texto en 
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Highlights:
Nests of L. apiculatum were estimated in crassicaule scrub (CS), desert microphyll scrub (DMS) and rosetophyll (RS).
Nest density was 4.8 nests∙ha-1 in CS, 2.2 nests∙ha-1 in DMS and 2.3 nests∙ha-1 in RS.
Slope direction, rock, bare soil, and soil type influenced L. apiculatum nests.
The correlation of nest density with the presence of agaves was weak.
Introduction
Ants are social insects distributed in several ecosystems (Szewczyk & McCain, 2018). One of these species is Liometopum apiculatum Mayr, commonly known as the 'escamolera' ant (escamoles are the larval stages of males and females), a species that has successfully colonized some environments of riparian vegetation, pine forest and deserts in North America (Lara-Juárez et al., 2015). In Mexico, L. apiculatum is distributed in 24 states with diverse ecosystems; for example, deserts and rainforests (Berumen Jiménez et al., 2021; Lara-Juárez et al., 2015). In semi-arid areas, L. apiculatum is located from 1 800 to 3 000 m in subterranean nests (Melo-Ruiz et al., 2016) in crassicule scrub, desert microphyll scrub and rosetophyll scrub (Cruz-Labana et al., 2014; Figueroa-Sandoval et al., 2018). As with other ants, for L. apiculatum, ground nesting provides it with a microhabitat suitable for rearing its larvae and shelters a population of up to 250 000 individuals per nest (Hoey-Chamberlain et al., 2013). The life cycle of nests is divided into four stages: foundation, growth, reproduction and disa[p]earance (Lara, 2013). Nest foundation occurs after the mating flight, in the first rain of March to April; the fertilized females that fall to the ground burrow and oviposit the first generation of workers (Lara-Juárez et al., 2015). A nest of L. apiculatum can reach 40 years of longevity or 12 years on average and produce up to 3.54 kg of escamoles (80 164 individuals) in preserved or well-managed nests; in central Mexico, the collection season is before the mating flight (Lara-Juárez et al., 2015; Ramos-Elorduy et al., 1984).
In San Luis Potosí and Zacatecas, L. apiculatum nests at the base of Agave spp., Opuntia spp. and Yucca spp. (Cruz-Labana et al., 2014; Figueroa-Sandoval et al., 2018; Hernández-Roldan et al., 2017), where hemiptera are found on which the ant feeds via trophobiosis (Cruz-Labana et al., 2018; Lara-Juárez et al., 2015). The habitat where the ant nests has cover of 12.1 to 13.8 % rocky soil, 26 to 56.4 % bare soil, 14.6 to 16.3 % basal area (vegetation), 46.9 to 11.7 % mulch, and edaphic differences and plant physiognomy (favorable and unfavorable conditions) associated with the presence of nests in areas of desert scrub (Lara-Juárez et al., 2016). The interaction of ground covers, nesting plants and food availability (hemiptera) and other habitat variables are important for insect populations (Cruz-Labana et al., 2014; 2018; Hernández-Roldan et al., 2017; Lara-Juárez et al., 2016). For example, deserts have reported 3.8 nests∙ha-1 (Hernández-Roldan et al., 2017) under conserved habitat conditions, 2.25 to 4.14 nests∙ha-1 under disturbed conditions (Figueroa-Sandoval et al., 2018) and even densities of up to 14 nests∙ha-1.
The semi-arid ecosystems of north-central Mexico, where L. apiculatum is distributed, are agricultural and livestock pastures with diversified activities such as Agave spp. harvest for commercial purposes (Hernández-Roldan et al., 2017; Lara-Juárez et al., 2016). In this scenario, escamoles are extracted excessively for culinary preparations (Berumen Jiménez et al., 2021). On the other hand, the collection of L. apiculatum larvae in rural areas of high marginalization represents complementary economic income (Melo-Ruiz et al., 2016), which has caused an overexploitation of this natural resource. This trend is aggravated by other factors such as the absence of legal guidelines for the extraction of escamoles (Lara-Juárez et al., 2015), non-standardized collection practices, and lack of information on the ecology and biology of the ant. Therefore, the objective of this research was to estimate nest density according to vegetation type and to identify habitat variables associated with nest presence. The hypothesis was that nest density is homogeneous for crassicule scrub, desert microphyll scrub and rosetophyll scrub and that the variables related to occurrence may be the presence of agaves and soil type. This research complements the existing information on the ‘escamolera’ ant in north-central region of Mexico, to contribute to its sustainable use.
Materials and Methods
The study area was located in the limits of San Luis Potosí and Zacatecas, northwest of the Mexican Altiplano region. Specifically in the communities of Villas de Ahualulco (22° 26’ 56.53’’ N and 101° 13’ 29.84’’ W), Ipiña (22° 27’ 1.40’’ N and 101° 18’ 35.15’’ W) and Santa Teresa (22° 20’ 9.71’’ N and 101° 21’ 48.70’’ W) in the municipality of Ahualulco in San Luis Potosí; Tolosa (22° 30’ 18.05’’ N and 101° 22’ 4.94’’ W) and Santiago (22° 27’ 49.99’’ N and 101° 27’ 50.06’’) in the municipality of Pinos, and two private properties (22° 37’ 59.664’’ N and 101° 58’ 57.042’’ W and 22° 38’ 16.37’’ N and101° 58’ 7.735’’ W) in the municipality of Villa González Ortega in Zacatecas (Figure 1).
The region is made up of highlands, plains and valleys with altitudes between 1 000 and 2 500 m in the so-called Mesa Central (Instituto Nacional de Estadística y Geografía [INEGI], 2021). The climate is semi-arid with BS-BW classifications, passing through subtypes BS1kw as temperate, BS1 the least dry and BS0 as the driest, with an average annual precipitation of 240 to 770 mm with rainfall in summer (Comisión Nacional del Agua [CONAGUA], 2017; INEGI, 2013). The vegetation corresponds, for the most part, to crassicaule scrub (CS) with species such as Opuntia rastrera F. A. C. Weber, Cylindropuntia imbricata (Haw.) F. M. Knuth, Cylindropuntia tunicata (Lehm.) F. M. Knuth, Yucca carnerosana (Trel.) McKelvey and Yucca filifera Chabaud; desert microphyll scrub (DMS) with colonies of Larrea tridentata (Sessé & Moc. ex DC.) Coville and rosetophyll scrub (RS) with Agave salmiana Otto ex Salm-Dyck, Y. carnerosana and Y. filifera as the most common species (INEGI, 2014a; Zavala-Hurtado & Jiménez, 2020).
The density of L. apiculatum nests according to vegetation type was estimated during January and February 2016 by means of field trips with the support of local guides (escamoles collectors). The activity consisted of recording with a GPS the largest number of nests (subject to escamoles extraction or without extraction), avoiding any type of land preferences (easy access or areas close to villages) to reduce bias. At the end of the field work, 201 nests were identified with their coordinates. With this information and a Geographic Information System (GIS) ArcMap 10.5.1 (Environmental System Research Institute [ESRI], 2017) the spatial points of the nests were created and projected on a vegetation and land use map at a scale of 1: 250 000 (Figure 1).
Based on the distribution map (Figure 1), the vegetation types with the highest number of nests were selected. Using this criterion, the spatial points of the nests were superimposed and gridded grids (200 × 200 m cells) were created, stratifying the study area into the three vegetation types: CS = 87 cells, DMS = 35 cells and RS = 18 cells (Figure 2).
Figure 1 Vegetation types and land use with presence of Liometopum apiculatum nests in north-central Mexico.
Figure 2 Distribution of Liometopum apiculatum nests, cells and sampling transects used for nest density estimates and measurement of habitat variables in three shrubland vegetation types in north-central Mexico.
In December 2016 and January 2017, habitat variables were measured, and nest density was estimated using stratified random sampling (SRS). These activities were performed with the geospatial information in Figure 2, which was transformed to a spreadsheet to count the number of nests per cell and calculate their population variance according to vegetation type. Using SRS with proportional allocation, 95 % probability and 0.2 % standard error (Calderón et al., 2019; García-García et al., 2013), we obtained a total size of 27 cells distributed across vegetation types and a size between strata in CS = 16, DMS = 7 and RS = 4. A 200 m line was drawn to each cell in a random direction (cardinal: N, S, E, W and ordinal: NW, NE, SW and SE; Figure 2). Habitat variables were measured on the 27 lines drawn in the cells, collecting information on 19 habitat variables (Table 1) in circular plots of 20 m in diameter at distances of 0, 50, 100, 150 and 200 m. Sampling was complemented by estimating ground cover using 10 m long Canfield lines established in north, south, east and west directions (Figure 3) (Canfield, 1941; Hernández-Roldan et al., 2017).
Figure 3 Circular plots distributed along 200 m lines for the measurement of Liometopum apiculatum habitat variables according to shrubland vegetation type in north-central Mexico.
Table 1 Liometopum apiculatum habitat variables evaluated in random cells according to vegetation type (crassicaule scrub, desert microphyll scrub and rosetophyll scrub) in north-central Mexico.
| Variable | Evaluation method |
|---|---|
| Type of vegetation (INEGI, 2013) | Stratification based on the highest concentration of nests according to vegetation type |
| Elevation | GPS (GARMIN®, GPSmap® 60CSx) |
| Slope (%) | Clinometer (Suunto®, PM-5/360) |
| Slope direction | Compass (Brunton®, 8099 Eclipse) |
| Type of soil | Overlay of the edaphological series layer (INEGI, 2014b) with the sampling cells |
| Land cover: herbaceous, grasses, bare soil, rock, woody material, shrubs, cacti and rosette plants | 10 m Canfield lines (Canfield, 1941) distributed along the transects at 0, 50, 100, 150 and 200 m in four directions (N, S, E and W) |
| Number of plants Agave spp., Opuntia spp., Yucca spp., Acacia farnesiana (L.) Willd and Prosopis glandulosa Torr. | Plant counts along the Canfield lines established in the circular plots (Ø = 20 m) |
Nest density was estimated along the lines (200 m) where habitat variables were measured (CS = 16, DMS = 7 and RS = 4); 100 m to the left and 100 m to the right of the transect centerline were considered and their distances perpendicular to this line were recorded. Nest density was calculated with Distance 7.2 Release 1 software by selecting the best model, using Akaike's minimum ranking criterion (Buckland et al., 2015).
Habitat variables that had the highest relationship with the presence of escamolera ant nests were identified with a Canonical Correspondence Analysis (CCA) with XLSTAT 2018.5 software (Addinsoft, 2018). This multivariate analysis technique (Liang et al., 2015) allowed the analysis of the correlation between the set of variables (discrete and nominal) and the occurrence data (nests) by creating canonical variables, presented graphically to facilitate the interpretation of the results.
Results and Discussion
Nest density
The nest density of L. apiculatum was 4.8 nests∙ha-1 in CS, 2.2 nests∙ha-1 in DMS and 2.3 nests∙ha-1 in RS (P = 0.05, AIC = 2.72). In this regard, Lara-Juárez et al. (2016) reported densities of 6 nests∙ha-1 in a scrub with favorable edaphic conditions and plant physiology (associated with the number of active nests). The difference in nest density between the two studies can be attributed to sampling strategies. In this research, grids with nests coordinates were used in which they drew random transects, while Lara-Juárez et al. (2016) used the point-centered quarted on the closest plant stratum with nests present; however, the results are not so contrasting, since the method for measuring vegetation is similar (transects), and the land use and ownership are the same (ejido rangeland and private properties). Figueroa-Sandoval et al. (2018) reported 2.25 nests∙ha-1 classified based on their conservation status (preserved vs. disturbed nests), using a modified relative density equation in Pinos and Ahualulco. This result is similar to that observed in DMS (2.2 nests∙ha-1) and RS (2.3 nests∙ha-1); both investigations coincided in data collection in two ejidos (Santiago and Ipiña) with the same vegetation classification (microphilous shcrub, crasicaule shcrub and rosetophilous scrub) (Figure 1).
Differences in L. apiculatum ant nest density among vegetation types may be due to plants regulating the amount and composition of resources available from the soil biota (Kardol et al., 2013). One of these resources is food; for example, Cylindropuntia imbricata (Haw.) has been reported to provide extrafloral nectar to L. apiculatum (Avila-Argáez et al., 2019). It can also be assumed that nest density is affected by trophic interactions with plants (plant- ant). As an example, L. apiculatum controls the population of hemiptera on which it feeds (via trophobiosis), in return, the plants get as a benefit a pest control. In other species this behavior has already been described by Rosumek et al. (2009) and Wurst et al. (2018), who mention that formicides reduce the population of nematodes that feed on plants, so ant-free plants have 50 % more herbivory compared to those that grown with ants.
Predation by arthropods considered pests affects plant yield (reduction in biomass production up to 24 %) with cascading effects on trophic levels (Rosumek et al., 2009; Wurst et al., 2018). Anthropogenic events can cause high levels of hemipteran infestation and as a consequence impair biomass production in plants and modify vegetation structure leading to declines in Formicidae populations. Anthropogenic activities in the semiarid zones of central Mexico are agriculture, livestock, firewood extraction and mescal production, which has modified the vegetation structure and decreased the native species subject to exploitation (Hernández-Roldan et al., 2017; Lara-Juárez et al., 2016; Ríos-Casanova et al., 2012) to the detriment of L. apiculatum populations.
Hernández-Roldan et al. (2017) and Cruz-Labana et al. (2014) estimated the density of escamolera ant nests (presence records) in transects of CS, DMS and RS vegetations in semi-desert areas and two land uses (private property and ejido), the authors modeled the data with the program Distance (Buckland et al., 2015) and measured habitat variables in circular plots. Hernández-Roldan et al. (2017) reported 5.5 nests∙ha-1 in an area near the private properties where data were taken for this research (Unidad de Manejo El Milagro, 4.5 to 5.0 km to the southeast) and documented that nest density was associated with CS, agave density, and shrub density. In this context, Cruz-Labana et al. (2014) estimated a density of 6.8 nests∙ha-1 in Charcas, San Luis Potosí, and reported that the occurrence was explained by agaves (pineapple width and percentage of agaves infested with scale insects), woody plants, land cover, and slope of the land. In this study, the CCA indicated a weak association of nest density with agaves, so this variable was not a good predictor. This contrasts with that reported by several authors, as this succulent plant is one of the main nesting and foraging substrates of L. apiculatum (Rafael-Valdez et al., 2017, 2019). Agaves were counted in circular plots (absolute frequencies) and L. apiculatum nesting substrates were not recorded as a variable, so it is likely that the results of the statistic could detect a substantial correlation with such plants by including this information.
Differences in nest densities can also be attributed to site effects and habitat condition, as well as land use history (Lara-Juárez et al., 2016). Agriculture and livestock have caused changes in the edaphic soil component and as a consequence have increased bare soil, which has greater susceptibility to erosion and solar radiation (Ríos-Casanova et al., 2012), which can negatively affect ant foraging activity and ant nest density.
Habitat variables associated with the presence of nests
The first two axes of the CCA reached values of 0.188 and 0.094, explaining 93.6 % of the cumulative variance of nest presence and its association with vegetation type (Figure 4). The inertias between axes indicate that the dimensional analysis was a good predictor between the relationships of habitat variables, number of plants and sampling sites. The CCA shows that the variables associated with the presence of nests according to vegetation type (CS, DMS and RS) are slope direction, rock, bare soil and the soil unit eutric leptosol + lithic leptosol medium texture.
Figure 4 Canonical Correspondence Analysis (CCA) between habitat variables and the presence of Liometopum apiculatum nests. Vegetation types: C1:C16 = crassicaule scrub; M17:M23 = desert microphyll scrub; R24, R25, R26 and R27 = rosetophyll scrub. Plants: AGA = Agave ssp., OPUS = Opuntia ssp, YUC = Yucca ssp. Habitat: %P = slope steepness, ALT = altitude, S. D. = slope direction, HER = herbs, G = Grass, B. S. = bare soil, ROC = rock, W. M. = woody material, S = shrubs, CAC = cacti, ROS = rosette plants, and seven soil units (Les + Ll2 = eutric [skeletic] Leptosol + medium-textured lithic Leptosol; Les + Rse2 = eutric [skeletic] Leptosol + medium-textured [epileptic] skeletic Regosol; Les + Rse + Lm = eutric [skeletic] Leptosol + skeletic [epileptic] Regosol + medium-textured mollic Leptosol; Lvc2 = medium-textured chromic Luvisol; Lcm + Lvce2 = calcaric [mollic] Leptosol + medium-textured chromic [epileptic] Luvisol; Rce + Lcr2 = calcaric [epileptic] Regosol + medium-textured calcareous [rendzic] Leptosol; Le + Ll2 = eutric Leptosol + medium-textured lithic Leptosol).
Slope direction
At the sampling sites, the northwest slope direction was the most frequent. This agrees with that observed by Rafael-Valdez et al. (2019), who place this direction as a usual direction of L. apiculatum foraging paths in a DMS. It is possible that this direction is associated with thermal regulation processes inside the nests. Microclimatic factors caused by ant activity, should provide temperature and humidity thresholds that facilitate larval care and ensure their growth (Römer et al., 2018). In this regard, it has been documented that the temperature and relative humidity (annual) inside the nests (trabeculae) of L. apiculatum in xerophytic scrub are higher (20.8 °C and 88.4 %, respectively) than those recorded in the environment (18 °C and 53 %; Cruz-Labana et al., 2014; Lara-Juárez et al., 2015), so it is very likely that the ant conforms a favorable microclimate for the development of its larvae.
In other formicids, it has been pointed out that the direction and inclination of the land offer greater opportunities for sunlight use. Ants as thermophilic organisms use this energy in foraging (Nobua-Behrmann et al., 2017; Sondej et al., 2018). Cedeño et al. (1999) studied the spatial distribution of nine ant species in a rainforest in Venezuela and found that they have abiotic relationships (nesting, foraging, seed dispersal, and nutrient flow affectations) with four plant families (Araceae, Bromeliaceae, Cactaceae, and Gesneriaceae) and identified ant hillsides with northwest direction, mean annual temperature of 26 °C, and solar radiation of 227.1 W∙m-2.
The results of Cedeño et al. (1999) have some similarity with that pointed out by Hernández-Roldán (2017), who reported that, in a crassicaule scrub, L. apiculatum travels its foraging paths with an average solar radiation of 282 W∙m-2 and temperature of 17 °C. This author points out the possibility that the northwesterly direction offers suitable conditions for the ant to travel to its feeding areas without facing risk of dehydration.
Rocky cover
Rocky ground provides thermal cover for L. apiculatum during the hours of highest solar radiation (Lara-Juárez et al., 2015). Hoey-Chamberlain et al. (2013) describe that the foraging activity of L. apiculatum is inhibited at temperatures ≤8 °C to ≥38 °C. Possibly, the ants find temporary refuge from the higher temperatures in the rocks during foraging activity.
For the escamolera ant, a butyric acid odor has been reported in their foraging paths. This compound is possibly part of a pheromone that induces recruitment and tracking of chemical trails (Hoey-Chamberlain et al., 2013); however, these may volatilize so ants may use another guidance system. For example, Temnothorax albipennis Curtis uses rocks on the ground as panoramic landmarks during their transfers from the nest to feeding sites and vice versa (Hunt et al., 2018).
Bare soil
Bare soil is a frequent variable interacting with the distribution of L. apiculatum according to vegetation type. This result is similar to that described by Cruz-Labana et al. (2014) and Hernández-Roldán et al. (2017), who mention that the possibility of finding ant nests decreases with increasing areas of bare soil. A terrain with these characteristics can cause a negative effect on foraging patterns because the lack of protection from high temperatures decreases the performance of ant workers to gather food, to the detriment of the production and development of larvae (Ramos-Elorduy et al., 1986). The ant L. apiculatum has foraging paths in several directions (Rafael-Valdez et al., 2019), so it is likely to set a course on bare ground, as it represents the optimal route to food sources.
Type of soil
The eutric leptosol soil was identified as a variable associated with the presence of nests in the CCA. Many ant species build their nests in soil, because it provides necessary habitat for larval development, as well as offering protective cover for their castes (Halboth & Roces, 2017).
Information on L. apiculatum and its relationship with the edaphic component is scarce. However, it is known that the beginning of the excavation and the foundation of the nest by the queen take place in moist soil (Ramos-Elorduy et al., 1984). Nest architecture depends on soil type, moisture gradients, textural properties, food availability, and the increasing size of the anthill (Guimarães et al., 2018; Kwapich et al., 2018). For example, Atta sexdens rubropilosa Forel and Atta vollenweideri Forel prefer moist soils because they are easier to excavate (Pielström & Roces, 2014; Swanson et al., 2019). The queen of L. apiculatum begins nest construction with a conical chamber 4.0 to 5.5 cm in major diameter, 3.5 to 4.5 cm in minor diameter and 1.5 cm in height (Ramos-Elorduy et al., 1984), therefore, moist soil conditions are likely to make it easier for her to excavate and transport material (jaw-sized aggregates) to the outside of the nest.
After nest foundation, the first and second generation of L. apiculatum (worker caste) initiate the construction of the brood chamber (trabeculae) and nest expansion. Cardoso et al. (2021) mention that soil texture is one of the factors that impact ant nest architecture and excavation habits. In this regard, it has been documented that L. apiculatum selects soils (in addition to certain plant materials) with textural properties of 50 to 70 % sand, 10 to 35 % silt, 15 to 20 % clay and 26 to 31 % organic matter content for trabeculae construction (Ramos-Elorduy et al., 1984).
The soils with greatest presence in the study area were eutric leptosol, cambisol leptosol, calcaric regosol and chromic luvisol (INEGI, 2014b). This result is similar to that reported by Lara-Juárez et al. (2015), who identified leptolsol, feozem and regosol as the soil units occupying the largest area in CS, DMS, RS, chaparral, oak forest, pine forest and pine-oak forest with nests of L. apiculatum distributed in the Altiplano potosino. Eutric leptosols (high saturation in most of the profile) and cambisol (poorly developed, intermediate between other soil units) are shallow soils (<25 cm) and of high stoniness (Palma-López et al., 2017). The leptosol, regosol and luvisol are medium textured soils characterized by being moderately compact with significant presence of aggregates. It is possible that these properties facilitate the construction of chambers, galleries and tunnels, decrease the effort of soil and debris removal, and provide stability throughout the architecture of the nest of L. apiculatum (Lara-Juárez et al., 2015).
Conclusions
There are 4.8 nests∙ha-1 in the crassicaule scrub, while 2.2 and 2.3 nests∙ha-1 were found in the desert microphyll scrub and rosetophyll scrub, respectively, showing that nest density is heterogeneous according to vegetation type. Slope direction, rocky cover, bare soil, and soil type (eutric leptosol + lithic leptosol [medium texture]) are variables associated with the presence of nests; canonical correspondence analysis showed weak correlation with agaves. This research identified some of the conditions that sites should meet for the establishment and development of the nest life cycle and contribute to their sustainable use. With the current knowledge of the habitat of L. apiculatum, research can be directed to establish artificial nests in the distribution areas of the species, in Wildlife Conservation Management Units, and even in properties and facilities that manage wildlife.
