Neotrop Entomol (2012) 41:95–104 DOI 10.1007/s13744-011-0009-5
ECOLOGY, BEHAVIOR AND BIONOMICS
Effect of Biotic Factors on the Spatial Distribution of Stingless Bees (Hymenoptera: Apidae, Meliponini) in Fragmented Neotropical Habitats MM FIERRO1, L CRUZ-LÓPEZ1, D SÁNCHEZ1, R VILLANUEVA-GUTIÉRREZ2, R VANDAME3 1
El Colegio de la Frontera Sur, Unidad Tapachula, Tapachula, Chiapas, Mexico El Colegio de la Frontera Sur, Unidad Chetumal, Chetumal, Quintana Rôo, Mexico 3 El Colegio de la Frontera Sur, Unidad San Cristóbal, San Cristóbal de Las Casas, Chiapas, Mexico 2
Keywords Cluster distribution, disperse distribution, foraging overlap, nesting site Correspondence Leopoldo Cruz-López, El Colegio de la Frontera Sur, Unidad Tapachula, Carretera Antiguo Aeropuerto km 2.5, Tapachula, CP 30700 Chiapas, Mexico;
[email protected] Edited Fernando B Noll — UNESP Received 21 June 2011 and accepted 9 November 2011 Published online 15 February 2012 * Sociedade Entomológica do Brasil 2012
Abstract We recorded stingless bee colony abundance and nesting habits in three sites with different anthropogenic activities in the Soconusco region of Chiapas, Mexico: (1) agroforestry (7 hacacao crop), (2) grassland (12 ha), and (3) urban area (3 ha). A total of 67 nests were found, representing five stingless bee species, Tetragonisca angustula angustula (Lepeletier), Trigona fulviventris (Guérin), Scaptotrigona mexicana (Guérin), Scaptotrigona pectoralis (Dalla Torre), and Oxytrigona mediorufa (Cockerell). The most abundant stingless bee in each site was T. angustula angustula (>50%). The primary tree species used by the bees were Ficus spp. (Moraceae, 37.8%) and Cordia alliodora (Boraginaceae, 13.5%). The nest entrance height of T. angustula angustula (96±19 cm) was different than the other species, and this bee was the only one that used all different nesting sites. Volatiles analyzed by gas chromatography from pollen collected by the stingless bees differed between bee species, but were highly similar in respect to the fragrances of the pollen collected by the same species at any site. Our data indicate that T. angustula angustula experienced low heterospecific and high intraspecific foraging overlap especially in the urban site. We observed cluster spatial distribution in grassland and in agroforestry sites. In the urban site, T. angustula angustula presented random distribution tended to disperse. Trigona fulviventris was the only overdispersed and solitary species.
Introduction Meliponines also known as stingless bees are important native pollinators and widely distributed in the Neotropics. Nevertheless, literature is scarce concerning to the role biotic factors plays on spatial distribution of stingless bees. The few studies on stingless bee spatial distribution are commonly related with tree nesting site preference and nest site availability (Hubbell & Johnson 1977, Batista et al 2003). However, another factor that could influence spatial distribution, which has not been considered in previous studies, is the overlap for pollen (the main source of protein) foraging between stingless bees. Stingless bees are considered polylectic, but some species of stingless bees may frequent the same plant species
(Ramalho et al 1990, Johnson 1983, Nagamitsu & Inoue 1997). For example, the fact that species of Trigona are generalist flower visitors and exhibit a substantial overlap in the taxa of the floral resources they exploit, suggests that Trigona spp. may display competition interspecifically for floral resources (Heithaus 1974, Steffan-Dewenter & Tscharntke 2000). Intraand heterospecific comparisons should be helpful in recognizing some of the biotic factors that may influence abundance and spatial distribution in stingless bee communities. In addition, accelerated environmental transformation all around the world due to change of land use, e.g., agroforestry, farming, and urban development, is causing changes in microclimates, and in insect and forest plant community structures as well (Perfecto & Vandermeer 1996, Kremen et al 2002).
96
These changes have negatively affected the provision of pollinator services by reducing the number of pollinators (Ghazoul 2005, Steffan-Dewenter et al 2005, VillanuevaGutiérrez et al 2005). The region of the Soconusco, in the state of Chiapas, Mexico, is part of the Mesoamerican biological corridor (Ramírez 2003), which includes extensive areas of Neotropical fragmented landscapes, consisting of mosaics of forest patches, pastures, and agroforestry (Salgado et al 2007). However, in the Soconusco, there is no information in respect to stingless bee nest site preferences or their abundance. In this study we investigated both nesting preferences and abundance within a framework that included land use and resource overlap (pollen) within the habitat as important variables that may indicate possible influence on the spatial distribution of stingless bee communities.
Material and Methods Study area The study was carried out from May 2009 to April 2010 in a humid Neotropical climate in the Soconusco region, Chiapas, Mexico. We selected three sites with different levels of degradation caused by anthropogenic activity:
Fierro et al
(1) agroforestry featuring cocoa (site C—7 ha; 14°53′07″ N, 92°11′52″ W), (2) grassland (site G—12 ha; 14°52′55″ N, 92°12′18″ W), and (3) urban area (site U—3 ha; 14°53′11″ N, 92°17′13″ W; Fig 1). Site descriptions Sites were similar in altitude (132–188 m asl), temperature (25–38°C), and annual rainfall (2,500–3,000 mm), but they had been impacted by different levels of anthropogenic activity over the last 40 years. Site C is limited to the north and south by pastures and seasonal crops, to the east are houses belonging to plantation workers and to the west there is a cocoa plantation and a fragment of primary forest (Fig 1). There were six feral colonies of honey bees Apis mellifera (L.) (Hymenoptera: Apidae) within this site. Anthropogenic activity was moderate and was related to the harvest and trimmer maintenance. Site G is situated 1 km southwest from habitat C, and is limited to the north by cocoa plantations, to the east by seasonal crops, to the south by grassland, and to the west by seasonal crops and mangos (Fig 1). Site G has few adult trees towards the interior, but there are hundreds of trees within the live fences. Within this site there were five feral colonies of honey bees settled on tree branches. Human activity was low. Site U is a complex of school buildings of
Fig 1 Map of the Soconusco region tropical humid forest showing the study sites. The three sites were extensively explored looking for stingless bee nesting sites. Upper right agroforestry site (C); bottom left grassland site (G); bottom right urban site (U). A cluster of stingless bee nests sampling for pollen appear inside the rectangle in each site with at least one of each species, with the exception of Trigona fulviventris nest in site G that appear solitary.
Spatial Distribution of Stingless Bees in Fragmented Neotropical Habitats
3 ha surrounded by a few large, old trees, mainly Ficus spp. (Moraceae, Fig 1). The anthropogenic activity all around was high. Nest location We carried out the survey by walking through the sites to find colonies. We recorded the genus and species of the stingless bees, coordinates of each colony, and nesting substrate (trees, trunks, subterranean, or the denominated “artificial” such as walls or rocks). As regards tree substrates, we recorded the genera and species, diameter at breast height (dbh), and nest entrance height. We counted and classified all trees with dbh ≥15 cm (Hubbell & Johnson 1977). Meliponines and tree species identification We collected several samples of bees from each nest for identification in the laboratory, using established keys (Ayala 1999). The few uncommon species of trees were identified using the keys developed by Pennington and Sarukhán (2005). Samples of bees and plants species were deposited at the “El Colegio de la Frontera Sur” insect collection and herbarium, respectively. Pollen collecting Stingless bees were sampled between 08:00 and 11:00 hours (the main period of the day in which foragers are more active), using manual vacuuming of the returning pollen foragers at the bee’s nest entrance. Samples of pollen were taken monthly during the flowering season from December 2009 to March 2010. The three sites were sampled in one single day per habitat in similar weather conditions registering date, time, and bee species as well as the nest site. A cluster of stingless bee nests with at least one of each species was selected for sampling in order to evaluate foraging overlapping in real time under similar circumstances. Oxytrigona mediorufa (Cockerell) was not sampled due to the height of the nest entrance and the bee’s highly defensive behavior. Pollen sample handling Foragers carrying pollen were kept frozen at −20°C for analysis. Pollen pellets were easily removed from the bee’s corbiculae in all species and separated by color and appearance. In most cases we used a full load of pollen (total pollen per bee) in each analysis, however, with Trigona fulviventris (Guérin), due to the large amount of pollen carried per bee, half of the total pollen load was used.
97
Chemical analysis In order to determine pollen volatiles profile, gas chromatographic separation of pollen sample volatiles was performed on a Varian Star model 3400 CX GC (Palo Alto, CA, USA), using the Varian ChromatoProbe sample introduction device (Palo Alto, CA., USA), using a DB-5 column (30×0.25 mm ID) with oven temperature programmed from 50°C (held for 2 min) to 250°C at 15°C min−1. The injection port temperature was held at 200°C. Data analysis The Kruskal–Wallis test was used to calculate species nest entrance height and species abundance differences. The Pielou’s evenness index (J′) was calculated through the formula J′ 0H′/H′max. Nest spacing was estimated using nearest neighbor (NN) distance from the nesting site GPS data. Spatial distribution patterns for T. fulviventris were analyzed by calculating an index of aggregation (R), using the method of Clark and Evans. The index of aggregation (R) was corrected for lack of boundary strips around the sites applying the Donnelly correction. R>1 indicates a disperse distribution; R<1 indicates a cluster distribution; R01 indicates a random distribution. The Riplay’s K function method determines the spatial pattern at ten different radius (t) intervals of 10 m each and requires at least 15 separate nests for the analysis in each site. Both methods are contained in an ArcGIS Desktop Arc View version 9.2 (Environmental Systems Research Institute 1999–2006). For statistical analysis a significance level of 5% was used.
Results Stingless bee species and nest density A total of 67 stingless bee nests representing five species were located. Among these, Tetragonisca angustula angustula (Lepeletier, 38 colonies) was the most abundant species followed by Scaptotrigona mexicana (Guérin, 12 colonies), Oxytrigona mediorufa (9 colonies), T. fulviventris (4 colonies), and Scaptotrigona pectoralis (Dalla Torre, 4 colonies; Table 1). Nest density differed among the stingless bee species (χ2 09.9, df04, P00.04). The highest nest density was observed with T. angustula angustula in site U with 5.3 nests per ha (Fig 2). Availability of nest sites and substrate preferences Species of bees differed in nesting habit. The frequency of nests per substrate type varied among the five
98
Fierro et al
stingless bee species (Table 1). In general, these bees preferred holes in trees (65.7%), followed by holes in other substrates like rocks and cement blocks (mentioned here as “artificial,” 16.4%), a few nests were found in dead trees (trunks, 10.5%). However, T. fulviventris always uses subterranean cavities to build their nests (Table 1). Preferences for tree species for the four species of stingless bees were similar. However the nest entrance height patterns were different (χ2 0 12.7514, df03, P<0.01). One special nesting characteristic was observed in T. angustula angustula which always established their nest closer to the ground, while S. mexicana, S. pectoralis, and O. mediorufa nested higher (Table 1). Furthermore, T. angustula angustula was capable of using the majority of substrates for nesting, observing the highest evenness index (J′, Table 1). Of the 16 tree species used to establish their nests; T. angustula angustula used 11 (Table 2). Ficus spp. was used primarily by T. angustula angustula (50%), followed by O. mediorufa (37%) and S. mexicana (30%). Cordia alliodora (Boraginaceae) was the second most used tree species by S. mexicana and T. angustula angustula recording 40% and 11%, respectively. The nest sites established in Ficus spp. were most numerous (χ2 025.341, df015, P<0.05), in relation to the other tree species visited and the density of this species in the three sites (Table 2). Tree diversity We surveyed a total of 1,193 trees with dbh >15 cm including fruit trees, timber trees, and others distributed in 24 families and 54 taxa. Although the number of trees surveyed in sites C and G were approximately the same (518 and 526, respectively), with similar average tree height (20 m), with the exclusion of Ficus spp. (Moraceae) (dbh0233±39.5 cm), both sites had similar dbh (63 ± 13.5 and 67 ± 11.5 cm, respectively). In the urban site, most of the trees were young trees between 10 and 15 years old (dbh034±7.5 cm). The only old trees (40 to 60 years old), were a small group of
Fig 2 Nest density of stingless bees in three nesting sites in the Soconusco region Chiapas, Mexico (N067).
palms and a dozen of Ficus spp. trees. Not considering Ficus spp. trees (preference taxon used for nesting at site U) in the three sites, the tree dbh used for nesting was 65.6±27.2 cm (Table 2). Floral resource preference according to pollen foraging and gas chromatography analysis A total of 73 pollen samples from 18 nests sites distributed in the three habitats were analyzed during the sampling period (December to March, Table 3). According to the chromatogram profiles, specific foraging preferences were observed for each of the stingless bees’ species (Fig 3). Each stingless bee species showed several floral foraging preferences (Table 3). These preferences were different between species, but highly similar within each species. This stingless bee behavior was observed in each site. Tetragonisca angustula angustula, the most abundant species, with the largest number of samples for analysis, displayed six different chromatographic profiles from 39 samples (71% of the bee pollen loads were from two floral
Table 1 Stingless bee nesting site preferences at the three sites in the Soconusco region, Chiapas, Mexico. Stingless bee species T. angustula angustula (n038) S. mexicana (n012) O. mediorufa (n09) S. pectoralis (n04) T. fulviventris (n04)
Tree diversity
neh (±SD)
Artificial
Subt.
Trunks
Total (%)
Evenness index (J′)
31.3 16.4 12.1 5.9 NF
96±19 224±88 288±148 253±121 ND
16.4 0 0 0 0
1.5 0 0 0 5.9
7.5 1.5 1.5 0 0
57.5 17.9 13.6 5.9 5.9
0.75 0.21 0.25 ND ND
Tree nest entrance height (neh in centimeters). Subt. subterraneous, ND not determinated, NF not found.
99
Spatial Distribution of Stingless Bees in Fragmented Neotropical Habitats
patterns, we included two nests sites located outside the study area, between sites C and G, resulting in a space between nests of 293.2±19.1 m.
taxa; Table 3, Fig 3a). Trigona fulviventris and S. mexicana presented a polylectic behavior, resulting in seven taxa visited (Table 3) with different chromatographic profiles from 13 to 12 bee pollen load, respectively (Table 3, Fig 3c, d). Finally, the least abundant stingless bee species, S. pectoralis, with just four nests, collected only in sites C and G (two nests each), had three chromatographic profiles (three taxa visited), from nine bee pollen load (Table 3), showing a majority of them representing one particular profile (Fig 3b).
Discussion Stingless bees species, tree diversity, and nest density The stingless bee species found in the three study areas used 16 tree taxa to establish their nests. The most used trees for bee nests were Ficus spp. and C. alliodora, with 28% and 22% of the total tree nests, respectively. Martins et al (2004) registered the use of 12 tree species in a Meliponini community in caatinga, Brazil. They reported that 75% of the stingless bees established nests in two taxa of trees: Caesalpinia pyramidalis (Caesalpiniaceae, 41.9%) and Commiphora leptophloeos (Burseraceae, 33.9%). A similar situation was observed by Rêgo and Brito (1996) in an ecosystem in Cerrado, Brazil and by Copa-Alvaro (2004) in La Paz, Bolivia, where 47% of the T. angustula angustula nests were established in just one single tree species Astronium urundeuva (Anacardiaceae). It is not always possible to determine the preference for a particular tree species (Nates-Parra et al 2008), but these results suggest that the stingless bee community explores and expresses preference for tree taxa with nesting sites available (hollows) and that tree diversity may influence stingless bee
Spatial distributions of nests The patterns of spatial distribution as determined by the K function registered at different 10-m intervals offer detailed information (Fig 4a–d). Only the number of total nests in each site and T. angustula angustula in site G and U were present in sufficiently high numbers to determine spatial patterns (Fig 4b, d).The nests found within the grassland site (G), and the agroforestry (C) showed a clustered distribution (10–70 m), tending to be located in “clumps” (Figs 1 and 4a, c). However, the spatial distribution for T. angustula angustula at site G was totally random (10–100 m, Fig 4b).Whereas the nests located within the urban (U) site were randomly distributed (10–40 m), they tended to disperse (50–100 m, Fig 4d). The subterraneous T. fulviventris nests were observed overdispersed (R01.75) and solitary. To further analyze T. fulviventris nesting
Table 2 Tree diversity used by stingless bees for nesting in three sites at El Soconusco region, Chiapas, Mexico. Family
Anacardiaceae Bignoniaceae Boraginaceae Burseraceae Fabaceae Moraceae
Palmae Rutaceae Sapindaceae Sapotaceae Meliaceae Leguminosae
Species
Mangifera indica Tabebuia donnell-smithii Tabebuia pentaphylla Cordia alliodora Bursera simaruba Diphysarobinioides Tamarindus indicus Ficus sp. Ficus involuta Brosimunco staricanum Cocos nucifera Citrus sp. Cupania dentata Chrysophyllum caimito Trichilia martiana Platymiscium dimorphandrum
dbh (±SD)
T. angustula angustula
S. mexicana
S. pectoralis
O. mediorufa
C
C
G
U
C
G
U
C
G
U
G
Total
U
78±0.0 64±4.2
1 0
0 1
0 0
0 0
0 0
0 0
0 0
0 1
0 0
0 0
0 0
0 0
1 2
83±4.0 57±8.8 68±0.0 74±0.0 48±0.0 87±0.0 247±63 92±7.5 39±3.5 37±5.7 67±0.0 118±0.0 84±0.0 54±0.0
0 1 1 0 0 0 0 0 2 1 0 0 0 0 6
2 3 0 1 0 0 2 0 0 1 1 0 0 1 12
0 0 0 0 0 0 7 0 0 0 0 0 0 0 7
0 1 0 0 0 0 1 1 0 0 0 1 0 0 4
1 5 0 0 0 1 0 0 0 0 0 0 0 0 7
0 0 0 0 0 0 2 0 0 0 0 0 0 0 2
0 0 0 0 1 0 0 0 0 1 0 0 0 0 2
0 0 0 0 0 1 0 0 0 0 0 0 0 0 2
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 2 1 0 0 0 1 0 0 4
0 1 0 0 0 0 0 0 0 0 0 0 3 0 4
0 0 0 0 0 0 1 0 0 0 0 0 0 0 1
3 11 1 1 1 2 15 2 2 3 1 2 3 1 51
C agroforest, G grassland, U urban area, dbh diameter breast height, SD standard deviation (in centimeters).
100
Fierro et al
of disturbance. Our results show the same floral preferences by intraspecific stingless bee species in each site as well as no heterospecific foraging overlapping, at least during the main blooming season (December to March). GC analysis showed similar chromatogram profiles from pollen samples collected by the same species of bees from the same or different sites, while GC profiles between species from different or same sites were different. This is probably due to an innate evolutionary coupling, an oscillating system that tends to coordinate species survival when coupled (Vandermeer 2006).This point enhances the idea that pollinators, in this case stingless bees, may be flexible and may change foraging preferences so as to avoid interference or exploitative competition, especially between foragers of aggressive species, providing that there is a rich and abundant supply of resources (Butz 1997). Similar results were obtained by Martínez et al (1994), who carried out a melissopalinological analysis of honey and pollen from four stingless bee species (including T. angustula angustula and S. mexicana) in the Tacaná region of Chiapas, Mexico and found just one foraging niche overlap between S. mexicana and T. angustula angustula in which Sapindus saponaria (Sapindaceae) was shared as a source of nectar and pollen. Although the four species we analyzed showed a polylectic foraging behavior for a specific group of floral taxa, T. angustula angustula was the greatest generalist foraging for pollen throughout the sampling period. According to our pollen chromatographic profiles, T. angustula angustula had more than 70% preference for just two floral resources. Such differences can be ecologically important since polylectic bees can become more specialized when conditions favor monopolization of certain resources (Heithaus 1979). The trophic foraging niche overlap seems not to occur to an important degree during the blooming season, but since some specimens of stingless bee species (S. mexicana and S. pectoralis) did not return with pollen on the sampling day, it was not possible to evaluate their floral preferences during that month. In order to accurately evaluate the real level of foraging overlap at specific times, we suggest sampling additional in situ pollen
spatial distribution. In site C, Ficus and C. alliodora represented only 0.2% and 4.4%, respectively, of the diversity, but they were inhabited by 31% of the total stingless bee nests. At site G, these two tree species represented 0.7% and 16%, respectively, of the tree diversity, but they were hosts to 44% of the stingless bees present (mostly C. alliodora, 36%). At site U, where C. alliodora was not available, 100% of the stingless bees used Ficus spp. even though it only represented 9.5% of tree diversity. It is important to mention that Ficus spp. tree trunks are large (dbh0233±39.5 cm), and the dbh of the trees used for nesting at sites C and G was 65.6±27.2 cm. This size relates to older trees that usually have more hollows to be used as nesting sites (Eltz et al 2003). In contrast, the dbh at site U (excluding Ficus spp. tree) was 34±7.5 cm, typically of young trees. According to Hubbell and Johnson (1977), the major factor involved is security, and safe nesting sites usually occur in older trees, especially those that have been preconditioned by birds, rodents, and/or natural decay and a process that might be occurring with Ficus and C. alliodora trees. Floral resource preference One unexpected characteristic found was the speciesspecific preferential foraging behavior. According to Imperatriz-Fonseca et al (1984), stingless bees are very efficient generalist pollinators. So, this trait, especially in Trigona sp. (Heithaus), could produce high foraging resource competition, especially in areas where high-quality resources of protein (pollen) have been reduced because of a change in land use. Competition influences the spatial nesting patterns in the stingless bee community (Slaa 2006a). There are a few studies that compare foraging ecology in Meliponini communities (e.g., Martínez et al 1994, Sosa-Najera et al 1994), but none that include intraand interspecific resource overlapping in situ in stingless bee communities present in habitats with different levels
Table 3 Stingless bee floral resources preferences according to pollen gas chromatography (GC) analysis in the three sites in the Soconusco region, Chiapas, Mexico. D
Species/site T. angustula angustula S. mexicana S. pectoralis T. fulviventris
C 2
a
G
2 2 2a 2a 1a 2 1a
J
U a
2 1a NF 1
F
Total bee Total nests sampling per pollen load analyzed month
M
C
G
U
C
G U
C
G
U
2 2 2 2
2 2a 1 1
2 1 NF 1
2 1 2 1
2 2 1 1
2 1 2 1
2 2 1 1
2 1a NF 1
2 1 NF 1
6 5 3 4
39 12 9 13
Total taxa % taxon % next visited (GC) (mv) taxon (mv)
6 7 3 7
D December, J January, F February, M March, C agroforestry, G grassland, U urban, mv most visited, NF nest not found. a
Bees returning with no pollen load on the sampling day.
40.5 25 44.4 23
35.1 16.6 33.3 15.4
Spatial Distribution of Stingless Bees in Fragmented Neotropical Habitats
101
Fig 3 Gas chromatographic profiles of pollen volatiles from floral preferences most visited from each species: Tetragonisca angustula angustula (a), Scaptotrigona pectoralis (b), Trigona fulviventris (c), and Scaptotrigona mexicana (d).
Fig 4 Spatial Spatial nesting pattern in in the the three three sites. sites. aa Spatial Spatial pattern pattern of of nests nests found found at at site site G G (grassland), (grassland), b b Tetragonisca T. angustula angustula pattern at Fig 4 nesting pattern angustula spatial angustula spatial site G (grassland), c spatial pattern of nests found at site C (agroforestry), d spatial pattern of nests found at site U (urban area). Observed values pattern at site G (grassland), c spatial pattern of nests found at site C (agroforestry), d spatial pattern of nests found at site U (urban area). between andbetween high confidence are considered a random distribution pattern. Values observed the high confidence Observedlow values low andcurves high confidence curvesasare considered as a random distribution pattern.outer Valuesofobserved outer of thecurve high are cluster distribution pattern.distribution Values observed below the observed low confidence consideredinterval a disperse Thedistribution function K confidence curve are cluster pattern. Values belowinterval the lowareconfidence are distribution considered pattern. a disperse as independent variable, calculatedvariable, considering circle of radius (t) for aeach interval, K(t).each The10 derived sample statistic √[K(t)/π] as a pattern. The function K aswas independent wasacalculated considering circle10ofmradius (t) for m interval, K(t). The derived sample dependent variable plotted against t. Black diamond expected, square observed, triangle low confidence, high statistic √[K(t)/π] as ais dependent variable is plotted against t. Black white diamond expected, whiteblack square observed, black trianglegray lowsquare confidence, confidence. gray square high confidence.
102
loads and analyzing their volatiles with more periodicity during one whole year. Spatial distribution The stingless bee’s spatial nesting distribution has not been widely studied. Their level of sociability (Chavarría & Carpenter 1994, Noll 2002) observed in their strategies for mating success (Van Veen & Sommeijer 2000) as well as their sophisticated communication systems for defense and foraging among stingless bees should be playing an important role in their spatial distribution. Usually stingless bee species mark food sources and lay trails with pheromones, sound production, or flight piloting (Sánchez et al 2004), some of them are foraging solitary and aggressive (Hubbell & Johnson 1977). Scaptotrigona mexicana is a species which show highly sophisticate strategy for foraging (Sánchez et al 2004), while T. angustula angustula seems to be less sophisticate in their foraging system (Villa & Weiss 1990). Nevertheless, the most common mechanism of nest dispersion seems to be direct aggressive interaction for food and nesting site, provoking territoriality (Johnson & Hubbell 1974). This behavior has been found at least in three stingless bee species, and T. fulviventris is one of them. However, the spatial nesting pattern of a nonaggressive stingless bee like S. pectoralis, T. angustula angustula, S. mexicana, and O. mediorufa seems different and opposite to the aggressive bees. Nonaggressive bees usually establish their nests in clusters and with a random spatial distribution except when the competition for food is high (Hubbell & Johnson 1977, Slaa 2006a). In our study area, for the result for site G (Fig 4a), with fewer old trees and reduced nest density (2.1 nest per ha), the distribution pattern [as defined by Slaa (2006a) for nonaggressive foragers such as T. angustula angustula, S. mexicana, S. pectoralis, and O. mediorufa] revealed a highly clustered nest distribution (10–70 m). Only Tetragonisca angustula angustula (Fig 4b) had a random spatial distribution (10–100 m). Site C (Fig 4c) had the largest density of stingless bee nests (3.3 nests per ha) and cluster spatial distribution (10–70 m), typical in stingless bee communities without competition. Tetragonisca angustula usually established cluster or random nesting patterns as observed in site C and G, but in site U (Fig 4d), tended to be dispersed (50–100 m). The aggressive foraging bee T. fulviventris, which nests underground, often among tree roots, was over dispersed (symmetric distribution). To evaluate this behavior, we included two T. fulviventris nesting sites located between site C and G. The average near neighbor distance (NN) was (293.2±19.1 m), suggesting that T. fulviventris maintains a territorial area around the nest with a radius of approximately 270 m. Trying to understand the factor influencing the interspecific spatial distribution as a whole is very complicated. We
Fierro et al
observed a significant cluster distribution of nests in site G, characterized by lower nest density. However, the significant landscape changes (perturbation), as found in site U, reduced tree diversity and nesting site availability may possible affect T. angustula angustula spatial pattern distribution, which tended to be dispersed. In social insects, disperse nesting patterns are generally believed to possibly indicate competition mainly for food or nesting sites (Hubbell & Johnson 1977). The overlapping foraging ranges among neighboring colonies may generate competition (SteffanDewenter & Tscharntke 2000), which should affect the spatial nesting patterns in aggressive species, but not in T. angustula angustula (Slaa 2006a). Only a few nests per species were sampled in each site, so it was not possible to determine the intraspecific level of foraging overlapping for food resources. However, it should be pointed out that in site U, the bees established the usual random and eventually dispersed spatial nesting pattern, even when living near an apiary with 50 permanent honey bee colonies foraging similar floral resources (unpublished data). Nesting of T. angustula angustula was most numerous at site U, indicating that this species was not significantly constrained by competition for food resources. In our study area, T. angustula angustula was the most successful in nesting. Due to their ecological traits, T. angustula angustula is one of the most interesting stingless bee species. Their great adaptability enables them to establish nests in a wide variety of places with the highest evenness index, as has been widely shown (Batista et al 2003). Compared to the majority of the Meliponini, T. angustula angustula is highly resistant to ecosystem perturbation (Batista et al 2003, Slaa 2006b). It has become one of the most widely distributed bee species in the neotropics (Nates-Parra et al 2006). Our results show that T. angustula angustula was the only stingless bee species capable of establishing nests at any substratum, especially at site U, where 53.3% of the total T. angustula angustula nests were established in artificial nesting sites. However, most of the Meliponini community needs trees for nesting and their flowers as a source of food. In summary, the factors that most affected the spatial distribution of stingless bee nests in fragmented areas were related mainly to the level of perturbation (e.g., deforestation), tree age (successional stage of the habitat), and tree species diversity (Eltz et al 2003). These factors may affect nesting site availability due to micro-climate changes and consequently lead to the reduction of many stingless bees species (e.g., S. mexicana, T. fulviventris, S. pectoralis, O. mediorufa), enabling others bees to dominate (e.g., T. angustula angustula). Future research considering these factors at different landscape altitudes should provide complementary knowledge of stingless bee spatial
Spatial Distribution of Stingless Bees in Fragmented Neotropical Habitats
distribution. Our documentation of these results is expected to contribute to the development of biodiversity conservation projects (Parrish et al 1999), especially those related to forest regeneration, landscape fragmentation management, including agroforest/farm, and the use of stingless bees for domestic crop pollination. Acknowledgments We deeply appreciate Dr. Gerald M. Loper for his comments and assistance with the translation of the manuscript and Professor Javier Valle for his support in the statistical analysis, and Adrian Peña and Arón Gamboa for their assistance in locating the study sites. This work was made possible through a doctoral scholarship (57515) sponsored by El Consejo Nacional de Ciencia y Tecnología (CONACYT). Finally, thanks are given to the Research and Postgraduate Division of the Universidad Autónoma de Chiapas (UNACH) for the permission and support.
References Ayala R (1999) Revisión de las abejas sin aguijón de México (Hymenoptera: Apidae: Meliponini). Fol Entomol Mex 106:1–123 Batista MA, Ramalho M, Soares EA (2003) Nesting sites and abundance of Meliponini (Hymenoptera: Apidae) in heterogeneous habitats of the Atlantic Rain Forest, Bahia, Brazil. Lundiana 4:19–23 Butz HVM (1997) Ecological impacts of introduced honey bees. Q Rev Biol 72:275–297 Chavarría G, Carpenter JM (1994) “Total evidence” and the evolution of highly social bees. Cladistics 10:229–258 Copa-Alvaro M (2004) Patrones de nidificación de Trigona (Tetragonisca) angustula y Melipona rufiventris (Hymenoptera: Meliponini) en el norte de La Paz, Bolivia. Ecol Appl 3:82–86 Environmental Systems Research Institute 1999–2006. ArcGIS: Release 9.2 (software). Redlands Eltz T, Zamrie I, Linsenmair KE (2003) Nesting and nest trees of stingless bees (Apidae: Meliponinae) in lowland dipterocarp forests in Sabah, Malaysia, with implications for forest management. For Ecol Manag 172:301–313 Ghazoul J (2005) Buzziness as usual? Questioning the global pollination crisis. Trends Ecol Evol 20:367–373 Heithaus ER (1974) The role of plant–pollinator interactions in determining community structure. Ann Mo Bot Gard 61:675– 691 Heithaus ER (1979) Flower feeding specialization in wild bee and wasp communities and seasonal Neotropical habitats. Oecologia 42:179– 194 Hubbell SP, Johnson LK (1977) Competition and nest spacing in a tropical stingless bee community. Ecology 58:949–963 Imperatriz-Fonseca VL, Kleinert-giovannini A, Cortopassi-Laurino M, Ramalho M (1984) Hábitos de coleta de Tetragonisca angustula angustula Latreille (Hymenoptera, Apidae, Meliponinae). Bol Zool Univ S Paulo 8:115–131 Johnson LK, Hubbell SP (1974) Aggression and competition among stingless bees: field studies. Ecology 55:120–127 Johnson LK (1983) Foraging strategies and the structure of stingless bee communities in Costa Rica. In: Jaisson P (ed) Soc. insects in the tropics, vol 2. Université Paris-Nord, Paris, pp 31–58 Kremen C, Williams NM, Thorp RW (2002) Crop pollination from native bees at risk from agricultural intensification. Proc Natl Acad Sci U S A 99(26):16812–16816
103
Martínez E, Cuadriello J, Ramirez E, Medina M, Sosa M, Melchor J (1994) Foraging of Nannotrigona testaceicornis, Trigona (Tetragonisca) angustula, Scaptotrigona mexicana and Plebeia sp. in the Tacana region, Chiapas, México. Grana 33:2005– 2217 Martins CF, Cortopassi-Laurino M, Koedam D, Imperatriz Fonseca VL (2004) Espécies arbóreas utilizadas para nidificação por abelhas sem ferrão na caatinga (Seridó, PB; João Câmara, RN). Biota Neotrop 4:1–8 Nagamitsu T, Inoue T (1997) Aggressive foraging of social bees as a mechanism of floral resource partitioning in an Asian tropical rainforest. Oecologia 110:432–439 Nates-Parra G, Rodríguez A, Vélez ED (2006) Abejas sin aguijón (Hymenoptera:Apidae: Meliponini) en Cementerios de la Cordillera Oriental de Colombia. Acta Biol Col 11:25–35 Nates-Parra G, Palacios E, Parra H (2008) Efecto del cambio del paisaje en la estructura de la comunidad de abejas sin aguijón (Hymenoptera: Apidae) en Meta, Colombia. Rev Biol Trop 56:1295–1308 Noll F (2002) Behavioral phylogeny of corbiculate Apidae (Hymenoptera; Apinae), with special reference to social behavior. Cladistics 18:137–153 Parrish J, Reitsma R, Greenberg R, McLarney W, Mack R, Lynch J (1999) Los cacaotales como herramienta para la conservación de la biodiversidad en corredores biológicos y zonas de amortiguamiento. Agroforest Am 6:16–19 Pennington TD, Sarukhán J (2005) Árboles tropicales de México: Manual para la identificación de las principales especies. UNAM-Fondo de Cultura Económica, México Perfecto I, Vandermeer J (1996) Microclimatic changes and the indirect loss of ant diversity in a tropical agroecosystem. Oecologia 108:577–582 Ramalho M, Kleinert-Giovannini A, Imperatriz-Fonseca VL (1990) Important bee plants for stingless bees (Meliponini and Trigonini) and Africanized honeybees (Apis mellifera) in Neotropical habitats: a review. Apidologie 21:469–488 Ramírez G (2003) El corredor biológico Mesoamericano. Biodiversitas 47:1–7 Rêgo MMC, Brito C (1996) Abelhas sociais (Apidae: Meliponini) em um ecossistema de cerrado s.l. (Chapadinha MA, BR): distribuição dos ninhos. In: encontro sobre abelhas, 3°, Ribeirão Preto, Anais., Ribeirão Preto. FFCLRP-USP 3:238–247 Salgado M, Ibarra G, Macias J, López O (2007) Diversidad arbórea en cacaotales del Soconusco, Chiapas, México. Interciencia 32:763– 768 Sánchez D, Nieh JC, Hénaut Y, Cruz L, Vandame R (2004) High precision during food recruitment of experienced (reactivated) foragers in the stingless bee Scaptotrigona mexicana (Apidae, Meliponini). Naturwissenschaften 91:346–349 Slaa EJ (2006a) Spatial nesting patterns in a Neotropical stingless bee community: do bees compete for food? Prod Neth Entomol Soc Meet 17:71–78 Slaa EJ (2006b) Population dynamics of a stingless bee community in the seasonal dry lowlands of Costa Rica. Insect Soc 53:70–79 Sosa-Najera MS, Martínez-Hernandez E, Lozano-García MDS, Cuadriello-Aguilar JI (1994) Nectaropolliniferous sources used by Trigona (Tetragonisca) angustula in Chiapas, southern México. Grana 33:225–230 Steffan-Dewenter I, Tscharntke T (2000) Resource overlap and possible competition between honey bees and wild bees in central Europe. Oecologia 122:288–296 Steffan-Dewenter I, Potts SG, Packer L (2005) Pollinator diversity and crop pollination services are at risk. Trends Ecol Evol 20:651–652
104 Van Veen J, Sommeijer MJ (2000) Observations on gynes and drones around nuptial flights in the stingless bees Tetragonisca angustula and Melipona beecheii (Hymenoptera, Apidae, meliponinae). Apidologie 31:47–54 Vandermeer J (2006) Oscillating populations and biodiversity maintenance. Bioscience 56:967–975
Fierro et al Villa JD, Weiss MR (1990) Observations on the use of visual and olfactory cues by Trigona spp. foragers. Apidologie 21:541– 545 Villanueva-Gutiérrez R, Roubik DW, Colli-Ucán W (2005) Extinction of Melipona beecheii and traditional beekeeping in the Yucatán peninsula. Bee World 86:35–41