Printer Friendly

Metacommunity structure of small mammals in western Mexico: is the San Pedro-Mezquital River a biological corridor?

Community composition is the combined result of local and regional processes acting in a local context (Ricklefs, 1987; Holt, 1993; Leibold et al., 2010). The realization that community attributes at the local scale are the result of a continuum of ecological processes at larger spatial scales has brought about a change in focus from local, community-oriented research to a regional large-scale perspective focused on populations that are distributed over ecological and geographic gradients (Rodriguez and Arita, 2004; Cottenie, 2005; Lewinsohn et al., 2006; Stevens et al., 2007; Presley et al., 2009, 2010; Keith et al., 2011; Willig et al., 2011). Community ecology has witnessed the rapid development of a new paradigm that unites local and regional processes to understand more clearly the spatial structure of groups of communities within regions (Leibold and Mikkelson, 2002; Leibold and Miller, 2004; Leibold et al., 2004; Holyoak et al., 2005; De la Sancha et al., 2014).

The metacommunity concept (Leibold et al., 2004; Presley et al., 2010) provides a framework with which to evaluate the organization of species composition along environmental gradients. Pattern-based approaches to metacommunity analysis (Leibold and Mikkelson, 2002; Presley et al., 2009, 2010) evaluate characteristics of species distributions along latent environmental gradients that emerge as a result of combinations of mechanisms that create particular metacommunity structures (random, checkerboard, nested, evenly spaced, Gleasonian or Clementsian structures; Leibold and Mikkelson, 2002). Each nonrandom structure assumes that species distributions are molded by a combination of biotic interactions (e.g., competition, habitat associations) and responses to abiotic factors (e.g., temperature, elevation) that vary among sites along environmental gradients. Because the theoretical underpinnings of each idealized structure are unique (Clements, 1916; Gleason, 1926; Diamond, 1975; Tilman, 1982; Patterson and Atmar, 1986), it is possible to simultaneously test multiple hypotheses associated with metacommunity structure.

[FIGURE 1 OMITTED]

The San Pedro-Mezquital River (SPMR) is located on the southern portion of the Sierra Madre Occidental (SMO) mountain range in Mexico. The river cuts across the mountains from northeast to southwest, along over 400 km from its headwaters on the western edge of the semiarid, temperate Mexican Plateau to the tropical Pacific coastal plain of Nayarit (Fig. 1). The highlands of the SMO can reach over 3,000 m in elevation, creating a barrier that isolates tropical western vegetations from eastern xerophytic plant associations. In contrast, the river bed descends gradually to the Pacific Ocean. The microclimate thus generated within the canyon allows for the development of a continuum of semiarid to tropical environments, which has led to the proposition that the river is a potential biological corridor (Gonzalez-Elizondo et al., 2007). If true, this means that the SPMR potentially allows faunas from both slopes of the SMO to enter in contact. Recently, the SPMR basin has been identified as a region of high mammalian species richness due to the convergence of tropical, temperate, and desert faunas (Lopez-Gonzalez et al., 2014). Moreover, for closely related populations, interbreeding could be possible, which might render this area a potentially important center of biological diversification.

[FIGURE 2 OMITTED]

In this work we analyze the spatial structure of a set of small-mammal communities along the SPMR to provide insights on the potential role of the river as a biological corridor. Previous research has shown that for rodents and bats there is a strong relationship between species occurrence and environmental factors (e.g., Arita, 1997; Sanchez-Cordero, 2001; Lopez-Gonzalez et al., 2012, 2015); therefore, the characteristics of the proposed environmental gradient should be reflected by the metacommunity structure of the mammalian fauna. In this case, as the river bed descends from around 1,800 m on the Mexican Plateau to sea level, and the local environment changes from arid to tropical, we would expect a high degree of mammal species turnover from arid-land species to tropical forms, resulting in Clementsian (discrete communities replacing each other as a group) or Gleasonian (species replace each other randomly across the gradient) metacommunity structures (Leibold and Mikkelson, 2002). Also, a significant relationship would be expected between mammal species composition along the gradient and vegetation composition as well as between species composition and elevation. On the other extreme, if the river bed was a biological corridor in which all mammal species can move freely throughout, no turnover or significant relationships between local mammal assemblages and elevation and vegetation composition would be found, and nested (species loss across the gradient) or random structures would be expected (Leibold and Mikkelson, 2002). Because different responses to environmental factors have been found between volant and nonvolant species (e.g., Rodriguez and Arita, 2004), they were analyzed separately.

MATERIALS AND METHODS--Study Area--The SPMR is one of the two riverine systems that cross the SMO from east to west. It flows southeastward along the SMO from its headwaters north of the city of Durango to the village of Nombre de Dios (Fig. 1). In this portion the river valley is relatively shallow, with a difference of less than 100 m between stream bed and ground level. Dominant vegetation includes thorn scrub dominated by mesquite (Acacia) and open oak (Quercus) forest; a few tropical elements occur (e.g., Ficus), but arid and temperate elements dominate (site 1). In site 2 vegetation includes Chihuahuan Desert thorn scrub, mesquite, and tropical dry forest elements (Bursera, Ficus), all of which intermingle in some areas; at elevations above 1,700 m open oak forest also occurs. The river turns south and flows into the mountains through a canyon that can be a few hundred meters wide and from 800 to 1,000 m deep in the central part of the mountain range (sites 3 and 4). Site 3 is dominated by tropical deciduous forest; site 4 is dominated by tropical deciduous forest, with some elements of semideciduous tropical vegetation at the bottom of the canyon and oak and pine-oak forest in the higher portions. At site 5 the canyon becomes less deep and more open, and a mixture of open oak forest, tropical deciduous, and tropical semideciduous forest occurs. On the western slope of the SMO, about 80 km from the river mouth, the terrain levels again and the stream runs into an open valley on the coastal plain (site 6), an agricultural area with fragments of semideciduous forest and palm groves that reaches the ocean at the Marismas Nacionales wetland, where mangrove dominates. Although the SPMR basin extends northward as far as the municipality of Canatlan, in Durango, and as far southeast as Sombrerete, Zacatecas (World Wildlife Fund, 2010), this work focuses only on the north-south portion of the river.

Specimen Collection--As part of an inventory of the basin, sampling was conducted in the six sites described above (Figs. 1 and 2). With the exception of site 4, located on the Lajas River, a tributary, all sites were on the SPMR itself. The number of sites examined is lower than that recommended for metacommunity analysis (Presley et al. 2010). Unfortunately, access to the SPMR is extremely difficult in two thirds of its span, specifically the central portion of the canyon. Further sampling demanded time and effort that was not available at the time this project was conducted (2009-2011). Since 2009 the SPMR has been under threat of damming for a hydroelectric complex (http://www. nayarit.gob.mx/ transparenciafiscal/administracion/2008/ beneficiarios/2010/mega_proyectos_seplan.pdf). This is the only basin left in the Pacific versant of Mexico that so far has not been dammed. It is also one of the least-explored areas of the country. Therefore, it was urgent to document the biodiversity of the area, especially those portions of the river that would be flooded. Sampling was conducted systematically to be able to get not only an inventory of species but an idea of turnover and metacommunity processes in the basin, as well as its role as a biological corridor. Although there has been considerable opposition to the hydroelectric project from many concerned parties (e.g., http://defiendemuxatena.org/), the project is ongoing (http://www.seplan.gob.mx/ds/proy_estra/ 2014/proyectos_estrategicos_noviembre2014.pdf). Thus, despite their limitations these data have become the only documentation available to estimate the metacommunity dynamics of this system before the change, and they will be the basis of further analysis of the effects of damming on the local mammalian fauna.

We sampled small mammals at each site for 10 nights between January 2009 and March 2011. At each site, sampling was conducted at several locations, selected to cover all vegetation associations present in the site, from the river bed to 1,800 m above sea level (Fig. 2). Nonvolant mammals were captured using Sherman traps baited with a mixture of oats and peanut butter. Traps were left open for 2 nights at each location. Bats were collected using ground-level mist nets placed along streams, ponds, or in trails within forests or they were taken from roosts. Mist nets were left open for 5 h after sunset. Specimens collected were processed and identified following conventional techniques in mammalogy, and were deposited in the Mammal Collection, CIIDIR-IPN Unidad Durango (SEMARNAT collection permit FAUT-0085 to C.L.G.). Animals were handled and processed following guidelines of the American Society of Mammalogists (Sikes et al., 2011). Taxonomic arrangement follows Ramirez-Pulido et al. (2014). From collection data we built matrices of six sites by 43 volant species (Chiroptera) and six sites by 18 nonvolant species (Rodentia and the mouse opossum Tlacuatzin canescens).

Environmental Data--Using the digital version of INEGI (1991) 1:250,000 vegetation maps included in the GIS of BIOTICA 5.0 (CONABIO, 1999, 2008), we generated a 12-km radius circular buffer for each collection site. Buffer size was chosen to avoid overlap between sites, to include all vegetational variation sampled, and to describe vegetation at a scale that was appropriate for animals with relatively large home ranges (bats), as well as for animals with a smaller home range (rodents). Within each buffer, percent of area covered by each vegetation type was calculated. Percent cover of each vegetation type was rescaled to exclude vegetation occurring above 1,800 m. From these data we generated a matrix of nine vegetation types (oak forest, agriculture and livestock areas, tropical and subtropical thorn forest, desert scrub, palm grove, natural grassland, savanna, tropical deciduous forest, and tropical semideciduous forest) by six sites.

To estimate elevational variation, a matrix of 18 elevation categories by six sites was built from digital elevation models (scale 1:50,000; INEGI, 2003, http://mapserver.inegi.org.mx/ DescargaMDEWeb). For each buffer we estimated the percent area covered between 100-m contours (0-100, 100-200 m, and so forth) from 0 to 1,800 m above sea level. Data were generated using ArcView GIS version 3.2 (ESRI, Inc., Redlands, California) and Idrisi Kilimanjaro version 14.01 (Clark Labs, Worcester, Massachusetts).

Metacommunity Analyses--We followed Leibold and Mikkelson (2002) in defining a metacommunity as a set of ecological communities at local sites that are potentially, but not necessarily, linked by dispersal, with a community defined as a collection of species that occupy a given site. We used the framework of metacommunity structure analysis of Leibold and Mikkelson (2002) and Presley et al. (2010) to determine if the volant and nonvolant metacommunities have nonrandom structure and, if so, to identify which idealized metacommunity structure best described empirical patterns (Lopez-Gonzalez et al., 2012: Table 1).

Each matrix was ordered via reciprocal averaging (RA), which maximizes the proximity of sites with similar species compositions and the proximity of species with similar patterns of occurrence at sites. RA (= correspondence analysis) is appropriate for identifying patterns in response to latent environmental gradients because similarity in species patterns of occurrence determines the position of sites and species along an axis of correspondence (i.e., a latent environmental gradient) without a priori knowledge of, or assumptions about, the particular factors that govern the response of each species (Leibold and Mikkelson, 2002). All analyses were conducted for the first axis of correspondence. We ran the analysis to explore patterns along a second axis of variation (Presley et al., 2009). Coherence was nonsignificant (results not shown), which indicates a random arrangement of sites and species (Presley et al., 2010). Ordered incidence matrices were tested against null distributions for coherence, turnover, and against a V distribution for boundary clumping to find the best-fit pattern of idealized metacommunity structures (Lopez-Gonzalez et al., 2012: Table 1). Overviews of these tests and their interpretation are presented in Leibold and Mikkelson (2002) and Presley et al. (2010).

Coherence--We assessed coherence by comparing the number of embedded absences (i.e., absences that have at least one presence toward each extreme in a row or column) in the ordinated data matrix to a distribution of embedded absences produced by a null model in which species richness per site is fixed and species occurrence is equiprobable (Leibold and Mikkelson, 2002). A metacommunity was considered significantly and positively coherent if the likelihood of having fewer embedded absences than observed was <a/2 (a two-tailed test). A metacommunity was considered significantly and negatively coherent if the likelihood of having more embedded absences than observed was <a/2. Negative coherence is characteristic of a "checkerboard" pattern (resulting from species pairs with mutually exclusive distributions) whereas positive coherence is characteristic of 12 other structures (Lopez-Gonzalez et al., 2012: Table 1). Nonsignificant coherence indicates random structure. Species occurrences in random structures are scattered along the latent environmental gradient such that measures of species turnover and range boundary clumping do not effectively reflect the concepts that they are intended to measure. Consequently, range turnover and boundary clumping were only evaluated when metacommunities exhibited significant coherence.

Turnover--Species ranges were made perfectly coherent by filling in all embedded absences prior to analyses of turnover (Leibold and Mikkelson, 2002). Species range turnover was evaluated based on the number of replacements of one species by another along the gradient. The observed number of replacements was compared to a null distribution of replacement values created from 1,000 matrices that contained randomly shifted species ranges to determine significance (Leibold and Mikkelson, 2002). Significantly negative turnover is indicative of nested distributions. Significantly positive turnover is indicative of Gleasonian, Clementsian, or evenly spaced structures. Nonsignificant turnover is characteristic of quasi-structures (Lopez-Gonzalez et al., 2012: Table 1).

Boundary Clumping--Morisita's index measures the clumping of species distributional boundaries by counting the number of terminal boundaries at each site. Significance was determined via a [chi square] goodness-of-fit test that compared the observed distribution of range boundaries to an expected uniform distribution. Range boundaries that occurred at random have a Morisita's index around 1.0 and are consistent with Gleasonian, quasi-Gleasonian, or stochastic species loss in nested structures (Lopez-Gonzalez et al., 2012: Table 1). Morisita's index values >1.0 with a significant [chi square] test indicate clumped boundaries and are consistent with Clementsian, quasi-Clementsian, or clumped species loss in nested structures. Index values <1.0 with a significant [chi square] test indicate hyperdispersed boundaries and are consistent with evenly spaced, quasi-evenly spaced, or hyperdispersed species loss in nested structures (Lopez-Gonzalez et al., 2012: Table 1). For some Clementsian structures, large numbers of clumped species boundaries were identified along the ordination of sites; these boundaries defined compartments (i.e., distinctive groups of species along portions of an environmental gradient; Lewinsohn et al., 2006). The structure of each of these compartments was analyzed separately. All analyses were conducted using MATLAB version 7.10.0.499 Release 2010a (MathWorks, Natick, Massachusetts). Null hypotheses were rejected at [alpha] = 0.05. Script files for MATLAB are available for download at http://faculty.tarleton. edu/higgins/metacommunity-structure.html.

Relationship Between Species Composition and Environmental Variables--Previous work has shown that vegetation can be an effective surrogate for environmental variation because it reflects the combined effects of the entire suite of environmental factors (Lopez-Gonzalez, 2004; Stevens et al., 2007; Lopez-Gonzalez et al., 2015). We related the observed pattern of species distributions among sites to the vegetational and elevational composition of sites using Redundancy Analysis (RDA; ter Braak and Prentice, 1988; ter Braak and Smilauer, 2002) as implemented in CANOCO for Windows v. 4.5 (Microcomputer Power, Ithaca, New York; ter Braak and Smilauer, 2002). For each species group, significance of the relationship between species composition and environmental variables (the matrix of vegetation or elevation characteristics by site) was tested as the correlation between site scores that are linear combinations of species data and site scores that are linear combinations of vegetation attributes (ter Braak and Prentice, 1988; ter Braak and Smilauer, 2002). Only the first canonical axis of variation was tested. We tested the null hypothesis of no relationship using the Monte Carlo test available in CANOCO. The test statistic (F) for this test is the ratio of the variance explained by the first canonical correlation and the residual sum of squares (ter Braak and Smilauer, 2002).

RESULTS--Sixty-one species of small mammals were recorded (43 bats and 18 nonvolant species) during our inventory. They represent 87 and 39%, respectively, of the species so far recorded for the SPMR basin (49 volant and 46 nonvolant species, Lopez-Gonzalez et al., 2014). The difference in collection success between groups is in part related to the elevational distribution of small nonvolant mammals, which are more species-rich at higher elevations than those examined (below 1,800 m). The nonvolant assemblage includes only cricetid rodents and the mouse opossum Tlacuatzin canescens; no squirrels (Sciuridae) or gophers (Geomyidae) were included.

The bat metacommunity of the Mezquital exhibited highly significant coherence and significant species turnover, but nonsignificant boundary clumping (Table 1), which is consistent with a Gleasonian structure characterized by the individualistic responses of species to environmental variation. The RA ordination diagram shows an arrangement of species and sites that makes these results evident (Fig. 3A). The left side of the diagram shows species occurring only on sites 6 and 4 (dominated by tropical forest), which are gradually replaced by species able to occur in more sites. To the right of the diagram are species that occurred toward the higher, more arid sites but not in the tropical sites. However, distributional boundaries of species are not clumped (i.e., tropical species are not replaced as a group by nontropical species), but in the center of the diagram are species that occur in most sites, and species appear and disappear from the gradient individually or in small groups. Moreover, the arrangement of sites did not correspond to the elevational or vegetational gradient, which is consistent with a nonsignificant relationship between species composition and vegetation composition across the gradient (F = 0, P = 1) and between species composition and elevation (F = 0, P = 1). No species was captured in all sites, but the nectar-feeding bats Glossophaga soricina and Leptonycteris yerbabuenae, and the insectivore Macrotus californicus, all tropical species, reached as far inland as site 2. Conversely, the vespertilionid Myotis yumanensis (insectivore), a bat common in the Mexican Plateau and highlands of the SMO, reached as far as site 5.

For rodents coherence was significant, but turnover and boundary clumping were nonsignificant, which is consistent with a quasi-Gleasonian structure (Table 1). Individualistic responses of species with small niche breadths (relative to the extent of the gradient) likely result in greater positive turnover and a Gleasonian structure, whereas species with larger niche breadths can result in nonsignificant positive turnover and a quasi-Gleasonian structure (Presley et al., 2010). RA ordination of sites and species resulted in a diagram that is similar to that of bats (Fig. 3B); the arrangement of sites did not correspond to the spatial arrangement of sites or to the vegetational gradient, and thus relationships between species composition-vegetation and species compositionelevation also were nonsignificant (F = 0, P = 1; F = 0, P = 1, respectively). For nonvolant mammals, turnover seemed to be less marked because several species occurred in at least four of six sites, and Sigmodon arizonae occurred along the whole gradient (i.e., nonvolant mammals had larger niche breadths).

[FIGURE 3 OMITTED]

DISCUSSION--Although the number of sites examined is relatively small, metacommunity analysis revealed clear patterns of species turnover along the river (Fig. 3). Two main factors allowed us to detect a pattern of species replacement using a small number of sites. On one hand, the basin is bound by the Pacific Ocean on the west and by the Mexican Plateau to the east, therefore there is a sharp discontinuity on the west side, and a comparatively uniform environmental situation on the east. On the other hand, the spacing between sites was enough to recover the environmental transitions occurring in the canyon.

Different metacommunity structures were identified for volant (Gleasonian) and nonvolant (quasi-Gleasonian) mammals. In both cases species along the gradient replace each other; nonetheless replacement is not by groups, but rather each species or small groups of species seem to be responding idiosyncratically to the local environment. Quasi-Gleasonian metacommunity structures are related to species with larger niche breadths as compared with Gleasonian structures (Presley et al., 2010). Rodents experience little movement throughout their lives, and they are strongly linked to characteristics of their microhabitats (Morris, 1987; Lambert et al., 2006); therefore, they would have smaller niche breadths than do bats, which due to their vagility can potentially use a larger variety of resources. We found the opposite pattern in the SPMR, where results suggest that there are proportionally more species of rodents able to use resources throughout the river basin than there are bats. Thus, different metacommunity structures suggest that nonflying mammals are able to use a wider variety of resources than are volant species, or else that the resources they use occur in a wider variety of habitats.

Previous research has shown that for rodents and bats there is a strong relationship between species occurrence and environmental factors (e.g., Arita, 1997; Sanchez-Cordero, 2001; Lopez-Gonzalez et al., 2012, 2015). Nonetheless, we found no relationship between species composition and vegetation or elevation. The lack of relationship and the metacommunity structures found can be explained by the unique characteristics of the river valley. The geographic position and physiographic (and therefore climatic and vegetational) complexity of the canyon create more-humid conditions within the canyon than in the surroundings (Brito-Castillo et al., 2010). Environmental conditions at any point on the river valley are the result of local elevation, depth, width, and orientation of the basin, which collectively create an unusually diverse suite of potential habitats.

Although for practical purposes we used the vegetation categories of INEGI (1991) modified by CONABIO (1999), in reality there are no clear boundaries between plant associations. The microclimate in the canyon allows some plant species to penetrate the mountain range farther than they would outside of the canyon; as elevation increases and climate becomes drier, the proportion of tropical elements decreases and that of xerophytic elements increases along the climate-elevation gradient. For instance, in the area around the town of Mezquital (site 2, Fig. 1), where the river valley is wide (about 7 km), we observed that elements of xerophytic vegetation occur in tropical and subtropical associations and vice versa (also see Gonzalez-Elizondo et al., 2007). In the central portion of the basin (where the river valley narrows) the mountains rise over 2,400 m above sea level, whereas the river bed is 1,000 m lower (Fig. 2). In these areas a strong vertical vegetation gradient occurs, from pine to pine-oak, to oak, and to tropical deciduous forest at the river bed. This transition occurs in short distances, and elements of any given association are mixed with elements of the one below and above it. In this scenario, mammal species that otherwise would be confined to one vegetation association can occur in more than one at the SPMR, probably tracking specific resources that are able to occur in more than one plant association along the elevational-vegetational gradient. This explains the lack of relationship between mammal species composition and vegetation types and suggests that as elevation increases, mammalian species respond to these changes idiosyncratically depending on the kind of resources they use and how far those resources reach into the canyon. The observed Gleasonian and quasi-Gleasonian metacommunity structures detected are consistent with this hypothesis.

Some insectivorous bats occur on both the slopes and the highlands of the SMO (e.g., Lasiurus cinereus, Tadarida brasiliensis) and, therefore, for them the canyon is not a corridor because they can move freely across the Sierra. Nonetheless, for most species examined the river canyon is a partial corridor in which tropical species stay associated to tropical elements as far as they reach into the basin, but not beyond, and the opposite happens with species of arid environments in the opposite direction. Examples of this are the mouse opossum T. canescens or the mustached bat Pteronotus parnellii, which inhabit the tropical deciduous and semideciduous forests of the coastal plain, but which penetrate far north into the mountains following the tropical elements as far as Candelaria del Alto, Durango (site 3, Fig. 1). Other species such as the bats Choeronycteris mexicana and L. yerbabuenae and rodents such as Reithrodontomys fulvescens and Baiomys taylori (not collected on the coastal plain for this project but recorded near Huajicori and Acaponeta, Nayarit, about 50 km from the river valley) occur on both sides of the SMO, but eastern and western populations are usually considered as isolated by the mountain barrier. However, they were collected within the canyon, which demonstrates that at this latitude gene flow is possible between the eastern and western slopes of the SMO.

Our analysis of the SPMR basin revealed Gleasonian metacommunity structures that likely are linked by dispersal. The geographic position of the SPMR basin produces an environmental gradient, unique to this system, in which dispersal of populations throughout the basin is likely mediated mostly by specific resource availability. Thus, at the SPMR basin local communities are structured by the combination of available resources and species-specific potential for dispersion (although the magnitude of the effect of interspecific interactions has not been assessed). Local biotic and abiotic conditions are critical not only to determine species composition, but very likely they also affect population densities. Because these environments are seasonal, availability of resources varies in space and time. Thus, this is a system in which mass effects and species sorting might occur (Holyoak et al., 2005), with local extinctions likely occurring within the river valley, particularly in the narrow central part, and with populations being replaced with individuals from the coastal plains or Mexican Plateau.

Human activities in the area should be carefully considered for the impact they could have on the biomes throughout the basin. For instance, the current project to dam the river might have unforeseen results throughout the basin that could affect biodiversity and human activities in areas far distant from the dam itself. At the regional level, changes in local climatic conditions can change the dynamics of the whole basin with consequences that could affect precipitation regimes as far as the Mexican Plateau (Brito-Castillo et al., 2010), where human activities are strongly dependent on rainfall. Local environmental conditions result in the unique mammalian species combinations observed within the basin (Lopez-Gonzalez et al., 2014); this is probably true for other organisms as well. The metacommunity structures observed at the SPMR system further support what previous authors have pointed out regarding megadiverse Mexico: it is not species richness in itself that accounts for the Mexican high diversity, but the high degree of environmental heterogeneity that in turn produces high species turnover rates at relatively low scales, resulting in high [beta] and [gamma] diversity (Arita and Rodriguez, 2002; Rodriguez et al., 2003).

Funding for this research was provided by Secretaria de Investigation y Posgrado, Instituto Politecnico Nacional (SIP 2010-0434, 2011-0349, 2012-1104 to C.L.G.) and CONABIO (GT015 to C.L.G.). We extend our gratitude to all the students who collaborated with us in the field, particularly to T. C. Monterrubio Rico and his students from Universidad Michoa cana de San Nicolas de Hidalgo, Michoacan, Mexico. We appreciate the facilities granted to us by the people of the communities we visited to conduct our project.

LITERATURE CITED

ARITA, H. T. 1997. The non-volant mammal fauna of Mexico: species richness in a megadiverse country. Biodiversity and Conservation 6:787-795.

ARITA, H. T., AND P. RODRIGUEZ. 2002. Geographic range, turnover rate and the scaling of species diversity. Ecography 25:541-553.

BRITO-CASTILLO, L., E. R. VIVONI, D. J. GOCHIS, A. FILONOV, I. TERESHCHENKO, AND C. MONZO N. 2010. An anomaly in the occurrence of the month of maximum precipitation distribution in northwest Mexico. Journal of Arid Environments 74:531-539.

CLEMENTS, F. E. 1916. Plant succession: an analysis of the development of vegetation. Carnegie Institution of Washington, Washington, D.C.

CONABIO. 1999. Uso de suelo y vegetacion modificado por CONABIO. Escala 1:1,000,000. Comision Nacional para el Conocimiento y uso de la Biodiversidad (CONABIO), Mexico City.

CONABIO. 2008. Sistema de informacion BIOTICA, ver. 5.0. Comision Nacional para el Conocimiento y uso de la Biodiversidad (CONABIO), Mexico City.

COTTENIE, K. 2005. Integrating environmental and spatial processes in ecological community dynamics. Ecology Letters 8:1175-1192.

DE LA SANCHA, N. U., C. L. HIGGINS, S. J. PRESLEY, AND R. E. STRAUSS. 2014. Metacommunity structure in a highly fragmented forest: has deforestation in the Atlantic Forest altered historic biogeographic patterns? Diversity and Distributions 20:1058-1070.

DIAMOND, J. M. 1975. Assembly of species communities. Pages 342-444 in Ecology and evolution of communities (M. L. Cody and J. M. Diamond, editors). Harvard University Press, Cambridge, Massachusetts.

GLEASON, H. A. 1926. The individualistic concept of the plant association. Bulletin of the Torrey Botanical Club 53:7-26.

GONZALEZ-ELIZONDO, M. S., M. GONZALEZ-ELIZONDO, AND M. A. MARQUEZ LINARES. 2007. Vegetacion y ecorregiones de Durango. Plaza y Valdez, Mexico D.F.

HOLT, R. D. 1993. Ecology at the mesoscale: the influence of regional processes on local communities. Pages 77-88 in Species diversity in ecological communities (R. E. Ricklefs, and D. Schluter, editors). University of Chicago Press, Chicago, Illinois.

HOLYOAK, M., R. D. HOLT, AND M. A. LEIBOLD. 2005. Metacommunities: spatial dynamics and ecological communities. University of Chicago Press, Chicago, Illinois.

INSTITUTO NACIONAL DE ESTADISTICA GEOGRAFIA E INFORMATICA (INEGI). 1991. Cartas de Uso de Suelo y Vegetacion, Escala 1:250,000.

KEITH, S. A., A. C. NEWTON, M. D. MORECROFT, D. J. GOLICHER, AND J. M. BULLICK. 2011. Plant metacommunity structure remains unchanged during biodiversity loss in English woodlands. Oikos 120:302-310.

LAMBERT, T. D., J. R. MALCOLM, AND B. L. ZIMMERMAN. 2006. Amazonian small mammal abundances in relation to habitat structure and resource abundance. Journal of Mammalogy 87:766-776.

LEIBOLD, M. A., E. P. ECONOMO, AND P. PERES-NETO. 2010. Metacommunity phylogenetics: separating the roles of environmental filters and historical biogeography. Ecology Letters 13:1290-1299.

LEIBOLD, M. A., M. HOLYOAK, M. MOUQUET, P. AMARASEKARE, J. M. CHASE, M. F. HOOPES, R. D. HOLT, J. B. SHURIN, R. LAW, D. TILMAN, M. LOREAU, AND A. GONZALEZ. 2004. The metacommunity concept: a framework for multi-scale community ecology. Ecology Letters 7:601-613.

LEIBOLD, M. A., AND G. M. MIKKELSON. 2002. Coherence, species turnover, and boundary clumping: elements of meta-community structure. Oikos 97:237-250.

LEIBOLD, M. A., AND T. E. MILLER. 2004. From metapopulations to metacommunities. Pages 133-150 in Ecology, genetics and evolution of metacommunities (I. A. Hanski and O. E. Gaggiotti, editors). Elsevier Academic Press, Burlington, Massachusetts.

LEWINSOHN, T. M., P. I. PRADO, P. JORDANO, J. BASCOMPTE, AND J. OLESEN. 2006. Structure in plant-animal interaction assemblages. Oikos 113:174-184.

LOPEZ-GONZALEZ, C. 2004. Ecological zoogeography of the bats of Paraguay. Journal of Biogeography 31:33-45.

LOPEZ-GONZALEZ, C., A. LOZANO, D. F. GARCIA-MENDOZA, AND A. I. VILLANUEVA-HERNANDEZ. 2014. Mammals of the San Pedro-Mezquital River Basin, Durango-Nayarit, Mexico. Check List 10:1277-1289.

LOPEZ-GONZALEZ, C., S. J. PRESLEY, A. LOZANO, R. D. STEVENS, AND C. L. HIGGINS. 2012. Metacommunity analysis of Mexican bats: environmentally mediated structure in an area of high geographic and environmental complexity. Journal of Biogeography 39:177-192.

LOPEZ-GONZALEZ, C., S. J. PRESLEY, A. LOZANO, R. D. STEVENS, AND C. L. HIGGINS. 2015. Ecological biogeography of Mexican bats: the relative contributions of habitat heterogeneity, beta diversity, and environmental gradients to species richness and composition patterns. Ecography 38:261-271.

MORRIS, D. W. 1987. Ecological scale and habitat use. Ecology 68:362-369.

PATTERSON, B. D., AND W. ATMAR. 1986. Nested subsets and the structure of mammalian faunas and archipelagos. Biological Journal of the Linnean Society 28:65-82.

PRESLEY, S. J., C. L. HIGGINS, C. LOPEZ-GONZALEZ, AND R. D. STEVENS. 2009. Elements of metacommunity structure of Paraguayan bats: multiple gradients require analysis of multiple axes of variation. Oecologia 160:781-793.

PRESLEY, S. J., C. L. HIGGINS, AND M. R. WILLIG. 2010. A comprehensive framework for the evaluation of metacommunity structure. Oikos 119:908-917.

RAMIREZ-PULIDO, J., N. GONZALEZ-RUIZ, A. L. GARDNER, AND J. ARROYO-CABRALES. 2014. List of recent land mammals of Mexico, 2014. Special Publications of the Museum of Texas Tech University 63:1-69.

RICKLEFS, R. E. 1987. Community diversity: relative roles of local and regional processes. Science 235:167-171.

RODRIGUEZ, P., AND H. T. ARITA. 2004. Beta diversity and latitude in North American mammals: testing the hypothesis of covariation. Ecography 27:547-556.

RODRIGUEZ, P., J. SOBERO N, AND H. T. ARITA. 2003. El componente beta de la diversidad de mamiferos de Mexico. Acta Zoologica Mexicana (nueva serie) 89:241-259.

SANCHEZ-CORDERO, V. 2001. Elevation gradients of diversity for rodents and bats in Oaxaca, Mexico. Global Ecology and Biogeography 10:63-76.

SIKES, R. S., W. L. GANNON, AND THE ANIMAL CARE AND USE COMMITTEE OF THE AMERICAN SOCIETY OF MAMMALOGISTS. 2011. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. Journal of Mammalogy 92:235-253.

STEVENS, R. D., C. LOPEZ-GONZALEZ, AND S. J. PRESLEY. 2007. Geographical ecology of Paraguayan bats: spatial integration and metacommunity structure of interacting assemblages. Journal of Animal Ecology 76:1086-1093.

TER BRAAK, C. J. F., AND I. C. PRENTICE. 1988. A theory of gradient analysis. Advances in Ecological Research 18:271-317.

TER BRAAK, C. J. F., AND P. SMILAUER. 2002. CANOCO reference manual and CanoDraw for Windows user's guide: software for canonical community ordination (version 4.5). Microcomputer Power, Ithaca, New York.

TILMAN, D. 1982. Resource competition and community structure. Monographs in Population Biology, Princeton University Press 17, Princeton, New Jersey.

WILLIG, M. R., S. J. PRESLEY, C. P. BLOCH, I. CASTRO-ARELLANO, L.M. CISNEROS, C. L. HIGGINS, AND B. T. KLINGBEIL. 2011. Tropical metacommunities along elevational gradients: effects of forest type and other environmental factors. Oikos 120:1497-1508.

WORLD WILDLIFE FUND (WWF). 2010. La Cuenca alta del rio San Pedro-Mezquital, caudal de vida y cultura. WWF/Nokia/Coca Cola.

Submitted 18 March 2015.

Acceptance recommended by Associate Editor, Michelle L. Haynie, 3 August 2015.

CELIA LOPEZ-GONZALEZ * AND ABRAHAM LOZANO

Instituto Politecnico Nacional, CIIDIR Unidad Durango, Sigma 119, Fraccionamiento 20 de Noviembre II, Durango, Durango 34220, Mexico

* Correspondent: celialg@prodigy.net.mx
TABLE 1--Results of analyses of coherence, species turnover, and
boundary clumping for small mammals along the San Pedro-Mezquital
River. Abs = number of absences; Rep = number of replacements; M =
Morisita's index; SD = standard deviation. *Significant results
([alpha] = 0.05).

                   Coherence

Assemblage   Abs      P        mean    SD

Bats         29    0.00 *      68.9    9.2
Rodents       9    0.0001 *    22.2    3.9

                     Species turnover

Assemblage    Rep      P       mean     SD

Bats         2381    0.04 *   1275.8   552.9
Rodents       377    0.19      232.9   110.9

              Boundary
              clumping

Assemblage    M      P         Pattern

Bats         1.13   0.42   Gleasonian
Rodents      1.33   0.28   Quasi-Gleasonian
COPYRIGHT 2015 Southwestern Association of Naturalists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lopez-Gonzalez, Celia; Lozano, Abraham
Publication:Southwestern Naturalist
Article Type:Report
Geographic Code:1MEX
Date:Dec 1, 2015
Words:5889
Previous Article:Wild pig (Sus scrofa) reproduction and diet in the rolling plains of Texas.
Next Article:Test and evaluation of various techniques to study refuged lizards in the field.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters