Introduction

Mountain systems meet environmental conditions related to climatic and edaphic variations that occur at small temporal and spatial ranges, which ensures the conservation of biodiversity (^{Körner et al., 2005}; ^{Nogués et al., 2007}). Temperature, rainfall, solar radiation and humidity favor the development of some more than others, based on their adaptation to the characteristics of the environment (^{Gutiérrez and Canales, 2012}).

In general, environmental factors modify species richness along some gradient because the available resources vary; thus, in the more favorable areas there will be greater possibilities for the existence of a higher diversity of species (^{McCarthy et al., 2001}). Altitudinal gradients in particular are closely associated with environmental modifications, which implies, in turn, a change in species diversity.

The study of this gradient in relation to the presence of plant species is important to know the differentiation between habitats with respect to their composition; to determine the distribution areas of each of the species and their spatial arrangement; to calculate the dimensions of the spaces of the endemic species; to understand how the structure of overlap and assembly between species; projecting the fundamental and realized ecological niche of species and the movement of species caused by natural environmental changes and anthropogenic activities (^{Koleff et al., 2008}). In this sense, the evaluation of the diversity of species in different altitudinal ranges contributes to the understanding of the changes that biodiversity undergoes in this context.

Although the region includes contrasting topographic conditions typical of mountainous areas, there is little information on the diversity of species and their distribution patterns in relation to the altitudinal gradient. Therefore, it is considered that the study described below constitutes an important contribution to the knowledge of the structuring of the species and the determinants in their composition for incorporation into the management programs for the purposes of harvesting and conservation of timber species. Thus, the objective of the present work was to evaluate and characterize the changes in the diversity of the arboreal vegetation in relation to the altitudinal gradient of the region.

Materials and Methods

The study area

Field work was carried out in the El Salto region at the southwestern part of the state of Durango (Figure 1), which is located in the mountainous massif of the Sierra Madre Occidental. The forest region of El Salto covers around 507 127 ha. The heights above sea level are distributed between 1 400 and 3 000 m (^{Inegi, 2017}).

According to the climatic classification of Köeppen, adapted for Mexico by ^{García (1981)}, the predominant climate in the study area is C (W₂), which corresponds to a subhumid temperate. The average annual rainfall is 1 200 mm, with rainfall in the months of June, July, August and September. This area has been subject to the application of the Método de Desarrollo Silvícola (Silvicultural Development Method) for about 35 years, a treatment that tends to the formation of regular forests.

Figure 1 Study area. 

Diversity and Equity Indices

Within each altitude range, species richness (S), Shannon- Wiener-Wiener proportional diversity index (H’), Simpson index (λ) and Pielou (J’) equity index and the Simpson for Dominance (E_λ) were estimated.

1

2

3

4

Where:

pi	= Proportional abundance of the number of individuals of each species (ni) divided by the total number of individuals (N)
H’max	= Maximum diversity of Shannon-Wiener, which equals the natural logarithm (ln) of the number of species (S)

Comparison of diversity

To determine if there were significant differences in the diversity between altitude ranges, the statistical test of ^{Hutcheson (1970)} was used, based on equation 5.

5

Variance was estimated from equation 6.

6

Once the value of t was known, the degrees of freedom (equation 7) were calculated to perform the hypothesis test.

Where:

H’1, H’2	= Shannon-Wiener indices of altitude ranges subject to comparison
VarH’1, VarH’2	= Shannon-Wiener variances of altitude ranges subject to comparison
Ni	= Total number of individuals in the ith gradient

Results

Species richness and the altitudinal gradient

Overall, six families, eight genera and 30 species were recorded; the most abundant families were Pinaceae with 12 and Fagaceae with 1 (Table 1).

Based on the different gradient ranges, the greatest number of species was found in the 2 100-2 400 m (S = 25 species), followed by the one of 1 800 to 2 100 m, by the one of 2 400 to 2 700 m, (S = 24 species each), 2 700 to 3 000 m (S = 20 species) and 1 500 to 1 800 m (14 species). This indicates that the greatest richness was located in the intermediate altitude ranges (1 800-2 700 m) and, in contrast, in the lowest (1 500-1 800 m) the smallest amount of taxa are found.

Table 1 Tree species richness by altitude ranges within the El Salto region. 

The species that were recognized along the altitude gradient studied were Arbutus tessellata P. D. Sørensen, Q. crassifolia Humb. & Bonpl., Q. rugosa Née, Q. sideroxyla Humb. & Bonpl., Pinus strobiformis Engelm., P. cooperi C. E. Blanco, P. durangensis Martínez, P. lumholtzii B.L.Rob. & Fernald, P. devoniana Lindl. and P. teocote Schiede ex Schltdl. In contrast, species restricted to a single range were Q. viminea Trel. (1 800-2 100 m), Pseodotsuga menziesii (Mirb.) Franco (2 700-3 000 m), Abies durangensis Martínez and Populus tremuloides Michx. (2 400-2 700 m). The other species had an in-between distribution range.

Proportional Diversity

Table 2 shows the values of the Shannon-Wiener index of 1.94 to 2.67, which means that there is greater diversity in the altitude ranges of 1 500-1 800 m, 1 800-2 100 m, 2 100-2 400 m and 2 400-2 700 m, compared to 2 700 to 3 000 m. Simpson’s dominance indicators varied from 0.82 to 0.91, and refer to the ranges 2 700-3 000 m and 1 800-2 100 m the lowest and highest dominance, respectively. On the other hand, the results of the Pielou Equity Index (J ‘) are between 0.54 and 0.74, corresponding to the lowest and highest homogeneity in the ranges of 2 700-3 000 m and from 1 800 to 2 100 m.

Table 2 Diversity and equity in different altitudinal ranges of the region of El Salto, Durango. 

The Shannon-Wiener indices of each altitude range and significance levels (P-value) define the difference in diversity between them. Most of them presented statistically different indicators of diversity of Shannon-Wiener (P-value <0.05). The opposite was verified (P-value> 0.05) among the 1500-1800 vs. 2 400-2 700 and 1 800-2 100 vs. 2 100-2 400 (Table 3).

Table 3 Hutcheson t test for each pair of assessed altitude ranges. 

Discussion

There is a variety of hypotheses that seek to explain the patterns of specific wealth in relation to altitude. Among them, ^{Stevens (1992)}, ^{Alvizu (2004)} and ^{Botero (2011)} point out that species richness decreases as altitude increases, while ^{Colwell and Hurt (1994)}, ^{Rahbek (1995)}, ^{Colwell and Lees (2000) Brown (2001)} and ^{McCain and Grytnes (2010)} argue that there is a greater number of species in intermediate range ranges. However, according to ^{McCain (2005)}, specific trends of species richness occurring in each region depend on climatic and edaphic factors that are linked to the range of altitude such as precipitation and temperature as well as soil type and Its moisture content. ^{Lamprecht (1990)} mentions that the temperature in particular descends between 0.4 to 0.7 °C per 100 m of increase in altitude. This variation is in function of the relative humidity of the air, even decreasing until 1 °C when the air of the atmosphere is completely dry. In relation to the present study, the pattern of distribution of specific richness as a function of altitude changes, shows the presence of a greater number of species in the intermediate ranges, which assume an altitude of 2 100 to 2 700 m, against high and low altitudes.

^{Margalef (1972)} reports that the Shannon-Wiener index usually ranges from 1 to 5, and interprets values lower than 2 as low diversity, from 2 to 3.5 mean diversity and greater than 3.5 as high diversity. According to this criterion, the results of the present study suggest that the forest communities of the ranges 1 500-1 800 m, 1 800-2 100 m, 2 100-2 400 m and 2 400-2 700 m present a mean diversity in contrast to the altitudinal range of 2 700 to 3 000 m, classified as little diverse. The Shannon-Wiener diversity indexes obtained in this study are much higher than those estimated in other studies in mixed coniferous forests with broadleaved conifers (^{Villavicencio et al., 2005}, ^{Díaz et al., 2012}).

Regardless of the variations in the environmental conditions attributed to the altitude gradient over changes in biodiversity, in the particular case of other mixed coniferous and broadleaved areas, their scarcity seems to be influenced by applied silvicultural practices. In this context, ^{Návar and González (2009)}, when studying the subject in neighboring forests of the study area, found that the Shannon-Wiener diversity index decreases as the cutting intensity is accentuated. The estimated maximum and minimum values were 1.33 and 0.53, respectively, while ^{Castellanos-Bolaños et al. (2008)} calculated results lower than 2.17 for the same index in a forest of Pinus patula Schiede ex Schltdl. & Cham. of the state of Oaxaca. Similarly, ^{García et al. (2012)} determined figures between 1.82 and 2.02, of which the maximum diversity corresponds to forests belonging to inaccessible areas of the Sierra Madre Oriental of the state of Nuevo León.

Although Simpson’s indicators suggest a high diversity in each of the altitude ranges (λ <0.17), their behavior is similar to the Shannon-Wiener index, since the maximum diversity is verified in the range of 1 800- 2 100 m (λ = 0.09) and the minimum in the range of 2 700 to 3 000 m (λ = 0.18). In this way the interpretation of the results of the Shannon-Wiener Equity Index (Pielou Equity Index) is congruent with those obtained with the Simpson Dominance Index (E_λ) ranging from 0.82 to 0.91. According to the two, the altitude range of 2 700-3 000 m brings together some dominant species, and in decreasing order those of 1 500-1 800 m, 2 400-2 700 m, 2 100-2 400 m and 1 800-2 100 m (Table 2).

Based on the Simpson indicators estimated in the present case, and if a community is considered more complex the greater the number of species that comprise it and the less dominance present one or a few species with respect to the others (^{Baca, 2000}), the communities evaluated on this occasion are classified as diverse.

The Hutcheson t test revealed statistically significant differences in most of the pairs of altitudinal intervals evaluated. In particular, the 1 500-1 800 m versus the 2 400-2 700 m, although suggesting a difference in specific richness (14 vs 24 species), the Shannon-Wiener index indicated homogeneity in proportional diversity. This result is attributed to the influence of dominant species, reflected in the indexes of equity and dominance. In this sense, ^{Alanís et al. (2010)} mentioned that the number of individuals affects the differentiation of the composition of the species present in each community between altitude ranges.

Conclusions

Within the altitudinal gradient studied 30 tree species distributed in the Betulaceae, Cupressaceae, Ericaceae, Fagaceae, Pinaceae and Salicaceae families were recorded. Species richness varied from 14 to 25, the Shannon-Wiener diversity indexes from 1.94 to 2.67 and Simpson’s from 0.09 to 0.18; the greatest richness and diversity at intermediate altitudinal intervals was found. Comparison of the Shannon-Wiener diversity index indicates that there is a significant difference between most of the altitude ranges. It was also observed that the dominance of tree species influences the differentiation of diversity between the altitudinal intervals. These results suggest that the environmental variability attributed to the altitudinal gradient plays an important role in the presence and abundance of the species.