Introduction

Forest ecosystems provide the habitat of wild plants and animals with the necessary resources to maintain their populations in space and over time (^{McComb, 2016}). The temperate forests of the Trans-Mexican Volcanic Belt (FVT, for its acronym in Spanish) are located in a transition zone where the Nearctic and Neotropical regions overlap, which is why they are distinguished by their high biological richness and endemicity (^{Luna et al., 2007}; ^{Suárez-Mota and Téllez-Valdés, 2014}; ^{Johnson et al., 2015}). The variety of climates and physiographic conditions present in the FVT favor a rich flora represented by at least 5 139 species of vascular plants, which is the largest number of taxa in these ecosystems of the main mountainous regions of Mexico (^{Villaseñor and Ortíz, 2007}). The diversity of wildlife is also high, in response to the variety of environments and floristic richness (^{Escalante et al., 2007}; ^{Flores-Villela and Canseco-Márquez, 2007}; ^{Navarro-Sigüenza et al., 2007}). The importance of terrestrial vertebrates (amphibians, reptiles, birds and mammals) in these ecosystems consists of the role they have as pollinators, seed dispersers, prey or predators and consumers of invertebrates and harmful insects and carrion (^{McComb, 2016}).

The Trans-Mexican Volcanic Belt (FVT) is one of the main regions of Mexico due to its social and economic importance; its biodiversity is subject to threats from anthropogenic pressures such as land use change, deforestation and human population growth.

Pines (Pinus L.) and oaks (Quercus L.) are among the groups of plants in the FVT with the greatest richness, and of high economic importance, since they are overexploited by illegal logging or sustainably harvested (^{Suárez-Mota and Téllez-Valdés, 2014}). For their harvesting, several technical systems are practiced with different degrees of intensity: the Forestry Development Method (MDS), which includes intensive practices that promote regeneration (total felling, liberation felling and thinning), it predominates and is applied in most of the area of the region where this study was carried out.

The above causes the resulting trees to be homogeneous, with uniform ages and few species. Furthermore, in the highest areas, from 3 000 masl, the Mexican Method for the Management of Irregular Forests (MMOBI) is applied, which is selective, for this reason, greater specific and structural diversity of tree vegetation is maintained, but the surface area with this type of management is small in the study area. Finally, to a lesser extent, the Successive Cutting (CS) method is used, which is similar to the MDS, which promotes the protection of seed trees and natural regeneration, which results in homogeneous masses (^{Morales, 2015}; ^{Ramírez, 2017}).

The aforementioned management systems have an impact on the abundance and richness of wild vertebrates, since they transform the spatial structure and floristic composition of the vegetation, as well as the local humidity and temperature (^{Palik et al., 2021}).

The Mexican legislation on the matter establishes as one of its priorities to integrate best practices for the conservation of forests and tropical forests biodiversity in their management programs (^{Semarnat, 2018}, ²⁰²⁰). To achieve this, it is necessary to know its composition and state. As a first phase, knowledge generated from a monitoring system is required with which those in charge of managing forest properties obtain basic information to identify areas at risk of loss of biological diversity, and to apply procedures that mitigate the impacts of the different forestry systems (^{Conafor, 2015}, ²⁰¹⁷). The second phase consists of analyzing the effect of different forestry systems on richness, diversity and occupation.

The region where this study was carried out is located in the Sierra Norte de Puebla and it is known as Cuenca de Abasto Chignahuapan-Zacatlán due to its forestry importance, since it is home to more than half of the forested area under management of the state of Puebla, for the annual volumes of harvested wood and for having almost 50 % of the installed forestry industry of the entity. Therefore, the application of intensive management techniques is promoted with the incorporation of best practices that allow the conservation of biodiversity in the intervened areas (^{Morales, 2015}; ^{Conafor, 2017}).

The importance of the conservation of biological diversity in areas subject to forest management has been promoted for a short time (^{Conafor, 2017}), including the Sierra Norte de Puebla (^{Barrón, 2021}); which has promoted the completion of academic studies, observations and faunal inventories to complement forest management programs. However, few have been recorded in the scientific literature since most of them remain in technical reports.

In this regard, formal and published studies in this area are scarce and partial; among them, the work of ^{Chávez-León (2019)} can mentioned, by phototrapping, he recorded the presence of nine species of medium and large mammals, in addition to seven birds, with a similar specific richness in stands where the MDS and MMOBI are applied. ^{López-Becerra and Barrón-Sevilla (2018)} documented 35 taxa of birds in managed forests of the Acolihuia ejido; they indicated greater richness in the silvicultural practices of the MDS than in those of the MMOBI. The herpetofauna of the region has received little attention, with specific reports of sightings of some species, such as that of ^{González-Hernández et al. (2016)} in which a snake: Thamnophis pulchrilatus (Cope, 1865), three lizards: Abronia graminea (Cope, 1864), Sceloporus microlepidotus (Herrera, 1890) (synonym of S. grammicus Wiegmann, 1828), and Lepidophyma sylvaticum Tailor, 1939, and two frogs: Dryophytes eximius (Baird, 1854) and Rheohyla miotympanum (Cope, 1863) were mentioned.

The objective of this work was to generate a general inventory of amphibians, reptiles, birds, and medium and large mammals, as well as estimate their richness in coniferous forests subject to three different forestry management methods in the region of Chignahuapan, Puebla, Mexico. Due to space limitations, the quantitative and comparative analysis of the effects of forest management on fauna diversity indicators will be documented in another article.

Materials and Methods

The study area is located in the region of the Trans-Mexican Volcanic Belt known as Sierra Norte de Puebla, mainly in the Chignahuapan municipality (Rinconada, Llano Verde, Villa Cuauhtémoc and Chignahuapan ejidos or common lands; private property Fraction III of the Former Ex Hacienda de Atlamaxac) and partially in Aquixtla municipalities, Puebla (El Manantial Multifunctional Forest Reserve) and Tlaxco, state of Tlaxcala (private property Fraction VI of the Former Ex Hacienda de Atlamaxac), in an altitudinal range of 2 400 to 3 550 m (Figure 1).

Figure 1 Location of wildlife sampling stations (yellow dots) on properties under forest management in Chignahuapan and Aquixtla, Puebla and Tlaxco, Tlaxcala, Mexico. 

The predominant ecosystem is coniferous forest with vegetation types of pine, pine-oak, oak-pine, fir-pine and fir in different phases of development of secondary succession (^{Inegi, 2017}). Up to three tree strata are distinguished, one bushy with different degrees of density and one herbaceous. The climate is temperate subhumid, and in the highest areas, it is semicold subhumid. Pinus patula Schltdl. & Cham. dominates these forests as a consequence of a long process of intervention and reforestation with this pine of high economic importance (^{Morales, 2015}).

Sampling was carried out in the years 2019 (October 14 to December 6), 2020 (March 30 to June 5), 2021 (September 20 to December 7) and 2022 (April 25 to July 8) in 32 stations distributed, mostly, in stands where the MDS thinning felling forestry treatment was applied (17), since this method is the most widespread in the region. To a lesser extent, it was sampled in stands subject to selection felling by the MMOBI (4) and in liberation felling by the CS (1) which are applied on few areas; in addition, stands intended for protection were included (10), which do not receive silvicultural management. Data collection was repeated for two years in most of the sites (15) and in some up to three (2) and four (6), although once in 9.

The study focused on four groups of terrestrial vertebrates: amphibians, reptiles, birds, and medium and large mammals. Data collection of birds, herpetofauna and mammals tracks was carried out on three occasions (with 30-day intervals) in each annual sampling period.

The field methods consisted of circular sampling stations of 1 ha, with a minimum distance of 2 km from each other, located inside forested areas more than 50 m from the nearest edge with open environments such as clearings, clear cuts, areas agricultural or grassland. Two or three were established per property, depending on its size. Each one was instrumented with a Cuddeback^® camera trap, models E3 (black flash) or C1 (white flash) to autonomously capture images of medium and large mammals, although birds were also recorded. They were kept operating continuously for a minimum of 60 days to simultaneously take 20 MP definition photos and 30-second videos. For recording, events of three synchronous shots were programmed with delay intervals between events of 1 minute. The status of the equipment was reviewed every 30 days.

The management, organization and analysis of the images was done with the camtrapR package in R (^{Niedballa et al., 2016}; ^{R Core Team, 2020}). To complement and confirm the phototrapping records, an intensive search for tracks was carried out for 30 minutes throughout the 1 ha circular station (^{Aranda, 2012}). The species were determined based on the descriptions of ^{Ceballos and Oliva (2005)}.

The visual and auditory recording of birds was carried out using counting points lasting 15 minutes from the center of the station (^{Pierce et al., 2020}). The identification of the observed species was corroborated by the descriptions of ^{Howell and Webb (1995)}, and was complemented with sound recordings for 15 minutes, distributed in three continuous periods of three minutes active and two minutes of pause, with a DR-40X Tascam^® portable equipment, which were analyzed online with the BirdNET application (^{Kahl et al., 2021}).

For amphibians and reptiles, an intensive search was carried out for 30 minutes at the monitoring station (1 ha). The capture of the specimens was manual or with a herpetological stick (^{Lips et al., 2001}). The individuals were released at the same site after taking photographs for later identification using the taxonomic keys of ^{Lemos-Espinal and Dixon (2016)}.

Abundance was calculated according to the different sampling methods used for each taxonomic group. The number of records of individuals per species of birds, amphibians and reptiles was used to estimate the relative frequency: proportion of the number of records of each species per sampling station among the total (^{Pierce et al., 2020}). In the case of mammals, the number of individuals detected by the cameras was calculated, standardized by 1 000 trap-days (^{Mandujano and Pérez-Solano, 2019}).

To determine how complete a biological inventory is, recently developed advanced statistics are applied (^{Chao et al., 2020}) as an alternative to traditional species accumulation curves. These statistics are parameterized by an order q to control the sensitivity to the relative abundances of the species. When q=0, species abundances are not considered and are reduced to the conventional measure of richness; and the completeness of sampling (C) is the relationship between the observed species richness and the expected richness (observed but not detected), which is expressed as a proportion or percentage. The completeness of the sampling and the expected richness of each group were calculated with the SpadeR program (^{Chao et al., 2019}).

The structure of the four terrestrial vertebrate communities was analyzed in terms of the homogeneity of their abundances, for which range/abundance graphs or Whittaker plots were generated (^{Magurran, 2004}) in which the species are ordered from highest to lowest quantity or its proportion.

Taxonomic lists of the four sampled groups were prepared, from which the taxa with some risk category in the Mexican Official Standard NOM-059-SEMARNAT-2010 (^{Semarnat, 2019}) and with restricted distribution (endemic to Mexico) were identified, which should be considered priorities for conservation in managed forests.

Results and Discussion

A total number of 13 species of medium and large mammals, 83 of birds, four of amphibians and 12 of reptiles were recorded (Table 1). This means 45 % of medium and large mammals (^{Escalante et al., 2007}), 14 % birds, except for those associated with aquatic environments (^{Navarro-Sigüenza et al., 2007}), 4 % amphibians and 9 % reptiles (^{Flores-Villela and Canseco-Márquez, 2007}) of the estimated total existence of taxa of these four groups in the FVT. If it is considered that the territory of the FVT is around 156 thousand km² (^{Ferrusquía-Villafranca, 2007}) and that of the forested area where the present study was carried out covers around 600 km² (0.4 %), it is inferred that the only best group represented was that of mammals.

The completeness analysis (C) indicated that sampling was enough to record most of the existing species in the study area (Table 1), since the total per taxonomic group varied from a minimum of 91.8 % for amphibians to a maximum of 99.6 % for mammals, with slightly less for birds (98.9 %) and reptiles (98.6 %). Regarding the different forest management methods, in MDS the highest values were achieved in the four groups, followed by CS and MMOBI, although in the areas under protection or without management, the expected completeness and richness were higher than in these two and similar to what was recorded in MDS.

A greater number of species from the four groups of terrestrial vertebrates were recorded in the MDS than in the stands where the other two forest management methods are used (Table 1). The above is mainly due to the differences in the size of the surfaces: those of MMOBI and CS are lower than those corresponding to MDS. The same happened in the case of mammals and reptiles, and the numbers of amphibians were minimal. It is notable that no amphibians were recorded in the protection areas, but reptiles were. It is possibly due to the changes in humidity and temperature caused by the application of the different stages of the felling cycle and the gradual opening of the canopy (^{Iglesias-Carrasco et al., 2023}; ^{Sanczuk et al., 2023}), that modify the conditions of the habitat for these ectothermic organisms, which depend on external sources of heat to regulate their body temperature.

Whittaker curves illustrate the community structure of each taxonomic group: dominance, evenness, and species richness (Figure 2). That of birds is concave, with a very pronounced slope at the beginning (Figure 2A), which indicates that three species, Myadestes occidentalis Stejneger, 1882, Myioborus pictus (Swainson, 1829) and Cyanocitta stelleri (Gmelin, 1788), are the most abundant and dominate this community with a proportion of 9 to 12 %, followed by five (proportion of 4 and 6 %), and the rest (75 species) with very low abundance. The graph of medium and large mammals tends to be more convex (Figure 2B) with two taxa whose proportional abundance is approximately 30 % (Sylvilagus Gray, 1867 spp. and Sciurus aureogaster F. Cuvier, 1829), another four between 7 and 10 %, and the rest (six species) with around 1 %. That is, at least half have a similar abundance.

A = Birds; B = Medium and large mammals; C = Amphibians; D = Reptiles.

Figure 2 Whittaker curves. 

In regard to reptiles, a lizard (Sceloporus grammicus) was the dominant one with 23 % of the records, four accumulated between 10 and 15 % and seven with less than 5 %; the slope is less marked than the previous two as a consequence of greater equity in the abundance of the least abundant reptiles (Figure 2C). The amphibian curve is an example of a community poor in species and abundance, which is noted by the straight slope and the proportional differences of 10 % between each one (Figure 2D). In comparison, the bird community was the richest and least uniform, followed by the mammal and reptile community with fewer species but greater uniformity, and the amphibian community was the least rich.

Conclusions

In areas subject to intensive forest management methods, such as the MDS, a richer community of mammals and birds is observed than MMOBI and CS, but similar to those in areas under protection, without management. The richness and abundance of amphibians and reptiles, most in some risk category, are low in all management methods, especially in MMOBI and CS, and almost absent in protection zones. Although this may be due to differences in sample sizes, completeness analysis indicates that sampling was sufficient to detect most taxa. It is also a reflection of the dimension of the areas where the mentioned methods are applied.

The bird and mammal communities sampled are dominated by a few species, two or three, with high abundance, while the rest had few individuals. Amphibians and reptiles have fewer species, but most have similar abundances.

Finally, it is important that those responsible for forest management programs consider species that are in some risk category in NOM-059-SEMARNAT-2010 and those endemic to Mexico as priorities for their conservation. This indicates that they are vulnerable to changes in their habitat conditions, such as those caused in the structure and composition of tree vegetation by forestry interventions and other human activities.

Taxonomic inventories are essential for understanding biodiversity in productive forests. They should be considered as a tool to base plans for the application of best practices for their conservation.