.

Del Toro I., Berberich G., Ribbons R.R., Berberich M.B., Sanders N.J., Ellison A.M. (2017) – Nests of red wood ants (Formica rufa-group) are positively associated with tectonic faults: a double-blind test. PeerJ, 5, e3903. DOI : 10.7717/peerj.3903

Dupuis J., Verger F. (1964) – Les microreliefs dus aux fourmis et leurs caractères pédologiques. Norois, 41, 5–15. DOI : 10.3406/noroi.1964.7225

Frouz J., Jílková V., Sorvari J. (2016) – Contribution of wood ants to nutrient cycling and ecosystem function. In Stockan J., Robinson E.J. (Eds.): Wood ant ecology and conservation. Cambridge University Press, Cambridge, 207–220.

Fürjes-Mikó Á., Csősz S., Paulin M.J., Csóka G. (2025) – The Role of Red Wood Ants (Formica rufa Species Group) in Central European Forest Ecosystems—A Literature Review. Insects, 16(5), 518. DOI : 10.3390/insects16050518

Godard G., Chevalier M., Bouton P., Mouroux B. (1994) – Un fleuve Yprésien du Berry à la Vendée, témoin de l’évolution paléogéographique et tectonique du Centre-Ouest de la France au Cénozoïque. Géologie de la France, 4, 35–56.

Gouraud C. (2020) – Enquête sur la Répartition des Fourmis Armoricaines. Bilan 2019. GRETIA (Rennes) et Antarea, 19 p.

Gouraud C. (2021) – Les fourmis rousses des bois (Formica rufa Linnaeus, 1761 & Formica polyctena Foerster, 1850 : Hymenoptera, Formicidae). Écologie et enjeux de conservation en Bretagne. Penn ar Bed, 241–242, 47-58.

Gouraud C., Raitière W., Galkowski C. (2021)Temnothorax pardoi (Tinaut, 1987) en Bretagne, une espèce sténotope inféodée aux landes ? Invertébrés Armoricains, 22, 11–20.

Lemperiere G., Bourbon G., Buray A., Franchini S. (2002) – Étude des populations de fourmis rousses dans cinq sites du bassin de Gap-Chaudun (Hautes-Alpes). Revue forestière française, 54(5), 419–428. DOI : 10.4267/2042/4931

Lenoir L. (2001) – Wood ants (Formica spp.) as ecosystem engineers and their impact on the soil animal community. Doctor’s dissertation, Swedish University of Agricultural Sciences, Uppsala, 36 p.

Mabelis A.A., Korczyńska J. (2016) – Long-term impact of agriculture on the survival of wood ants of the Formica rufa group (Formicidae). Journal of Insect Conservation, 20(4), 621–628. DOI : 10.1007/s10841-016-9893-7

Marage D., Lempérière G., Voreux C. (2017) – Trente ans plus tard… les fourmis rousses des bois, toujours un bon indicateur de l’état de conservation des forêts de montagne ? Revue forestière française, 69(2), 111120. DOI : 10.4267/2042/64090

Nageleisen L.M. (1999) – Étude de la densité et du rôle bioindicateur des fourmis rousses dans les forêts du nord-est. Revue forestière française, 51(4), 487495. DOI : 10.4267/2042/5457

ONF (2019) – Aménagement de la forêt départementale du Bois des Jarries 2019-2038. Office National des Forêts, Agence des Pays de la Loire, 28 p.

Parmentier T., Claus R. (2019) – A dazzling number of beetles in a hibernating red wood ant nest. The Coleopterists Bulletin, 73(3), 1–4. DOI : 10.1649/0010-065X-73.3.761

Parmentier T., Dekoninck W., Wenseleers T. (2014) – A highly diverse microcosm in a hostile world: a review on the associates of red wood ants (Formica rufa group). Insectes Sociaux, 61(3), 229–237. DOI : 10.1007/s00040-014-0357-3

Robinson E.J., Stockan J., Iason G.R. (2016) – Wood ants and their interaction with other organisms. In: Stockan J., Robinson E.J. (Eds.): Wood ant ecology and conservation. Cambridge University Press, Cambridge, 177–206.

Rolin P., Audru J.C., Bouroullec I., Wyns R., Thiéblemont D., Cocherie A., Guerrot C., Courtois N., Bernard E., Chantraine J., Lambert J., Terrier-Sedan M. (2000)Notice explicative, Carte géol. France (1/50 000), feuille Les Herbiers (537). Orléans, BRGM, 117 p.

Rouillon A., Longuet D. (1991) – Le site mésolithique du Bois des Jarries (Saint-Mars-La-Réorthe, Vendée). Revue archéologique de l’Ouest, 8(1), 47–59.

Schreiber U., Berberich G.M. (2011) – Red wood ant mounds as biological indicators for earthquake-bearing fault systems. EGU General Assembly 2011, Geophysical Research Abstracts, Vol. 13, EGU2011-1790.

Schreiber U., Brennholt N., Simon J. (2009) – Gas permeable deep reaching fracture zones encourage site selection of ants. Ecological Indicators, 9(3), 508–517. DOI : 10.1016/j.ecolind.2008.07.002

Sorvari J. (2022) – Biogeography and habitat preferences of red wood ants of the Formica rufa group (Hymenoptera: Formicidae) in Finland, based on citizen science data. European Journal of Entomology, 119, 92–98. DOI : 10.14411/eje.2022.010

Sorvari J., Haatanen M.K., Vesterlund S.R. (2011) – Combined effects of overwintering temperature and habitat degradation on the survival of boreal wood ant. Journal of Insect Conservation, 15, 727–731. DOI : 10.1007/s10841-010-9372-5

Torossian C. (1979) – Méthode d'étude quantitative des fourmis du groupe Formica rufa. Bulletin SROP, 2(3), 215-240.

Torossian, C., Humbert, P. (1982) – Les fourmis rousses des bois et leur rôle dans l'écosystème forestier. Revue forestière française, 34(1), 32–41. DOI : 10.4267/2042/21554

Viles H.A., Goudie A.S., Goudie A.M. (2021) – Ants as geomorphological agents: A global assessment. Earth-Science Reviews, 103469. DOI : 10.1016/j.earscirev.2020.103469

Whitford W.G., Eldridge D.J. (2013) – Effects of ants and termites on soil and geomorphological processes. In Shroder J., Butler D.R., Hupp C.R. (Eds.): Treatise on Geomorphology, Vol. 12, Ecogeomorphology. Academic Press, San Diego, 281–292.

Haut de page

Annexe

Version abrégée en français

Les fourmis rousses des bois (groupe Formica rufa) jouent un rôle fonctionnel essentiel au sein des forêts tempérées et boréales d’Europe (Frouz et al., 2016 ; Sorvari, 2022 ; Balzani et al., 2022 ; Fürjes-Mikó et al., 2025). Leur qualité de bioindicateur de l’état de conservation des écosystèmes forestiers est reconnue depuis longtemps, en particulier depuis les travaux de Claude Torossian (Torossian, 1979 ; Torossian et Humbert, 1982). En revanche, leur rôle comme agent géomorphologique a rarement été abordé en tant que tel (Bétard, 2021 ; Viles et al., 2021). Pourtant, en tant qu’espèces ingénieures, les fourmis rousses modifient leur habitat en creusant des galeries souterraines et en édifiant des dômes en surface, avec des densités pouvant atteindre plusieurs dizaines de nids par hectare (fig. 1). Ces dômes représentent des microreliefs biogéniques pouvant perdurer pendant plusieurs décennies, bien qu’en constante évolution. L’objectif de cette étude zoogéomorphologique est d’examiner la distribution spatiale et l’évolution temporelle des dômes de Formica rufa dans un massif forestier de basse altitude et de milieu tempéré océanique, à travers un suivi réalisé sur cinq ans (2020-2024).

Localisé dans le nord-ouest de la France, le site d’étude correspond au Bois des Jarries, une forêt départementale de 62 hectares située dans le Haut-Bocage vendéen (fig. 2). Son statut de forêt ancienne est attesté depuis au moins le XVIIIe siècle (carte de Cassini). Aujourd’hui propriété du département de la Vendée, le site est classé Espace Naturel Sensible depuis 1998 et sa gestion relève du régime forestier. Occupant deux collines granitiques dont la plus haute culmine à 262 m, le boisement est composé majoritairement de châtaigniers, de chênes pédonculés et de chênes sessiles gérés en taillis ou futaie sur souche (ONF, 2019). Ces peuplements de feuillus sont diversifiés par la présence de résineux, notamment de pins sylvestres, de pins maritimes et de cèdres de l’Atlas, ainsi que par une mosaïque de landes sèches à bruyères qui présentent un fort intérêt écologique et patrimonial (tab. 1).

L’étude qualitative et quantitative des dômes de fourmis rousses, conduite sur une période de 5 ans (2020-2024), comporte deux volets méthodologiques complémentaires : (i) un relevé GPS exhaustif des dômes sur les 62 ha du massif forestier, couplé à une analyse spatiale sous SIG, permettant le calcul de densités et un croisement avec différentes variables environnementales (types de peuplements forestiers, pente, exposition, géologie) ; (ii) un suivi morphométrique pluriannuel des dômes le long de 13 transects par bande (fig. 3) avec différents paramètres relevés et mesurés sur le terrain chaque année à la même période (grand et petit diamètres, hauteur, matériaux du nid, présence de bois mort, etc.), ces différentes mesures permettant de calculer une batterie d’indicateurs (pseudobiovolume, nécrovolume, biovolume total, biovolume moyen) et de classer les dômes selon leur taille (petit dôme < 32 dm3 ; dôme moyen entre 32 dm3 et 256 dm3 ; gros dôme entre 256 dm3 et 2048 dm3 ; très gros dôme > 2048 dm3).

L’inventaire GPS exhaustif des dômes de fourmis rousses a permis de recenser 767 nids sur le périmètre de la forêt départementale, avec des densités pouvant atteindre 20 nids/ha dans la partie centrale du boisement composée de peuplements mixtes feuillus-résineux (fig. 4, tab. 2). Elles restent élevées dans les parcelles de feuillus anciennes et diversifiées (les peuplements monospécifiques de châtaigniers sont clairement défavorables : fig. 5). Topographiquement, la concentration des dômes est plus importante sur pentes faibles ou nulles que sur pentes fortes (fig. 6., tab. 3) mais ne montre pas de relation significative avec l’exposition des versants. La géologie semble jouer un rôle important (fig. 7) : les granites arénisés apparaissent favorables au creusement des galeries souterraines et à l’édification des nids, tandis que les dômes de fourmis rousses sont absents sur les lithologies environnantes (schistes, cornéennes). La fréquence d’alignements de dômes orientés selon des directions tectoniques connues dans la région (N°120 et N°40) pourrait indiquer une relation possible avec un réseau orthogonal de failles armoricaines.

Le suivi morphométrique des dômes durant 5 années consécutives indique de fortes variations de la taille et de l’activité des dômes à l’échelle des transects étudiés (fig. 8). Les biovolumes moyens calculés montrent des valeurs plus élevées dans les parcelles de peuplements mixtes que dans les parcelles de feuillus (fig. 9). Ceci s'explique principalement par les matériaux disponibles et adaptés à la construction des dômes (aiguilles de pin), l'éclairage (peuplements forestiers plus ouverts favorisant un ensoleillement partiel) et la ressource alimentaire disponible (miellat excrété par les pucerons des aiguilles et autres insectes suceurs de sève). Les variations climatiques interannuelles expliquent en bonne partie les variations de l’activité des nids sur la période étudiée, les années chaudes et sèches (comme 2022) ayant un impact notable sur cette activité (fig. 10). Malgré cela, les résultats qui découlent de la présente étude indiquent à la fois la relative stabilité et la bonne santé des populations de fourmis rousses sur les 5 années du suivi, même si le statut de l’espèce demeure précaire et vulnérable dans le contexte de fragmentation forestière qui caractérise le Bocage vendéen et dans le contexte du changement climatique qui constitue une menace à moyen ou long terme pour des populations situées ici en limite d’aire. L’hétérogénéité géomorphologique du massif des Jarries apparaît comme l’un des facteurs pouvant permettre aux populations de Formica rufa de s’adapter à la variabilité climatique interannuelle, notamment grâce à la présence de versants exposés au nord qui jouent le rôle de microrefuges climatiques lors des années chaudes et sèches.

D’un point de vue opérationnel, les indicateurs obtenus, en termes de densités de nids et de pourcentage de gros dômes, ont permis de produire une carte de sensibilité à l’échelle de l’unité de gestion forestière (fig. 11), permettant de mieux prendre en compte l’enjeu « fourmis rousses » dans la gestion sylvicole et écologique du site. Parmi les perspectives de recherche future, l’analyse des relations entre la distribution des dômes et la trame tectonique constitue une piste intéressante à explorer, compte tenu des travaux récents ayant montré l’influence des failles actives et des gaz géogéniques dans les concentrations de nids de Formica rufa dans diverses régions d'Europe centrale et orientale (e.g., Schreiber et al., 2009 ; Berberich et Schreiber, 2013 ; Berberich et al., 2016 ; Del Toro et al., 2017 ; Berberich et al., 2019). Une autre perspective prometteuse est la réalisation d’un inventaire systématique des organismes myrmécophiles dans et autour des plus gros nids, susceptible de révéler une biodiversité extraordinaire associée aux dômes de fourmis rousses.

Haut de page

Table des illustrations

Titre Fig. 1 – Red wood ants and related biogenic landforms. A: worker of Formica rufa (photo: Michal Kukla, CC BY-SA 4.0); B: anthills created by Formica rufa in the study site (Bois des Jarries, Vendée, NW France); the mound in the foreground is  1 m high and is aligned with two other smallest mounds.Fig. 1 – Les fourmis rousses des bois et leurs constructions biogéomorphologiques. A : ouvrière de Formica rufa (photo : Michal Kukla, CC BY-SA 4.0) ; B: dômes construits par Formica rufa sur le site d’étude (Bois des Jarries, Vendée, France) ; le dôme au premier plan fait  1 m de hauteur et est aligné avec deux autres dômes plus petits.
Légende (photo: François Bétard).(photo : François Bétard).
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-1.jpg
Fichier image/jpeg, 303k
Titre Tab. 1 – Synthetic description of the 12 forest stands of the study site (after ONF, 2019).Tab. 1 – Description synthétique des 12 types de peuplements forestiers sur le site d’étude (d’après ONF, 2019).
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-2.jpg
Fichier image/jpeg, 167k
Titre Fig. 2 – Location of the study site. A: hypsometric map from the RGEAlti 1 m DTM (IGN/Géoservices); equidistance of contour lines 5 m; B: aerial photograph showing the studied forest massif isolated in an agricultural-dominated hedgerow landscape (IGN/Géoportail); the photograph was acquired on 10 May 2022; C: geological map at 1:50,000 scale showing the main lithologies (BRGM/Infoterre).Fig. 2 – Localisation du site d’étude. A : carte hyspométrique dérivée du MNT RGE Alti 1 m (IGN/Géoservices) ; équidistance des courbes de niveau 5 m ; B : photographie aérienne montrant le massif forestier étudié, isolé dans un paysage de bocage majoritairement agricole (IGN/Géoportail) ; la photographie a été acquise le 10 mai 2022 ; C : carte géologique à l’échelle 1/50 000 indiquant les principales lithologies rencontrées dans la zone d’étude (BRGM/Infoterre).
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-3.jpg
Fichier image/jpeg, 750k
Titre Fig. 3 – Forest stands and sampling plan locating the 13 strip transects at the study site (map base from ONF, 2019).Fig. 3 – Types de peuplements forestiers et plan d’échantillonnage localisant les 13 transects par bande sur le site d’étude (fond cartographique d’après ONF, 2019).
Légende 1: Scots pines with chestnuts; 2: Atlas cedars; 3: native oaks; 4: oaks with chestnuts; 5: chestnuts; 6: chestnuts with oaks; 7: native oaks; 8: oaks and chestnuts; 9: chestnuts; 10: chestnuts and maritime pines; 11: chestnuts and Scots pines; 12: dry heaths and stunted oak coppice on skeletal soils; 13: forest paths; 14: plot limits; 15: monitored transect. Numbers with white outline are plot numbers; numbers in italics are subplot numbers.1 : pin sylvestre sur châtaignier ; 2 : cèdre de l’Atlas ; 3 : chênes indigènes ; 4 : chênes et châtaignier ; 5 : châtaignier ; 6 : châtaignier et chênes ; 7 : chênes indigènes ; 8 : chênes et châtaignier ; 9 : châtaignier ; 10 : châtaignier et pin maritime ; 11 : châtaignier et pin sylvestre ; 12 : lande sèche et taillis rabougri sur sol rocheux superficiel ; 13 : sentiers forestiers ; 14 : limite de parcelle forestière ; 15 : transects étudiés. Les numéros sur fond blanc sont les numéros de parcelles forestières ; les numéros en italique sont ceux des sous-parcelles.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-4.jpg
Fichier image/jpeg, 593k
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-5.jpg
Fichier image/jpeg, 9,1k
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-6.jpg
Fichier image/jpeg, 8,4k
Titre Fig. 4 – Spatial distribution of RWA mounds according to the main forest stand types at the study site.Fig. 4 – Distribution spatiale des dômes de Formica rufa selon les principaux types de peuplements forestiers sur le site d’étude.
Légende 1: stumped-planted high forest of oaks and chestnut; 2: mixed conifer regularly planted high forest; 3: monospecific coppice of chestnut; 4: coppice of oaks and/or chestnut; 5: long-term unmanaged oak coppice; 6: clear-cutting in 2017; 7: clear-cutting in 2021; 8: very large mound; 9: large mound; 10: medium mound; 11: small mound.1 : futaie sur souche de chênes et châtaignier ; 2 : peuplement mixte résineux-châtaignier ; 3 : taillis de châtaignier monospécifique ; 4 : taillis de chênes et/ou châtaignier ; 5 : taillis de chênes laissé en libre évolution ; 6 : taillis récent (coupe rase en 2017) ; 7 : taillis récent (coupe rase en 2021) ; 8 : très gros dôme ; 9 : gros dôme ; 10 : dôme moyen ; 11 : petit dôme.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-7.jpg
Fichier image/jpeg, 388k
Titre Tab. 2Descriptive statistics of expected numbers of RWA nests, mapped nests and mound densities according to the main forest stand types. Tab. 2 – Statistiques descriptives sur les nombres de nids attendus, les nids inventoriés et les densités de nids selon le type de peuplement forestier.
Légende Expected nest values are obtained by multiplying the total number of nests (n = 767) by the percentage of the area of each class.Les valeurs de nids attendues sont obtenues en multipliant le nombre total de nids (n = 767) par le pourcentage de la superficie de chaque classe.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-8.jpg
Fichier image/jpeg, 89k
Titre Fig. 5 – Two adjacent forest stands in the northern part of the study site, which are very different in terms of their suitability for RWA nests. A: coppice of native oak trees with diverse undergrowth vegetation, conducive to the establishment of numerous ant nests; B: coppice of monospecific chestnut trees, with very poor undergrowth vegetation, an environment unfavorable to the establishment of Formica rufa.Fig. 5 – Deux peuplements forestiers mitoyens dans la partie nord du boisement, très contrastés en termes d’accueil de dômes de fourmis rousses. A : taillis de chênes indigènes avec une végétation diversifiée de sous-bois, propice à l’implantation de nombreuses fourmilières ; B : taillis de châtaigniers monospécifiques, très pauvre au niveau du sous-bois, un milieu défavorable à l’installation de Formica rufa.
Légende Photos: François Bétard, 15 April 2024.Photos : François Bétard, 15 avril 2024.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-9.jpg
Fichier image/jpeg, 553k
Titre Fig. 6 – Spatial distribution of RWA mounds according to the slope gradient at the study site.Fig. 6 – Distribution spatiale des dômes de Formica rufa selon la pente (en °) sur le site d’étude.
Légende 1: very large mound; 2: large mound; 3: medium mound; 4: small mound; 5: ENS perimeter.1 : très gros dôme ; 2 : gros dôme ; 3 : dôme moyen ; 4 : petit dôme ; 5 : périmètre ENS.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-10.jpg
Fichier image/jpeg, 672k
Titre Tab. 3 - Descriptive statistics of expected numbers of RWA nests, mapped nests and mound densities according to the slope gradient. Tab. 3 – Statistiques descriptives sur les nombres de nids attendus, les nids inventoriés et les densités de nids selon la classe de pente.
Légende Expected nest values are obtained by multiplying the total number of nests (n = 767) by the percentage of the area of each class.Les valeurs de nids attendues sont obtenues en multipliant le nombre total de nids (n = 767) par le pourcentage de la superficie de chaque classe.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-11.jpg
Fichier image/jpeg, 76k
Titre Fig. 7 – Spatial distribution of RWA mounds according to the main lithologies and possible relationships with tectonic faults at the study site.Fig. 7 – Distribution spatiale des dômes de Formica rufa selon la lithologie et relations possibles avec la trame tectonique sur le site d’étude.
Légende 1: leucogranite; 2: leuco-monzogranite; 3: hornfels and spotted schists; 4: schists; 5: sandstones; 6: mapped fault (BRGM); 7: inferred fault (this study); 8: very large mound; 9: large mound; 10: medium mound; 11: small mound; 12: ENS perimeter.1 : leucogranite ; 2 : leuco-monzogranite ; 3 : cornéennes et schistes tachetés ; 4 : schistes ; 5 : grès ; 6 : faille cartographiée (BRGM) ; 7 : faille interprétée (cette étude) ; 8 : très gros dôme ; 9 : gros dôme ; 10 : dôme moyen ; 11 : petit dôme ; 12 : périmètre ENS.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-12.jpg
Fichier image/jpeg, 417k
Titre Fig. 8 – Variations of size and activity of RWA mounds along two strip transects (50 x 20 m) during five consecutive years (between 2020 and 2024). Fig. 8 – Variations de la taille et de l’activité des dômes de fourmis rousses le long de deux transects par bande (50 x 20 m) durant cinq années consécutives (entre 2020 et 2024).
Légende Diameters are in cubic root of volume. Dark: active mounds; light gray: abandoned mounds. Note the aligned mounds along known tectonic directions in the region.Les diamètres sont exprimés en racine cubique du volume. Noir : dômes actifs ; gris clair : dômes abandonnés. Noter les alignements de nids selon des directions tectoniques connues dans la région.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-13.jpg
Fichier image/jpeg, 442k
Titre Fig. 9 – Comparative evolution of biovolumes of RWA mounds according to the two main forest stand types, between 2020 and 2024, on the 13 monitored transects.Fig. 9 – Évolution comparée du biovolume des dômes de fourmis rousses selon les deux principaux types de peuplement forestier, entre 2020 et 2024, sur les 13 transects étudiés.
Légende 1. Mixed-conifer forest stands; 2. Deciduous forest stands.1. Peuplements mixtes résineux-feuillus ; 2. Peuplements feuillus.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-14.jpg
Fichier image/jpeg, 93k
Titre Fig. 10 – Relationships between interannual climate variability and nest activity of RWA mounds during the 5-year monitoring study on the 13 strip transects. A: Average temperatures (°C) and total precipitation (mm) at the meteorological station of Les Herbiers (located at 8 km from the study site) between March and July corresponding to the period of activity of red wood ants preceding the surveys (source of climatic data: https://www.mellifere.com/​climat/​85500-les-herbiers.php); B: Level of nest activity during the surveys: N: many workers on the nest; M: moderate activity; R: few workers on the nest; A: abandoned nests. The dotted line represents the percentage of nests abandoned (A) and being abandoned (R).Fig. 10 – Relations entre variabilité climatique interannuelle et niveau d’activité des nids de Formica rufa selon les années sur l’ensemble des 13 transects étudiés. A : Températures moyennes (°C) et précipitations totales (mm) à la station des Herbiers (située à 8 km du site d’étude) entre les mois de mars et juillet correspondant à la période d’activité des fourmis rousses précédant les relevés (source des données climatiques : https://www.mellifere.com/​climat/​85500-les-herbiers.php) ; B : Niveau d’activité des nids lors des relevés : N : nombreuses ouvrières sur le nid ; M : activité moyenne ; R : quelques rares ouvrières sur le dôme ; A : nid abandonné. La ligne pointillée représente la part de nids abandonnés (A) et en cours d’abandon (R).
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-15.jpg
Fichier image/jpeg, 257k
Titre Fig. 11 – Sensitivity map to the “red wood ant” issue at the scale of the forest management unit on the study site.Fig. 11 – Carte de sensibilité à l’enjeu « fourmis rousses des bois » à l’échelle de l’unité de gestion sur le site d’étude.
URL http://journals.openedition.org/geomorphologie/docannexe/image/20529/img-16.jpg
Fichier image/jpeg, 448k
Haut de page

Pour citer cet article

Référence électronique

François Bétard, « Anthills as dynamic landforms: a 5-year monitoring study of red wood ant mounds in a temperate forest landscape of NW France »Géomorphologie : relief, processus, environnement [En ligne], 32-1 | 2026, mis en ligne le 07 novembre 2025, consulté le 09 décembre 2025. URL : http://journals.openedition.org/geomorphologie/20529 ; DOI : https://doi.org/10.4000/1547o

Haut de page

Auteur

François Bétard

Sorbonne Université, UR Médiations, 191 rue Saint-Jacques, 75005 Paris, France

Articles du même auteur

Haut de page

Droits d’auteur

CC-BY-SA-4.0

Le texte seul est utilisable sous licence CC BY-SA 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont susceptibles d’être soumis à des autorisations d’usage spécifiques.

Haut de page
Search OpenEdition Search

You will be redirected to OpenEdition Search