Keywords

14.1 Introduction

Of the whole set of arid vegetation types in the New World (deserts and xeric shrublands, sensu Olson et al. 2001), the avifauna of the North American region is the most diverse, being surpassed only slightly by that in the Namibian Karoo region in southwestern Africa (Riddle and Hafner 2006). The highest diversity within the North American arid regions is concentrated mainly in the Chihuahuan and central Mexican deserts (https://www.worldwildlife.org/biomes/deserts-and-xeric-shrublands). Deserts and other arid habitats in North America hold a large number of resident species (ca. 100 species; Berlanga et al. 2010; Navarro-Sigüenza et al. 2014), with breeding populations present year-round either throughout the region or in some desert patches. These habitats are also used by transient and wintering species coming from other open habitats located in the Great Plains (southern Canada-central USA). Moreover, the arid regions of the continent include areas with a high number of endemic species of taxonomic groups such as birds (Zink et al. 2001), herps (Morafka 1977), and mammals (Riddle et al. 2000; Leaché et al. 2007). This has been attributed to multiple events of in situ evolution, the presence of isolating barriers, and clear patterns of evolutionary diversification (Zink et al. 2001; Riddle and Hafner 2006).

The avifauna of the North American desert regions is fairly well known. Major ornithological surveys have been published for avifaunas present in regions such as Arizona (Phillips et al. 1964), Nevada (Alcorn 1988; Floyd 2007), New Mexico (Bailey 1928), Texas (Arnold 1984), Baja California Peninsula (Wilbur 1987; Erickson and Howell 2001), Sonora (Van Rossem 1945; Russell and Monson 1998; Villaseñor-Gómez 2008), Coahuila (Garza de León et al. 2007), and Nuevo León (Contreras-Balderas et al. 2008), among others. In addition, recent interest in the ecology and conservation of arid land birds have produced a good number of regional or local surveys that have analysed the distribution and community patterns between vegetation types (Tomoff 1974; Naranjo and Raitt 1993; Arizmendi and Espinosa de los Monteros 1996; Macías-Duarte et al. 2009; Suárez-García et al. 2017).

The CD includes the largest arid and semi-arid region in Mexico. Because of its size, topography, and habitat diversity, it holds a high number of species, including birds (Contreras-Balderas et al. 2004), reptiles, and amphibians (Mendoza-Quijano et al. 2006). The CCB is immersed in the CD and is an internationally important wetland included in the Ramsar list. It is also among the Priority Ecoregions for Conservation included in the World Wildlife Fund (Souza et al. 2006). Despite its relevance, there are very few studies on bird species composition (Contreras-Balderas 1984; García-Salas et al. 1995; Contreras-Balderas et al. 1997, 2004) and no studies concerning the species distribution among the main vegetation types found in the basin.

In this study, we first analyse biogeographic diversity patterns in some of the arid regions of North America. Once we establish this scenario, we compare the bird communities found in the CCB with local studies conducted in other arid regions and, finally, we analyse the avifauna found in the CCB and provide some reasons that could explain its diversity.

14.2 Methods

14.2.1 Study Area

For the first part of the study, we included all of the North American desert ecoregions (Olson et al. 2001) (Fig. 14.1) but restricted the analysis within Mexican boundaries, since we had the most complete datasets for this area. In addition, we included environmentally similar areas dominated by shrubland (Tamaulipas, central Mexico, and the Tehuacán Valley) or mezquite shrubland (Tamaulipas, northern Nuevo León, and northeastern Coahuila). This area is similar to the one considered by Hubbard (1973) in his seminal study. We also compared the number of species found in nine other study sites in North American deserts and assessed if richness was related to rainfall and latitude.

Fig. 14.1
figure 1

Arid ecoregions of Mexico. The black dot in the Chihuahuan desert represents the location of Cuatro Ciénegas

Full size image

The second part of the study focused on the CCB, which is part of the CD. The site is located between 26°45′00″ N and 27°00′00″ N and 101°48′49″ W and 102°17′53″ W (INE 1999) at 740 m asl (Fig. 14.1). Mean annual rainfall is ~150 mm. The valley hosts a large system of natural springs, streams, and ponds. The vegetation consists of (1) mezquite shrubland or matorral where Prosopis glandulosa is dominant (Mezquital), (2) wet grassland dominated by Sporobolus spp. and Distichlis spicata (Grassland), (3) open creosote bush (Larrea tridentata) scrub where Fouquieria splendens is an important element (Larrea), and (4) desert spoon (Dasylirion wheeleri) scrub (Sotolera) (Pinkava 1984). The birds were counted in four localities, three of which had water bodies. Creosote, mesquite, grasslands, and sotol scrub were included in these localities (Fig. 14.2).

Fig. 14.2
figure 2

Map of the Cuatro Ciénegas Basin showing the vegetation types and bird sampling sites. 1. Churince creosote bush 2. Churince mesquite thickets 3. Churince Dasylirion scrub 4. Churince grassland 5. Poza Azul grassland 6. Ejido La Vega mesquite thicket 7. Rancho Pozas Azules grassland 8. Rancho Pozas Azules mesquite thickets 9. Rancho Pozas Azules creosote bush scrubland

Full size image

14.2.2 Data Analysis

Data for the biogeographical analysis was obtained from the Mexican Bird Atlas (Navarro-Sigüenza et al. 2003); from public databases, such as eBird (http://ebird.org/content/ebird/) and the Avian Knowledge Network (http://www.avianknowledge.net/); as well as from VertNet (http://vertnet.org/), which includes presence records derived from scientific collections. All the records were downloaded in ArcView 3.2 (ESRI 1991–1995). We then overlapped the presence records with the desert ecoregions (Fig. 14.1). Seasonality was based on Berlanga et al. (2015) and Stotz et al. (1996). Bird data for the CCB was obtained from Suárez-García et al. (2017). We searched the literature to find information of the local bird communities found in other North American regions.

We compared the number of species of resident, migrants, and transient species between the CCB and the rest of the avifauna of the North American deserts with an X2 goodness-of-fit test. The Bray-Curtis coefficient with the UPGMA linkage method was implemented to assess the similitude of the bird communities among vegetation types. We assessed the proportion of species shared between vegetation with the Jaccard coefficient of similitude.

14.3 Results

14.3.1 North American Deserts

We found 498 species in the North American deserts in addition to 40 accidental or transient species. Geographically, the ecoregions with matorral (hereafter shrubland) as the dominant habitat contained the highest number of species, with 71% in the Central Mexican shrubland and 59% in the Tamaulipan shrubland ecoregion; on the other hand, the Sonoran and Chihuahua ecoregions concentrated 58% and 59%, respectively, of the avifauna considered in this study. Ecological distribution patterns, on the other hand, showed that only 71 species (14%) of the total avifauna are primarily restricted to arid environments, whereas 27 species (5%) are only marginal in these areas. Finally, most species in arid environments are resident species (56%), which increases to 78% if these species with both resident and migratory populations are included. Migratory species constitute 21% of the avifauna in arid ecoregions. Around 14% of birds found in North American deserts may be considered ecologically endemic to these environments according to their habitat preferences (Stotz et al. 1996).

14.3.2 Cuatro Ciénegas and Other North American Deserts

The complete checklist of the CCB includes 101 species (18.8% of the richness that has been recorded in North American deserts and 34.7% of the richness recorded in the CD). Seven of the recorded species are exclusive of arid environments: lucifer hummingbird (Calothorax lucifer), cactus wren (Campylorhynchus brunneicapillus), rock wren (Salpinctes obsoletus), black-tailed gnatcatcher (Polioptila melanura), curve-billed thrasher (Toxostoma curvirostre), rufous-crowned sparrow (Aimophila ruficeps), and black-throated Sparrow (Amphispiza bilineata). Forty-seven bird species were residents (46.5%), 15 were summer visitors (14.9%), and 30 were winter migrants (29.7%). Nine species were transients (9%) (Fig. 14.2). The number of migrants was higher than expected in the CCB with regard to the number of residents, migrants, and transients found in the North American deserts as a whole (x2 = 4.57, 1 d.f., p < 0.05). The main habitat of most species in the CCB was shrubland (38%), followed by open vegetation and grassland (38 and 16%) (Fig. 14.3). In contrast to the complete set of desert species, 38% of the species from the basin were mainly from desert environments.

Fig. 14.3
figure 3

Percentage of migratory status of birds from Mexican deserts and from the Cuatro Ciénegas Basin

Full size image

Arguably, the basin has more species than local sites of the Chihuahuan, Mojave, and Sonoran deserts. The only study that has reported more species within North American arid environments is that of Dávila et al. (2002) for the Tehuacán Valley (Table 14.1). The differences in the number of species among deserts are not due to precipitation since the correlation between the two variables was not significant (r2 = 235, p = 0.512). Furthermore, the relationship became negatively significant when the Tehuacán site, an outlier, was omitted (r2 = 0.725, p = 0.027) (Fig. 14.4). On the other hand, there was a negative relationship between latitude and richness (Fig. 14.5).

Table 14.1 Location, survey period, number of species, reference, and main vegetation types found in various local bird studies in North American deserts
Full size table
Fig. 14.4
figure 4

Main habitat (a) and percentage of species found in the main habitats were of the Cuatro Ciénegas Basin, Northern Mexico (b)

Full size image
Fig. 14.5
figure 5

Relationship between mean annual rainfall and bird species richness in different North American desert sites. ScRiv-N = Paradise Valley, Clark County, Nevada; ScColW-S = Organ Pipe Cactus National Monument, Pima County, Arizona; ScCol-S = Avra Valley and Houghton Road, Pima County, Arizona; ScGr-Ch = Jornada del Muerto, Dona Ana County, New Mexico; ScRiv-Ch = Tortugas Mountain, Dona Ana County; ScRiv-Chb = Tularosa Basin, Otero County, New Mexico; Gr-Ch = Sueco and Coyame Municipalities, Chihuahua; ScRiv-Ch = Sacramento Mountains, Otero County, New Mexico; ScW.Sch = Cuatro Ciénegas Basin, Cuatro Ciénegas Municipality, Coahuila; ScFCol-Teh = Tehuacán-Cuicatlán Biosphere Reserve, various municipalities, Puebla and Oaxaca

Full size image

14.3.3 Local Diversity

The Bray-Curtis classification, based on the bird relative abundances, grouped together the vegetation types within the CCB: the three mezquite thickets, the two larrea sites, and two of the grasslands (Churince and Rancho). The third grassland (Pozas Azules) was placed in the later cluster at a similarity level of 0.20. The sotol scrub, which is found on sandy dunes, was segregated from the other vegetation types (Fig. 14.6). According to the Jaccard coefficient, the mezquite thickets had the highest compositional similarity among them. In addition, two of the grasslands were more similar to each other than to the other types of plant associations. However, the third grassland (Poza Azul) shared more species with the mezquite thickets than with the other pastures (Table 14.2) and the two larrea sites had low values between them (Fig. 14.7).

Fig. 14.6
figure 6

Relationship between latitude and bird species richness among different North American desert sites. Legends for site locations are the same as in Fig. 14.5

Full size image
Table 14.2 Jaccard similarity coefficients of the bird species found in the Cuatro Ciénegas Basin, Mexico
Full size table
Fig. 14.7
figure 7

Bray-Curtis classification based on the bird species relative abundances found in the Cuatro Ciénegas Basin

Full size image

14.4 Discussion

The high number of species recorded in North American deserts is the result of the widespread distribution of arid ecoregions as well as of a north to south distributional gradient, in which some of these areas are bounded by different ecoregions, some of them with more mesic conditions that allow a higher number of species to invade ecologically transitional areas. On the other hand, when only endemic species (including ecological endemics) are considered (51 species), the number is high in comparison to other habitats in Mesoamerica (Navarro-Sigüenza et al. 2014). This process may have its roots since the Pleistocene (Hubbard 1973), when environmental changes may have caused extinction of some taxa but also may have forced others to adapt to xeric conditions, leading to diversification processes that may explain the levels of endemism in the avifauna of the North American deserts. Other processes may have been also benefited from these changes, such as the opportunities for avifaunal interchange between isolated xeric patches when environmental conditions favored the expansion of the arid zones. Thus, the avifauna of desert habitat is a mixture of endemic taxa (both ecological and geographical) with those that have dispersed from other formerly isolated patches, as well as from the avifauna that inhabits neighboring mesic habitats. In fact, as it was mentioned in the results, only a small fraction of the bird species of North America (14%) is restricted to arid environments. In contrast, we found that 38% of the CCB species is restricted to dry ecosystems. This is probably a reflection of the low rainfall and the severe conditions of the valley.

The number of species found in the CCB was high in comparison to what has been reported for other North American deserts (Suárez-García et al. 2017). This was particularly true for migratory species. Cuatro Cienégas is part of the North American Central Migratory Flyway, which runs from Alaska, Canada, and western USA to central and south-east Mexico. This important route contributes to the presence of a large number of North American migrant species in the CCB. In addition, the presence of bodies of water contributes to the high number of species found in the basin. Even though other studies in our comparison were made in sites with riparian vegetation or where water bodies were present, in the CCB the pool (pozas) and river systems are an important part of the landscape and have been there since they were isolated from the sea in the late Mesozoic (Souza et al. 2008). Historically, this may have favored the establishment of resident species and especially the presence of migratory birds that have established migratory routes throughout the region. In addition, there is a variety of discrete vegetation types in the CCB, including grassland and Larrea and mezquite scrub, and the percentage of species shared between sites was low (mean percentage = 22%). This suggests that each type of vegetation has its own composition and partially explains the elevated number of species found.

From a biogeographical perspective, shrublands had more species than grasslands and other desert habitats. This was also the case for the CCB. This could be due to the fact that the shrubs and small trees are structurally more complex and can harbor both ground and foliage species in contrast to grasslands.

Finally, we found a negative relationship between richness and precipitation across regions when the Tehuacán site was excluded. This could be an artifact because the narrow range of rainfall values, with the exception of Tehuacán, does not necessarily influence bird species richness. The fact that the number of species decreased at higher latitudes might be more relevant and the neotropical component could well play an important role in explaining the high number of species in the CCB and Tehuacán. In fact, the Tehuacán Valley shares only 30% of its species with the North American deserts, but when the tropical birds are excluded, the sites share 80% of the species (Arizmendi and Espinosa de los Monteros 1996; Dávila et al. 2002). In the CCB, the tropical component is not as high, but, according to reproductive ranges, there are at least 14 species of neotropical affinity (14% of the total).