Study area

The climate is semi-cold subhumid with rains in summer with average annual temperature between 5 °C and 12 °C, the temperature of the coldest month varies between -3 °C and 18 °C with a percentage of winter precipitation between 5 and 10.2 % (^{García, 1973}). The Edaphological Chart Series II indicates that the soil is classified as Regosol (^{INEGI, 2010}) and the arboreal vegetation is composed by species of the Pinus and Quercus genera (Soil Use and Vegetation Chart Series VI).

Figure 1 Location of the studied plantations at the ejido and state context. 

Sample selection

Sampling considered the selection of three areas in which mulch cuttings were applied with immediate planting in the summer of 2010 and the adjacent stands (RA) with tree vegetation established naturally and that represent the conditions that existed within the plantation areas before the application of clearcuttings. The species only planted at spacing of 3 × 3 m were Pinus cooperi Blanco and Pinus durangensis Martínez (50 % for both species), from nurseries in the study region, with one year of age, which are the species with the highest growth and economic value in the region. The plantation areas were named: plantation 1 (P1), with an area of 10.29 ha, plantation 2 (P2) with 12.45 ha and plantation 3 (P3) with 9.43 ha. Starting from the center of each plantation and in the direction of the cardinal and subcardinal points, 20 circular sites of 150 m² were located at a distance of 25 m between them (60 sites).

In order to evaluate the changes that occurred in the tree diversity of the plantations, a comparison was made of it with the vegetation of the RA, which were called RA1, RA2 and RA3. In function of the same directions of the plantations towards the RA, ten sites of 0.1 ha (30 sites) were located within each one.

At both the plantation and RA sites, all trees were identified by genus and species. In the case of the plantations, the diameter at the base was measured with a Truper™ Caldi-6MP caliper and for the RA the normal diameter at the height of 1.30 m with a Haglöf Mantax 59722 caliper, as the RA had regular diametric structures, with ages close to the forestry shift and, therefore, there is little presence of shoots.

Data analysis

From the information on the number of new species per site, the accumulated richness curves were obtained, to left that the sampling effort was enough and included most of the species present both in the plantations and in the RAs (^{Dzib et al., 2014}).

The description and comparison of the alpha diversity of the plantations and the RAs was carried out with the species richness indexes (S) and the Shannon-Wiener diversity index (H). The species richness index was defined as the total number of species present within each plantation and adjacent stand, while H 'was estimated from the sum of the multiplications of the proportions of the abundances of each species (pi) by the logarithm natural (ln⁡) of these proportions:

To evaluate the occurrence of significant changes in S and H´ between each plantation and its adjacent stand, considering the latter the one that represents the natural conditions of the forest, a rarefaction analysis was made from the number of species and abundances per sample, using the Estimates 9.1.0 program (^{Colwell and Elsensohn, 2014}). Through a Monte Carlo simulation process, this method estimates an average and variance of S or H´ for each sample size and, thereby, generates confidence intervals, which allow statistical comparisons to be made at a desired level of significance (0.05 % for this study) to the minimum sample size recorded (^{Gotelli and Colwell, 2011}). In this study, the minimum of samples was obtained in the RA (10 sites) compared to the plantations (20 sites).

Similitude indexes

To evaluate the level of species replacement, originated by the application of clearcutting with plantations against the RA, the qualitative similitude indexes of Jaccard (Ij) and Sörensen ((Is) were applied.

Where:

a = Number of species in the plantation

b = Number of species in the adjacent stand

c = Number of species common to both areas

The range of values of both indices goes from zero when there are no shared species, up to one, when the two sites share the same species. The Ij measures differences in the presence or absence of species, while the Is   relates the number of species in common in regard to all the species found in the two areas (^{Magurran, 1988}).

Importance Value Index

The species importance value index (IVI) was estimated to evaluate the existence of structural changes in the species composition as a consequence of the application of the slaughterhouse cut with plantation. The IVI, developed by ^{Curtis and McIntosh (1951)}, consists of adding the relative values of density (Abr), dominance (Dr) and frequency (Fr). This synthetic structural index is mainly oriented to rank dominance by species in mixed stands (^{Sánchez et al., 2018}). According to ^{Alvis (2009)}, the general expression of the IVI is the following:

Where:

Abr = Ratio of the number of trees per species (ni) over the total number of trees of all species (n).

Dr is equal to the basimetric area of each species ( ABi) over the total basimetric area of the species (AB).

Fr is the ratio of the number of sites in which each of the species is present (Fi) over the sum of the number of sites in which all the species are present (F).

Richness and diversity

The number of species in a community is an expression by which a quick and easy estimate of diversity is obtained (^{Jiménez, 2000}). Figure 2 shows the starting point of the upper asymptote of the species accumulation curves, deducing that the sample size was sufficient to record the richness of species present in the plantations and in the adjacent stands (^{Magurran, 2004}).

*RA1 curve cannot be appreciated as it is identical to RA2.

Figure 2 Species accumulation curves from the sampling of the three plantations and adjacent stands. 

Within the study area (plantations and adjacent stands), a total of eight species were recorded, distributed in the Pinus, Quercus, Juniperus and Arbutus genera. Between P3 and RA3 the greatest difference of species (two species) was detected, while P2 and RA2 differed only by one species, and no differences were found between P1 and RA1 (Table 1). In general terms, the total species richness recorded in this research is greater than the seven recorded by ^{Hernández et al. (2013)} for the forests of Chihuahua. In a similar way, ^{Návar and González (2009)} calculated 7 to 7.8 species average in areas without forestry treatment (0 % removal) in a study carried out in temperate forests of Durango. Likewise, they determined a gradual decrease in the number of species as the percentage of removal from the basal area increased, which was calculated up to 5.7 species on average with the clearcutting treatment (100 % removal).

The rarefaction curves indicate that the species richness of P3 is significantly lower than that of RA3. This agrees with the studies by ^{Corral et al. (2005)} and ^{Vásquez-Cortez et al. (2018)} coincided in the argument that the number of species that are shared can decrease in an area when carrying out forestry treatments. The curves of P1 and P2 with their adjacent stands indicate that the richness of species tends to maintain significantly the same number of taxa compared to the adjacent stands (Figure 3). In this sense, ^{Leyva-López et al. (2010)} and Hernández et al. (2013) pointed out that the application of intensive felling to promote regeneration (parent trees) does not only guarantee the establishment of the renewal of trees of the genus selected as parent tree, because within the intervened areas there are other species with different regeneration mechanisms. In this study, not only planted species of Pinus were established in both P1 and P2, but also the emergence of other pine species was confirmed.

Figure 3 Rarefaction graphics constructed for species richness (95 % confidence). 

In Figure 4 it is observed that there is no overlap in the confidence intervals in the Shannon proportional diversity index between the P2 and P3 plantations with their respective adjacent stands, which shows that there are significant differences between the forest diversity of these plantations and its adjacent stands. While between P1 and RA1 no significant changes were observed, indicating that both species richness and their proportion were similar.

Figure 4 Rarefaction graphics constructed for the Shannon-Wiener proportional diversity (95 % confidence). 

For P2 and its RA, the rarefaction curves based on species richness did not confirm significant differences (Figure 3). However, if they existed for H´. This characteristic coincides with that mentioned by ^{Del Río et al. (2003)} and Hernández et al. (2013), who recognized that the periodic application of forestry practices, although they can maintain the richness of tree species, tend to change the proportional abundance of the number of individuals of each of the species and, with it, the diversity.

Similtude indexes

When evaluating the similarity of the species present in the plantations with the adjacent stands, the Jaccard and Sörensen indexes supported the fact that the species composition between P3 and RA3 has greater similitude (five species in common) compared to the other plantations and its adjacent stands, which were less similar (Table 2).

Of the total number of species rerecorded in the study area, P. durangensis, P. cooperi, Quercus sideroxyla Humb.& Bonpl. and Juniperus deppeana Steud. are present both in the plantations and in the RAs (Table 1). In regard to the species registered in the areas treated with cuttings, most of the individuals of P. durangensis and P. cooperi come, mainly, from the plantation, while Q. sideroxyla, Q. rugosa Née, J. deppeana and Artus xalapensis Kunth sprouted by seeds contained in the soil or by vegetative reproduction. In this sense, the similarity of species as well as the difference in the number of them can be explained from the establishment and development strategies that each species shows in response to the kind and degree of disturbance to which it was exposed (^{Leyva-López et al., 2010}).

On the other hand, Hernández et al. (2013) determined that as the forest mass prospers after being intervened, a smaller number of species are shared, but their richness in the forest is maintained. This coincides with this study, although the similarity between P2 and RA2 was 0.57 for Ij and 0.72 for Is, the difference between these areas was only one species, and for P2 and RA2 the similarity was Ij = 0.66 and Is= 0.80; however, the same number of species was observed, while between P3 and RA3 there was a greater difference in richness (2 species) but the similarity was greater than for the other areas.

Importance Value Index

If RAs are considered as the original condition and that it was modified by the plantations, it is observed that P. cooperi remained as the species with the highest IVI in P2 and RA2; in the same way, P. durangensis did not undergo changes in the IVI between P3 and RA3, but in P1 it displaced P. cooperi from RA1 with the highest IVI. In contrast, the absence of P. engelmannii Carr. and P. leiophylla Schiede ex Schltdl. & Cham. stands out in the plantations with respect to their adjacent stands where they were present (Table 3).

J. deppeana did not undergo considerable modifications in the IVI in any of the plantations. Q. sideroxyla increased its IVI position in the three plantations and A. xalapensis emerged in P1 and P2, but was not identified in their RA (Table 3).

In general, it would be expected that only the species of P. durangensis and P. cooperi would be observed in the plantations in the same proportion as they were plants (50 % for each species). However, it is important to mention that both these species and other native ones can emerge spontaneously by seed deposited in the soil before and during cutting, seed from adjacent stands or the vegetative way, from the new conditions that arise in each area. Based on the above, ^{Oliver and Larson (1996)} and ^{Leyva-López et al. (2010)} mentioned that the presence of a greater or lesser number of species within the studied areas is due to the capacity of each of them to reproduce or regenerate in a vegetative, as well as its vulnerability to the level of disturbance, stress and competition. Furthermore, ^{Vásquez-Cortez et al. (2018)} recognized that the areas surrounding the clearcuttings can limit the dominance of the plants and the loss of tree and shrub diversity.

In regard to Juniperus, ^{Rzedowski (1981)}; ^{Bakker et al. (1996)} and ^{Biondi and Ferrante (2004)} mentioned that this genus has the ability to colonize degraded places and is part of secondary vegetation; these trees are early onset and their presence is associated with disturbances caused by human activities. This suggests a possibility of greater repopulation and development of the genus when the level of competition and shade are significantly reduced by silvicultural interventions. In a similar way, ^{Leverkus et al. (2014)} stated that Quercus species re-sprout very favorably in burned and degraded areas, which gives them resilience to the forests they comprise.

On the other hand, although there is little information on the regrowth capacity of Arbutus after cutting, a high incidence of new specimens was observed in the study area. In this regard, ^{Díaz-Hernández et al. (2014)} confirmed a high probability of Arbutus regrowth in an area with disturbance caused by a forest fire.