Introduction

Conservation efforts require the delimitation of taxa for their prioritization (Hortal et al., 2015; Moritz, 1994); therefore, it is essential to recognize the evolutionary independence of units, typically recognized as species (Cracraft, 1983; Haig et al., 2006; Moritz, 1994; Ryder, 1986; Zink, 2004), along with their historical relationships (Ryder, 1986). However, when we face species delimitation, the species concept itself has led to decades of controversy concerning the definition of the species category and methods for inferring the boundaries and species numbers (de Queiroz, 2007; Fraser & Bernatchez, 2001; Winker et al., 2007). The lack of agreement on this controversy and the dominance of the biological species concept (BSC) led to the recognition of subspecies as conservation units. Despite different points of view on their usefulness and applicability (e.g., Alström, 2006; Johnson et al., 1999; Remsen, 2010; Wheeler, 1999; Winker, 2010; Zink, 2004), subspecies have been considered as conservation entities by several international conventions and organizations such as CITES (UNEP & WCMC, 2014) and IUCN (IUCN, 2020), as well as in official endangered species lists of mega-diverse countries such as Brazil (MMA, 2014), Australia (EPBCA, 1999), and Mexico (DOF, 2010).

Inconsistent taxonomic criteria for species assignment can impact conservation priorities (Peterson, 2006; Peterson & Navarro-Sigüenza, 1999; Rojas-Soto et al., 2010; Zink, 2014), which limits the possibility of accurately evaluating the conservation state of independent evolutionary units in order to protect them effectively (Nori et al., 2020; Scherz et al., 2019). This issue is particularly exacerbated in Latin America because there is a lack of enough specialists, which generates geographically unbalanced biodiversity inventories (Hughes et al., 2021; Mora et al., 2011), and a shortage of understanding of population distribution and variation within species (Peterson & Navarro-Sigüenza, 1999), increasing the Linnean, Darwinian, and Wallacean shortfalls (Hortal et al., 2015). However, recently re-analyzed groups based on formal phylogenetic methods reveal that the actual diversity patterns are usually underestimated (Cordier et al., 2021). Therefore, splitting/synonymizing the species groups (filling Linnean and Darwinian shortfalls), generates immediate changes in species distributions (filling Wallacean shortfalls; Diniz-Filho et al., 2013; Hortal et al., 2015). These changes also update the species’ vulnerability (Scherz et al., 2019) and accurately identify richness and endemism patterns (Peterson & Navarro-Sigüenza, 1999, 2000).

The assignment of risk categories also requires complementary information, such as population sizes, decline rates, threats, land-use change, and vulnerability to climate change (Foden et al., 2018; IUCN, 2019; Mace & Lande, 1990). However, restriction of the species range is one of the most important criteria. The International Union for Conservation of Nature (IUCN) developed the Red List of globally Threatened Species to establish their risk category based on population trends, occupation areas, distribution areas, and the species extinction risk analysis (Cassini, 2011; Harfoot et al., 2021; IUCN, 2020). In Mexico, the Norma Official Mexicana NOM-059-SEMARNAT-2010 (DOF, 2010; NOM hereafter) is the legal instrument that describes species risk categories at the national level. The NOM bases its categorization on the Method for Extinction Risk Assessment of Mexico’s Wild Species (MER). This method includes assessing the species’ distribution, vulnerability, and the influence of anthropic activities on their survival.

Both lists, IUCN and NOM, base the taxonomic recognition of the conservation units by applying the Biological Species Concept (BSC), defined as “the groups of actually or potentially interbreeding natural populations, which are reproductively isolated from other such groups” (Mayr, 1942, 1957, 1963). Thus, the characteristic of evolutionary units is their reproductive discontinuity from other units (Cracraft, 1983), and whose criterion for identifying species is precisely the ability to produce fertile offspring (Dobzhansky, 1973; Mayr, 1942); although some authors have recently accepted within this concept a relaxation in the strict necessity of reproductive isolation (i.e., Coyne & Orr, 2004; Johnson et al., 1999). Nevertheless, the central issue of differentiation is not reproductive isolation because it has no consistent genotypic or phenotypic correlations to predict allopatric groups’ reproductive compatibility. Besides, extensive inbreeding tests are impractical, and one needs to assume that phenetic similarity is directly related to ease of interbreeding (Sokal & Crovello, 1970). The BSC fails to provide unambiguous criteria for grouping organisms (or assigning species rank). Consequently, the resulting species are inappropriate for comparative biology or analyses of evolutionary history, as recognized taxons are frequently conglomerates of several independent, not necessarily monophyletic groups. Also, hybrid zones do not necessarily contribute to clarifying species limits. Alternatively, the Phylogenetic Species Concept (PSC) defines a species as “the smallest diagnosable cluster of individual organisms within which there is a parental pattern of ancestry and descent” (Cracraft, 1983). Various formulations of a PSC (e.g., Cracraft, 1983; de Queiroz & Donoghue, 1988; Donohgue, 1985; Mckitrick & Zink, 1988; Mishler & Brandon, 1987; Nixon & Wheeler, 1990; Zink & Mckitrick, 1995) agree that species concepts and definitions should emphasize criteria of phylogenetic relationship (descent) and not reproductive relationship (interbreeding; Donohgue, 1985; de Queiroz & Donoghue, 1988). Although different versions of the PSC emphasize diagnosability, monophyly, or both (see Mayden, 1997; Mishler & Theriot, 2000), all recognize that the species constitute differentiated basal taxa by the presence of unique characters (autapomorphies) or unique combinations of characters that allow the delimitation of evolutionary units or lineages (Mckitrick & Zink, 1988; Mishler and Theriot, 2000; Nixon & Wheeler, 1990; Zink & Mckitrick, 1995). Thus, the theoretical and practical limitations of the BSC lead us to advocate a PSC, which emphasizes monophyly and where diagnostic characters are considered to flag independent evolutionary histories and are used to delineate species boundaries. Such basal evolutionary units should be used in phylogeny reconstruction, speciation analysis, biogeography, and conservation (de Queiroz & Donoghue, 1988; Donohgue, 1985; Mckitrick & Zink, 1988; Moritz, 1994; Ryder, 1986). Despite many other species concepts, those cited previously are the most commonly accepted in biology and conservation.

By applying the PSC delimitation criteria instead of the BSC in the risk lists, different allopatric and diagnosable populations within some of the current biological species would be recognized as independent evolutionary units (Rojas-Soto et al., 2010). Consequently, several taxa considered as subspecies under the BSC would be considered as species under the PSC since biological species could be divided into two or more phylogenetic species (Figure 1a), which means recognizing independent distribution areas (i.e., allopatric populations recognized as separate species; Cordier et al., 2021; Zink, 2014). This division implies that, when applying the PSC, a reevaluation for the assignment of the risk categories is needed in some cases, given the presumption that they could be at greater risk due to the intrinsic reduction of the distribution areas and their population sizes (Cordier et al., 2021; Rojas-Soto et al., 2010; Williams et al., 2009). In this sense, one might expect that the consideration of subspecies for protection would not represent a conservation problem if subspecies always agreed with phylogenetic species or evolutionary units (e.g., Glaucidium griscomi corresponds with Glaucidium palmarum griscomi) but this becomes problematic when subspecies do not correspond with phylogenetic species (e.g., the phylogenetic species Glaucidium palmarum include two subspecies: G. palmarum palmarum, and G. p. oberholseri), which masks the actual conservation target (Figure 1a).

Figure 1.

(a) Geographic representation of the actual recognized distribution of Glaucidium palmarum considering the BSC (solid light orange) and the three subspecies that conformed the species: Glaucidium palmarum palmarum (discontinuous red line), G. p. oberholseri (discontinuous purple line) and G. p. griscomi (discontinuous blue line) in contrast to the separate distribution of Glaucidium palmarum (solid green) and Glaucidium griscomi (solid blue) when considering the PSC. (b) Potential distribution area of G. p. oberholseri considering two different sources. (c) Potential distribution area of G. p. palmarum considering three different sources.

Since changes in the official taxonomic lists (e.g., American Ornithological Society; AOS) are generally made after the publication of systematic studies, omissions inevitably occur in recognition of new lineages (i.e., species) resulting from the application of the PSC. Furthermore, it is common to observe bias in the taxonomic update toward more studied regions (e.g., North America) compared to others (e.g., the Neotropics); and toward certain groups (Hortal et al., 2015), which displaces the study of other taxa of less interest (Ortiz-Pulido, 2018). On the other hand, the official lists (recognized by the AOS, Handbook of the Birds of the World (HBW) & Birdlife International, 2019a) are based on the BSC but tend to omit the 75% rule criterion for subspecies recognition on their classifications, which means that some of them could, or not, qualify as valid taxa (Remsen, 2010) and also, subspecies distributions are not known well enough, there is no explicit geographic information, or there is a disagreement between sources (Friedmann et al., 1950; Moore, 1937; Nelson, 1901; Schulenberg, 2020; Figures 1b and c). As expected, these committees frequently dismiss the studies that demonstrate and support the phylogenetic species status (AOU, 1998 and its supplements, last: Chesser et al., 2021), which can generate vulnerability for populations of species that are not included in the risk lists. Therefore, the assignment of risk categories using the PSC constitutes an alternative to the BSC to protect the uniqueness of avian diversity throughout the re-analyses and new interpretation of the variation of characters that ponders an evolutionary recognition of the units to be protected (Haig et al., 2006; Isaac et al., 2004; Moritz, 1994; Rojas-Soto et al., 2010; Ryder, 1986; Zink, 2004).

The main goal of this study was to analyze the implications of the application of the PSC in the risk status’ assignment in NOM-059-SEMARNAT-2010 and the IUCN Red List of Threatened Species, based on the ranges of the phylogenetic species recognized from the new taxonomic reassessments, using the Mexican endemic avifauna as an example. We compared the species recognized under the PSC and the traditional taxonomy based on the BSC as an input for their possible updating or inclusion in the current risk lists.

Materials and Methods

Results

From the 93 phylogenetic species evaluated, we found that 31 are currently under some conservation concern. Eleven are listed as species and 20 as subspecies according to the NOM ( Supplementary Appendix 1). There are 20 species found to belong to an at-risk category and that are also recognized as independent units based on IUCN: 18 as LC and two as NT. Of these 20 species, five have a decreasing population trend, nine are stable, and one is increasing, and trend is unknown for the other five. For the 73 remaining phylogenetic species, we considered the established population trend for their related species recognized under the BSC. That results in 34 phylogenetic species with a decreasing trend, 29 as stable, six increasing, and four with an unknown trend ( Supplementary Appendix 1).

The potential distribution models reveal that 68 species were highly restricted (73%), 16 species (17%) were restricted; eight are considered widely distributed, and only one resulted in the extensive distribution classification. 73.2% of the potential distribution models for the species meet high scores for criterion “A” of the MER. In terms of the IUCN classification and according to the “B1” criterion, for 62 species (66%), the distribution areas determined from the SDMs would remain in the LC category, nine species could be susceptible to having the NT category, 15 as vulnerable and seven as threatened in the EN category (see examples in Table 1).

Table 1.

Examples of Some Phylogenetic Species Whose Categories Could Be Modified by the Implementation of the PSC. Columns from Left to Right are Included: (1) The Phylogenetic Species, (2) The Population Trend According to IUCN, * The Category is Assigned to the Species from Which the Phylogenetic Species is Derived (PT); (3) Vulnerability Value According to PIF (Where Values Close to One are Low Vulnerability and Maximum Vulnerability Corresponds to 20); (4) The Potential Distribution Area of the Phylogenetic Species in Square Kilometers; (5) The Current Risk Category by IUCN Corresponding to the Species: LC= Least Concern, NT= Near Threatened, *= the Category Comes from the Taxon from Which it is Split (RC IUCN); 6) The Category When Considering the B1 Criterion: LC= Least Concern, NT= Near Threatened, VU= Vulnerable, EN= Endangered (PC IUCN). NOM Category: P= Endangered, A= Threatened, ** The Category is Assigned to the Subspecies, *** The Category is Assigned to the Species from Which the Phylogenetic Species is Derived (RC NOM); Value to Criterion A of the MER.

Of the 68 highly restricted species, 23 have a score between 15 and 18, according to the evaluation of the PIF conservation status, and most of them have decreasing population trends (Table 2) so they can be considered species with high vulnerability or risk of extinction ( Supplementary Appendix 1). Of these, only five have a risk category assigned by the NOM as subspecies equivalent to the phylogenetic species recognized by PSC.

Table 2.

List of Species With High Vulnerability or Risk of Extinction Columns from Left to Right are Included: (1) The Phylogenetic Species, (2) The Population Trend According to PIF, * The Category is Assigned to the Species From Which the Phylogenetic Species is Derived (PT) 1= Large Significant Increase, 2= Small Significant Increase, 3= Uncertain Population Change, 4= Possible Moderate Decrease or Moderate Significant Decrease, 5= Large Significant Decrease, * if from the Taxon from Which the Phylogenetic Species is Derived; (3) Vulnerability Value According to PIF (Where Values Close to One are Low Vulnerability and Maximum Vulnerability Corresponds to 20); (4) NOM Category: P= Endangered, A= Threatened, *The Category is Assigned to the Subspecies; (5) The Potential Distribution Area in Square Kilometers; (6) Percentage Relative to Distribution in México (%).

Finally, 31 species are susceptible to a reconsideration of the risk category assigned by the IUCN, of which 22 could be re-categorized based on the potential distribution models obtained, nine could change to NT due to the combination of factors, and 62 species would maintain their category under criterion “B1” when considering the PSC.

Discussion

Red lists have used the BSC as a unit to determine species risk categories; however, the use of traditional taxonomy has ramifications in cases of specific phylogenetic species (e.g., Cortés-Rodríguez et al., 2008; Domínguez-Domínguez & Vázquez-Domínguez, 2009; González et al., 2011; Puebla-Olivares et al., 2008; Zink et al., 1997), where delimitations are not accurate for conservation purposes since they fail to recognize species evolutionary history, masking the fragility of taxa, leading to biases in conservation efforts (Moritz, 1994; Peterson & Navarro-Sigüenza, 1999; Rojas-Soto et al., 2010). In the case of NOM and partially in IUCN, the application of the BSC recognizes two taxonomic categories (i.e., species and subspecies) that correspond to one evolutionary unit (i.e., lineages) under the PSC. For example, some of the phylogenetic species considered correspond to biological species, others to subspecies, and even some correspond to two subspecies, such as Cyrtonyx sallei, which includes the subspecies C. montezumae sallei and C. m. rowleyi (Navarro-Sigüenza & Peterson, 2004), of which the NOM protects only the first. In these cases, the NOM segregates some populations of C. sallei as it fails to recognize the entire lineage as a single evolutionary unit.

There are complexes of species such as Lampornis amethystinus (Cortés-Rodríguez et al., 2008), Pampa curvipennis (González et al., 2011), and Aulacorhynchus prasinus (Puebla-Olivares et al., 2008) that have restricted distributions and under the PSC are divided into independent species. In these cases, importance should first be given to the extent of distribution for the assignment of risk categories and then to analyze other aspects, such as habitat modification pressures generated by anthropic activities and land-use change (MER, Tambutti et al., 2001; IUCN, 2019). For this reason, it is essential to focus studies and efforts on including the consideration of specific phylogenetic species in the environmental policy agenda, increasing their visibility to policymakers and the general public. Furthermore, the application of the PSC as a taxonomic criterion also reduces the ambiguity of the patterns that lead to subspecies delimitation, for example, when the interpretation of the different characters is not distributed geographically in a concordant manner (Remsen, 2010).

We highlighted some cases to exemplify the implications of the results found following. In the IUCN list, there are five subspecies endemic to Mexico from the separation of Colinus virginianus (C. ridgwayi, C. coyolcos, C. pectoralis, C. graysoni, and C. godmani; Figure 2). If they were recognized as independent species (as suggested by the PSC), they would have an NT classification because they would be assigned the taxon category to which they are associated. However, C. coyolcos and C. pectoralis could be included in the vulnerable category as phylogenetic species with more restricted distribution. Regarding the NOM, only C. ridgwayi has an assigned category (Table 1). However, because the C. virginianus complex is susceptible to extractive use for human consumption which would further decrease their populations, the revision of the risk category for all phylogenetic species could be pertinent to ensure the maintenance and establishment of sustainable management programs for its populations.

Figure 2.

Potential distribution of five phylogenetic species separated from the biological species Colinus virginianus.

Other examples include the Glaucidium palmarum, Coturnicops noveboracensis and Campylorhynchus rufinucha complexes (Table 1). The IUCN considers the extensive distribution of these biological species as a continuous area without distinguishing between phylogenetic species with allopatric distribution. Furthermore, the area includes ecosystems that are unsuitable for the species. In the G. palmarum complex, the distribution considered areas of high mountain forests; however, this species shows a direct and strict association with arid and tropical ecosystems (Howell & Robbins, 1995; Stotz et al., 1996). Specifically, with the tropical dry forest and the cloud forest’s lowlands, the use of the distribution area proposed by the IUCN when using the BSC can lead to inefficiency in defining conservation strategies and actions. This complex splits in G. palmarum y G. griscomi under the PSC, and G. griscomi could be considered in the NT category due to the extension of its range, PIF score, and the cloud forest vulnerability to which it is associated (Ponce-Reyes et al., 2012; Sierra-Morales et al., 2021; Stotz et al., 1996).

There are cases, such as Coturnicops goldmani (separate phylogenetic species from C. noveboracensis), that turned out to be “very restricted” because the estimated potential distribution area for this species is 0.06% of the national territory. Moreover, it turned out to be the most restricted lineage of all those evaluated, indicating that C. goldmani could be considered CR (critically endangered). C. noveboracensis is considered an LC biological species due to its stable population trend and wide distribution; however, it is presumed that C. goldmani is a separate phylogenetic species extinct in the wild (Howell & Webb, 1995). Although there are elements to consider C. golmani as a different species due to its limited dispersal capacity, neither the NOM nor the IUCN has considered that designation, not even as a subspecies, which suggests omissions in the risk assessment when considering a minor taxon.

Another example is Campylorhynchus rufinucha, a resident species of the tropical dry forest; this complex has a continuous distribution that ranges from western Mexico to northwestern Costa Rica with an isolated population in the central Veracruz plains near the Gulf of Mexico (under the BSC). However, Vázquez-Miranda et al. (2009) confirmed the genetic differentiation between the three subspecies that comprise it, resulting in C. rufinucha and C. humilis as Mexican endemic new phylogenetic species and a third species: C. capistratus, with distribution in southern Mexico and Central America. The AOS has not considered the taxonomic implications of the evidence showed by Vázquez-Miranda et al. (2009), ten years after its publication; however, the IOC (Gill et al., 2020) and HBW (2019b) have already accepted the separation and currently consider them as independent species. The suggested changes could mean that C. humilis as a recognized species would merit a new category and be assigned special protection based on MER criterion “A,” as it is a species with a very restricted distribution (C. rufinucha has the category of threatened as a subspecies thus the category could be maintained). On the other hand, for IUCN criterion B1, C. rufinucha could be considered VU.

Likewise, Phaethornis mexicanus has not been considered for inclusion in either of the two lists of species at risk, despite being recognized as a separate species from P. longirostris (Arbeláez-Cortés & Navarro-Sigüenza, 2013). This species distribution area is equivalent to 1.82% of the national territory, suggesting it could be assigned to a risk category considering the precautionary principle, which states that where there are threats of irreversible damage, and lack of scientific certainty shall not be used as a reason for postponing conservation actions (United Nations, 1992). Therefore, based on its restricted distribution, population decline, and vulnerability to climate and land-use change (Prieto-Torres et al., 2021; Sierra-Morales et al., 2016), P. mexicanus could be categorized as NT and Pr (Subject to Special Protection) according to the IUCN and the NOM, respectively.

Because the species analyzed in this study are Mexican endemic, it is crucial to consider them when categorizing risk at the national level by applying the risk evaluation method. Of all the 392 bird species and subspecies, only 91 (23%) were evaluated by the MER (DOF, 2010), which means that the rest of the species in risk categories were assigned with a sense of urgency based on “expert” knowledge. This situation highlights the relevance of enhancing studies that provide inputs for the re-categorization of the species. Of the species considered in this study, only 11 were evaluated by the MER, four as subspecies consistent with phylogenetic species under the PSC (Arremon apertus, Chlorospingus wetmorei, Glaucidium hoskinstii, and Campylopterus excellens) and the remaining in the species category under the BSC (Amaurospiza relicta, Amazilia viridifrons, Crypturellus occidentalis, Grallaria binfordi, G. ochraceiventris, Glaucidium palmarum, and G. griscomi). The correspondence of the category after the splitting considering the range reduction and population size reduction could be discordant.

It is currently difficult to determine how many Mexican species are or should be included under a risk category (Ortiz-Pulido, 2018) due, among other causes, to the lack of taxonomic coincidence among sources (Rojas-Soto et al., 2010; Zink, 2004), which produces uncertainty in taxa’s recognition. In this sense, given that the BSC considers reproduction for the delimitation of species, it generates uncertainty in recognizing their risk status, which can be clarified by considering finer aspects of species limits, such as the ancestry-descent patterns that the PSC addresses (Isaac et al., 2004; Moritz, 1994; Ryder, 1986; Zink, 2004).

In the absence of information and lack of resources to conduct extensive studies on species status, risk categorization systems inevitably tend to arbitrariness, exemplifying the importance of using a species concept that favors the recognition of evolutionary units as a conservation target (Alström, 2006; Frankham et al., 2012; Mayden, 1997). Since the PSC recognizes the individuality of taxa based on the uniqueness of the ancestral descent pattern, it increases the stability granted by being defined as a base unit (Domínguez-Domínguez & Vázquez-Domínguez, 2009; Leslie, 2015; Wheeler, 1999; Zink, 2004), with more assertive impact on the definition of public policies on biodiversity. Besides, one major problem is the fact that several of the current officially recognized subspecies (under the BSC) are not diagnosable by any mean (morphologically, genetically, geographically, and environmentally), and thus, they do not are considered, neither correspond to any phylogenetic species (Rojas-Soto et al., 2010; Zink, 2004).

Author Contributions

The authors’ contribution was equal for the study’s concept and design; data collection, data analysis and interpretation, manuscript preparation, and critical revision, adding intellectual content.