Location of the study area

The study was conducted in the Aboreachi forest ejido, located in the municipality of Guachochi, southwest of the state of Chihuahua, Mexico, in the physiographic province of the Western Sierra Madre, whose coordinates are 27°11'46"N 107°23'13"W (Figure 1). The area corresponds to a non-even-aged cold temperate forest with diverse forest species; the soil types are Eutric Cambisol, EutricPlanosol and Eutric Regosol of medium to fine texture (^{INEGI, 2014}). The climate is C(E)(w₂)(x'), which corresponds to a semi-cold humid climate, with a mean annual temperature of 5 to 12 °C and a mean annual precipitation of 621.3 mm (^{INEGI, 2008}).

Figure 1 Location of the study area in the Aboreachi ejido, municipality of Guachochi, Chihuahua. 

Sampling method

The study area comprised an area of 11.44 ha, last intervened in 1992; in 2017 it was established as a Permanent Forest Research Plot (PFRP). The dasometric information was obtained by census or total registration in the summer of 2018. The database consisted of 5 092 trees, whose normal diameter (DN) was measured at a height of 1.3 m from ground level with a Forestry Suppliers Inc^® five-meter diameter tape, and the total height (H), with a Suunto^® PM5-1520 hypsometer. The species to which each individual belonged was also recorded.

Data analysis

In order to determine the alpha diversity, the Shannon-Wiener index was calculated using the expression described by ^{Shannon (1948)} (Equation 1):

Where:

H’ = Shannon-Wiener Index

S = Number of species

p _i = Proportion of individuals of species i with respect to the total number of individuals

N = Total number of individuals

n _i = Number of individuals of species i

Equation 2 was used to measure the species richness with the Margalef index (^{Margalef, 1972}):

Where:

D _Mg = Margalef's Index

S = Number of species

nl = Natural logarithm

N = Total number of individuals

The horizontal structure of the stand was determined by characterizing the distribution of the diameter classes present, and the number of individuals for each taxon was quantified. Abundance was calculated based on the number of trees per species, and dominance, based on the basal area. Since a census was carried out, it was not necessary to evaluate the frequency of the species. The Importance Value Index was estimated according to the above indicators (IVI) (Equation 3), with the average percentage from 0 to 100 of the previous ecological indicators (^{Alanís-Rodríguez et al., 2011}):

Where:

IVI = Importance Value Index

RA _i = Relative abundance

RD _i = Relative density

The characterization of the vertical structure of the stand was carried out with the Vertical Species Distribution Index (A) (^{Pretzsch, 2009}) by means of Equations 4, 5 and 6. According to ^{Jiménez et al. (2001)}, this index defines three height zones: Zone I covers 80 % to 100 % of the maximum tree height; Zone II, 50 to 80 %, and Zone III, 0 to 50 %. This index is used to determine the structural diversity in terms of vertical distribution of species (^{Alanís-Rodríguez et al., 2020}); ^{García et al. (2020)} point out that A _max corresponds to the maximum value for each of the species per stratum; finally, A _rel is the standardization of A:

Where:

A = Index of vertical distribution of species

A _max = Maximum value per species in each stratum

A _rel = Standardized value of A

S= Number of species present

Z= Number of height strata

P _ij = Percentage of species in each zone

nl = Natural logarithm

N = Total number of individuals

Table 1 shows the equations used to estimate the total tree volume (volume of stump, trunk and branches) of each individual of the recorded species; the expressions are based on the Shumacher-Hall model, collected from the biometric study of the Aboreachi ejido and from the study by ^{Graciano-Ávila et al. (2019)}. Subsequently, the results per individual were used to estimate the total tree volume per hectare and at the PFRP level.

According to ^{Aguirre-Calderón and Jiménez-Pérez (2011)}, the estimation of biomass and carbon content can be derived from equations based on the normal diameter at the individual tree level. Therefore, the allometric models shown in Table 2 were used to estimate the aboveground biomass per individual of the species recorded. The biomass of each individual was then scaled to the level of taxon, genus and family, which allowed calculations to be made in Megagrams per hectare, and, therefore, at the level of the entire PFRP. The carbon content was obtained by multiplying the biomass value by the conversion factor 0.50 suggested by the ^{IPCC (1996)}; the unit of measurement utilized to express the results for the biomass and carbon stored in the tree layer per unit area was the Megagram per hectare (Mg ha^-1).

Floristic diversity

Sixteen tree species belonging to six genera of the families Pinaceae, Fagaceae, Cupressaceae, Ericacea, and Betulaceae were registered in the PFRP; this composition of families was higher than those cited by ^{García et al. (2020)} and ^{Hernández-Salas et al. (2013)} in coniferous forests of the state of Chihuahua. The most representative family was Pinaceae, with seven species; this result is similar to the one documented by ^{Márquez et al. (1999)}, who registered eight Pinaceae of the genus Pinus in a forest in Durango, followed by Fagaceae, Cupressaceae, Ericacea and Betulaceae, with five taxa, one, two and one, respectively. ^{González-Elizondo et al. (2012)} and ^{Silva-Flores et al. (2014)} indicate that these families are representative of the temperate forests of the Western Sierra Madre.

Species diversity and richness

The Shannon-Wiener Index (H') normally ranges between 1 and 5. Values above 3.5 are considered as high diversity; 2 to 3.5, as medium diversity, and less than 2, as low diversity (^{Margalef, 1972}). A value of 1.52 was estimated for the PFRP, which is higher than the value estimated by ^{Solís et al. (2006)}, of 1.21. For their part, ^{Delgado et al. (2016)}, reported a maximum value of 1.58; in an analysis of altitudinal diversity in pine-oak forests of Durango, ^{Medrano et al. (2017)} estimated a minimum value of 1.94, which is higher than the value determined in the study documented herein. As a consequence of the physiographic and climatic conditions, the PFRP has a low species diversity, typical of temperate forest ecosystems that have been harvested in the medium term, with typical second-growth or regenerating stands.

Margalef’s richness index (D _Mg ) corresponds to a value of zero when there is only one taxon in the stratum, but it can approach 5 if the number of existing species increases the complexity of the ecosystem. The study area exhibited a value of 1.76, which is lower than the value obtained by ^{Domínguez et al. (2018)} of 1.79 for mixed coniferous forests in the region of Pueblo Nuevo, Durango; for their part, ^{García et al. (2020)} estimated a value of 1.52 in mixed forests in the state of Chihuahua, lower than that reached in the PFRP; however, the values correspond to a wealth of species similar to that described in the present study. This result is consistent with what has been cited in other research for temperate ecosystems, as the climatic and physiographic conditions of coniferous forests and pine-oak associations limit the development of a wide variety of tree taxa; the opposite is the case in tropical regions, where it is possible to observe more complex tree associations.

Horizontal structure

The horizontal structure of the tree stands in the PFRP showed an irregular distribution that is typical of a second-growth forest or stands in the process of regeneration, due to previous timber harvests, anthropogenic impacts such as land use change, fire, or environmental impact in the form of pests and diseases. The highest frequency was registered in trees with a diameter category of 10 cm, decreasing in proportion to larger diameters (Figure 2). In the PFRP, an average normal diameter of 16.72 cm was obtained for the total number of species recorded. The Pinus genus presented an average diameter of 19.21 cm and Quercus 13.88 cm; values similar to those reported by ^{Návar-Cháidez and González-Elizondo, (2009)}, as well as by ^{Delgado et al. (2016)}, for pine-oak forests in Durango, and by ^{García et al. (2019)} in pine-oak forests of the southern part of the state of Chihuahua.

Frecuencia (miles) = Frequency (thousands); Categoría diamétrica = Diameter class.

Figure 2 Frequency of tree species by diameter class determined in the PFRP. 

The commercial tree community that makes up the PFRP consists of 5 092 registered individuals, equivalent to 445 trees per hectare (trees ha^-1) (Table 3). Pinus had the highest density, with a total of 3 769 trees (329 trees ha^-1) representing 74.02 % of the total plot; the second place is occupied by the Quercus genus, with 1 062 specimens (93 trees ha^-1), amounting to 20.86 % of the stock. The Juniperus genus was present with 132 individuals (12 arb ha^-1), equivalent to 2.60 % of the total. Finally, the "other leaves" group accounted for 129 trees (11 trees ha^-1), which represented 2.52 % of the total number of individuals in the area. The Pinus and Quercus genera add up to 94.87 % of the tree cover; their values coincide with those described by ^{Delgado et al. (2016)}, ^{Domínguez et al. (2018)}, and ^{García et al. (2019)} in forests of the states of Durango and Chihuahua.

The species with the highest abundance was Pinus durangensis Ehren., with 2 539 individuals (222 trees ha^-1), equivalent to 49.86 % of the total tree cover; P. ayacahuite Ehren. registered 885 trees (77 trees ha^-1), with 17.38 % of the total, and Quercus sideroxyla Bonpl., 845 individuals (74 trees ha^-1), which amount to 16.59 % of the total. These three species together constitute 83.83 % of the existing trees within the PFRP (Table 4). ^{Zúñiga et al. (2018)} obtained similar results in the region of Pueblo Nuevo, Durango; according to these authors, P. durangensis tends to be the most representative taxon in pine-oak forests.

The largest basal area corresponded to P. durangensis, with 134.62 m², equivalent to 11.77 m² ha^-1; Q. sideroxyla, with 29.41 m² (2.57 m² ha^-1); P. ayacahuite, with 24.92 m² (2.18 m² ha^-1), and P. chihuahuana Engelm., with 11.93 m² (1.04 m² ha^-1). Together, these three taxa totaled 200.88 m² (17.56 m² ha^-1), i.e., 92.52 % of the total. ^{Valenzuela and Granados (2009)} point out that P. durangensis is found in two tree associations in the state of Durango; ^{Graciano-Ávila et al. (2020)} report that P. durangensis and Q. sideroxyla are the most dominant species in a forest of the same state.

At the PFRP, Pinus exhibited an IVI of 76.70 %, while the genus with the lowest value was Alnus, with 0.07 % of total. According to ^{López-Hernández et al. (2017)}, in pine-oak forests of the state of Puebla, Pinus spp. tends to be the most representative, with up to 85.20 % of the total. P. durangensis, Q. sideroxyla, P. ayacahuite, and P. chihuahuana were the species with the highest IVI: 55.93, 15.07, 14.43, and 5.90 %, respectively; these four taxa together accounted for 91.33 % of the total. Eight species were registered with values of less than one percent; in this regard, Q. candicans Née and Q. fulva Liebm. were the least outstanding. According to ^{Graciano-Ávila et al. (2017)}, in a forest of Durango, P. durangensis Q. sideroxyla, and P. teocote Schiede ex Schltdl. were the most prevalent species, with 62.43 % of the community's total.

Vertical structure

The PFRP exhibited a value of 1.8916 in the vertical distribution index of species, with an A _max of 3.8712 and an A _rel of 48.86 %. ^{García et al. (2020)} determined an A index of 2.58, an A _max of 3.40, and an A _rel of 75.56 % in the mature tree stratum of a mixed forest in Chihuahua; on their part, ^{Graciano-Ávila et al. (2020)} calculated a maximum value for A of 2.72, with an A _max of 3.87 and an A _rel of 70 % in managed forests of Durango.

The distribution of individuals by stratum obtained from the analysis may be a consequence of the effect of the previous application of silvicultural treatments to this irregular stand, where the aim was to take advantage of the over mature individuals that were mainly located in the high stratum (I), so that the opening of the canopy favored natural regeneration, increasing the number of individuals and species in the low stratum (III). In the distribution of strata (Table 5), P. durangensis was observed to be the only species present in stratum I, with a maximum height of 29.30 m; stratum II was represented by 10 tree species, among which, again, P. durangensis registered the highest proportion within the stratum, with 80.77 %, followed by P. ayacahuite with 10.06 %. The height range for this stratum was 14.70 to 23.43 m. Finally, stratum III presented 100 % of the registered taxa, among which P. durangensis was the most abundant, representing 44.99 % of the total number of individuals in the stratum and 38.92 % of the individuals in the entire area, followed by P. ayacahuite and Q. sideroxyla, which amount to 16.04 and 15.99 %, respectively, of the total tree inventory of the PFRP.

The vertical distribution analysis showed that species diversity decreases as the tree height increases; similar results were described by García et al. in mixed forests in 2019, and in a forest of Pseudotsuga in 2020, in both cases in the south of the state of Chihuahua.

Estimation of total tree volume, aboveground biomass and carbon stocks

A total tree volume of 1 810.38 m³ was quantified in the PFRP (Table 6). The genus Pinus contributed the largest volume, with 1 545.69 m³ equivalent to 85.37 % of the total; for the Quercus genus, a volume of 233.63 m³ was estimated, which corresponded to 12.90 % of the total. At the species level, a volume of 1 222.62 m³ (67.53 %) was calculated for P. durangensis, followed by P. ayacahuite, with 216.11 (11.93 %) and Q. sideroxyla, with 208.44 m³ (11.51 %). In terms of total volume at the hectare level, the PFRP exhibited an average volume of 158.25 m³ ha^-1, which is lower than that obtained by ^{Graciano-Ávila et al. (2019)} in temperate forests of Durango, of 207.36 m³ ha^-1; the volumetric difference is due to the fact that the study carried out in the El Salto area shows a greater density with a higher average diameter and height than those registered in this study.

Regarding the aerial tree biomass determined in the PFRP, a total of 1 066.44 Mg was calculated, of which P. durangensis registered 687.45 Mg (64.46 %), P. ayacahuite 150.74 Mg (14.13 %), and Q. sideroxyla 131.46 Mg (12.33 %) of the total; Q. candicans had the lowest biomass value, with 0.01 Mg. The estimated biomass at the hectare level was 93.22 Mg, which is within the biomass range of 75.43 to 176.06 Mg determined by ^{Martínez et al. (2016)} ha^-1 for pine-oak forests in the state of Durango.

The volume, biomass and above-ground carbon obtained in the species recorded in the research plot (30 to 45 cm) exhibited values of 60.62, 67.46, 63.42 and 64.99 Mg, respectively; these concentrated 47.99 % of the total carbon in the PFRP (Figure 3). The carbon stored by the genus Pinus was the highest (452.55 Mg), amounting to 84.68 % of the total PFRP; Quercus accumulated 73.71 Mg, which represents 13.79 % of the total stored.

Frecuencia (miles)= Frequency (thousands); Categoría diamétrica = Diameter class; Carbono = Carbon.

Figure 3 Aerial carbon stored by diameter class in the research plot. 

Of the existing species in the study plot (11.44 ha), P. durangensis had the highest accumulated carbon values (343.72 Mg), followed by P. ayacahuite (75.37 Mg) and Q. sideroxyla (65.73 Mg). These three species accounted for 90.72 % of the total carbon accumulated in that plot. The mean value of this parameter was 46.71 Mg ha^-1; the results are in agreement with those reported by ^{Graciano-Ávila et al. (2019)} in pine-oak forests of the state of Durango (65.14 Mg ha^-1), but are lower than those published by ^{Buendía-Rodríguez et al. (2019)}, in the range of 58.36 to 123.49 Mg ha^-1, for pine-oak associations in southern part of the state of Nuevo León.

Studies characterizing forest stands in terms of diversity, structure, volume, biomass and carbon content of the tree stratum, based on census information regarding large areas such as the one documented herein, will help managers to plan informed and oriented actions in order to achieve sustainable management of timber resources in northwestern Mexico. In addition, periodic monitoring of large permanent plots will reveal the dynamics of growth over time and, thus, contribute to the development of research programs for the use and conservation of forest resources.