Study area

The experiment was carried out at the Zoquiapan Experimental Forest Station (EFEZ) located between 19°12'30 "and 19°20'00" N and 98°42'30 "and 98°30'00" W. One part is extended in the State of Mexico and another in the state of Puebla, (Figure 1). It has an area of 1 638 ha and an average altitude of 3 284 masl. The climate is subhumid temperate with rain in summer, with an average annual temperature of 11.1 ºC and an average annual rainfall of 964.1 mm (^{Inegi, 1981}).

Figure 1 Location of the Zoquiapan Forest Experimental Station. 

The predominant relief is mountainous, with some plains, with a variation in the slope of 2 to 50 % (^{Vega, 1982}). The soils are classified as Molsic Andosols, with good aeration, good drainage, friable consistency and slightly sticky and plastic; the pH with values ranging between 5.06 and 6.3, moderately acidic and with good nutrient availability (^{Rey, 1975}, ^{Aguilar and Plateros, 2003}).

The vegetation consists mainly of three high mountain communities: pine forest (Pinus hartewegii Lindl.) Widely distributed throughout the area covering 65 %; fir forest (Abies religiosa (Kunth) Schltdl. et Cham.) which occupies 10 % of the area, with a very closed canopy; and aile forest (Alnus jorullensis Kunth), which covers 24 % of the area, widely distributed throughout the season and in some places forms pure masses (^{Herrera and Rodríguez, 1975}).

Location of sampling sites

To locate the sampling sites, in the first instance, the different types of vegetation or cover were obtained through the supervised classification of Landsat 8 images (Path 26 and Row 47). The types of vegetation were: oak-pine forest, fir forest, pine forest, pine-alder forest, pine-fir forest and pasture. Of these, three were selected: pine forest, firforest and the pine-alder forest association, because they cover about 90 % of the total area of the EFEZ.

To estimate the slope of the terrain, the digital elevation model (MED) with spatial resolution of 15 m was used (^{Inegi, 2013}); by the ArcGIS version 10.3^® software and the Slope tool, a raster file was generated with land slope values, which were classified into three categories: 0-10.10, 10.10-21.16 and 21.16-48.43 degrees. By means of the intersection of the layers of coverage and slope, a map of systems with the same coverage and slope was obtained. Table 1 describes the characteristics of the nine systems.

For the location of the sampling plots within the systems, homogeneous coverage and slope characteristics were considered, as well as the criteria established by NOM-021-Semarnat-2000 (^{Semarnat, 2002}).

Soil sampling

Once the systems were delimited, in each one of them 10 circular plots of 400 m² distributed regularly in the total area of the system were proposed, of which three were randomly chosen and georeferenced. In the field the plots were located and at the center of each of them the soil samples were extracted at 0 to 10, 10 to 20 and 20 to 30 cm deep; one for the determination of apparent density values, for which the cylinder auger (Eijkelkamp C) was used and the other for the analysis of the chemical properties of the soil, for which a Dutch type auger was used (Eijkelkamp C). In each site, 54 samples were collected (3 plots × 3 slopes × 3 depths × 2 types of sample). In total, 162 samples were collected.

Physical and Chemical Determinations

The samples were dried in an oven (THELCO 28) at 110 ± 5 °C until reaching a constant weight. Each sample was sifted and the bulk density determined by the test method (^{Sánchez et al., 2011}). For the concentration of COS and Nt, they were dried in the open air and sifted through a 2 mm mesh, and then sent to the Laboratorio Central Universitario de la Universidad Autónoma Chapingo (Central University Laboratory of the Autonomous University of Chapingo) to conduct tests to calculate organic matter and Nt concentrations.

The concentration of organic carbon (CO) was estimated from organic matter by the method of ^{Walkley and Black (1934)}; the concentration of CO was obtained by means of Van Benmelen factor 1.724, which was divided by the percentage of organic matter, since it is assumed that this component contains 58 % organic carbon. Such content was calculated based on the following equation (^{González et al., 2008}):

Where:

COS = Organic carbon concentration in the soil (Mg ha^-1)

Co = Carbon concentration in the soil (%)

Dap = Bulk density (g cm^-3)

Ps = Soil depth (cm)

The concentration of total nitrogen was obtained by steam distillation with the micro-Kjeldahl process, from the following equation:

Where:

N-tot = Nitrogen concentration in the soil (%)

Dap = Bulk density (g cm^-3)

Ps = Soil depth (cm)

Organic carbon content in the soil

Coverage influence

For most of the values in Table 2, regular behavior of COS contents according to coverage is observed; for the three slopes and depths, the values are lower in the pine forest, they increase in the pine-adler forest and decrease in the fir forest with respect to those of pine and adler.

Of the three highest COS values for the coverage, two corresponded to the one of pine-aile with 130.80 Mg ha^-1 and 122.47 Mg ha^-1; the third corresponded to the coverage of the fir community with 132.20 Mg ha^-1. These figures are similar to those mentioned by ^{Vela et al. (2012)} who found COS contents in fir and pine conservation forests in Mexico City of 145.5 and 119 Mg ha^-1, respectively. They also resemble those reported by ^{Pérez et al. (2013)}, who conducted a study of the COS content in conserved areas in the Monarch Butterfly Biosphere Reserve in the State of Mexico, and calculated an average value of 153 Mg ha^-1 in the soil. Regarding the three lower COS values, these corresponded to the pine coverage with 30.86 Mg ha^-1, 32.62 Mg ha^-1 and 36.39 Mg ha^-1.

When analyzing the behavior of the COS content for the coverages in the three depths, it can be seen that the coverage with the highest COS contents was that of the pine-aile forest with 130.80, 122.47 and 94.90 Mg ha^-1. The coverage with the lowest COS content for the three depths was that of pine with 49.65, 36.39 and 32.62 Mg ha^-1. This behavior can be explained by the degree of conformation of the areas and forest management, since the pine-alder and fir coverages were the most compact and with less degradation in their interior, which contrasts with that of the pine forest, which shows intervention of productive activities of the livestock type.

Slope Influence

In general, the COS content based on the slope factor had a regular behavior in the pine coverage at the three depths, lower for the low and high slopes, and with an increase in the intermediate. For pine-adler coverage, there was an irregular behavior for the three slopes and three depths. In the 0-10 cm depth of the fir community, the COS content increased as the slope did, but for the 10 to 20 and 20 to 30 cm depths, the opposite occurred.

Of the three most outstanding values of COS as a function of slope, one was found at the steepest slope with 132.20 Mg ha^-1, another at the intermediate with 122.47 Mg ha^-1 and the third at the lowest at 130.80 and Mg ha^-1.

It is noted that the slope with the highest COS content at the three depths was the low one with 130.80, 122.47 and 94.90 Mg ha^-1 with the pine-alder cover. And the opposite was verified in the minor, with 49.65, 36.39 and 32.62 Mg ha^-1 with the pine cover.

^{González et al. (2008)} noted that in most plant ecosystems there is a tendency to concentrate more COS in the lower position of the slopes. The above statement is valid in the analysis of the effect of the greater and lesser slope for the three depths in the pine-alder cover.

Depth Influence

In most cases the COS content decreased with depth. The previous behavior coincides with the findings of ^{Alvarado et al. (2013)}. According to the ^{IPCC (1997)}, in soils the change in the COS is relatively small and occurs mainly in the first 30 cm of the soil layer. According to ^{Schargel and Marvaez (2009)}, the above is due to the fact that organic carbon is linked to organic matter, so its abundance decreases with depth, since the accumulation of organic remains and the activity of microorganisms occurs in the first centimeters of the ground.

From the three highest COS values for depth, two were found from 0 to 10 cm with 132.20 and 130.80 Mg ha^-1, and the third at 10-20 cm with 122.47 Mg ha^-1.

Of the three lowest COS contents, two came from 20 to 30 cm, with 30.86 Mg ha^-1 and 32.62 Mg ha^-1, and the last one from 10 to 20 cm with 36.39 Mg ha^-1.

Carbon content from 0 to 30 cm depth

Table 3 shows the COS contents of 0-30 cm depth, obtained as the sum of the three partial depths and as a function of the coverage and slope grades. It is observed that the coverage with the three most outstanding COS values were those of pine-adler with the lowest slope, followed by the same coverage with the intermediate slope and finally the one of fir with the steepest slope. The coverage with the three lowest values of COS was that of pine with the lowest and highest slopes and that of pine-adler with the highest slope. The COS contents obtained are higher than those calculated by other authors such as by ^{Segura et al. (2005)} in humid forests with 110.5 Mg ha^-1. The differences may be due to the fact that temperate forests have lower temperatures compared to wet forests where mineralization rates are higher.

Total Nitrogen content in the soil

Coverage Influence

Similar to the COS, in Table 3 a regular behavior is observed for most of the Nt content values, depending on the coverage for the three slopes and depths, with lower values in the pine, increasing in pine- alder and diminishing in fir compared to that of pine and alder.

It is observed that the three covers with the highest Nt content in the soil were found in the fir community with 14.19 Mg ha^-1, which is followed by the pine-alder with 11.61 Mg ha^-1 and, thirdly, again is found in the pine-alder forest with 10.57 Mg ha^-1. The coverage with the three lowest values of Nt was that of pine with 3.37 Mg ha^-1, 3.81 Mg ha^-1 and 4.62 Mg ha^-1.

In the analysis of the Nt content for the three depths, it can be seen that the pine-alder coverage had the highest Nt content with 10.57, 9.93 and 9.33 Mg ha^-1, respectively, and the pine cover had the lowest Nt content., with 4.62, 3.81 and 3.37 Mg ha^-1. According to the classification of nitrogen and level of abundance in the soil (^{Semarnat, 2002}), the fir forest and pine-alder forest are rich in this element; the pine forest was defined in the middle level range. ^{Arzuaga et al. (2016)} evaluated the content of Nt in the soil under cover of preserved pine forest, and calculated values of 2.69 Mg ha^-1, while for the deforested areas there was a reduction of around 51 % in the first 30 cm.

Slope influence

The Nt values as a function of the slope showed the most irregular behavior. For the slopes with the pine cover and the lower and intermediate depths, it was lower for the lower slope, increased for the intermediate slope and decreased again with the steepest slope. For the greater depth, the content of Nt was smaller in the opposite direction to the slope. For the pine-alder cover at the depth of 0-10 cm, the content of Nt decreased when going from the lower slope to the intermediate slope, later it increased in the greater one. For the depths of 10 to 20 cm, the contents of Nt were lower as the slope increased. For the coverage of fir, the content of Nt was lower in the opposite direction to the slope for the depths of 10 to 20 and 20 to 30 cm, but to the 0 to 10 cm, it first decreased and then increased.

Two of the three highest Nt values were recorded on the higher slope with 14.19 Mg ha^-1 and 11.61 Mg ha^-1, and the third on the lower slope with 10.57 Mg ha^-1. The three lower contents of Nt (3.37 Mg ha^-1, 3.81 Mg ha^-1 and 4.62 Mg ha^-1) were recorded with the highest slope.

The slope with the highest Nt content at the three depths was the lowest, with 10.80, 9.93 and 9.33 Mg ha^-1. The opposite was presented for the three depths in the greater, with 4.62, 3.81 and 3.37 Mg ha^-1.

The results of the analysis and the comparison between the lower slope and the greater one allow us to affirm that the content of Nt decreased as the degree of inclination advanced, which is supported by the results obtained by ^{Lozano et al. (2016)}; these authors asserted that the topography, slope and orientation of the land are variables that should be included in the estimates of Nt and COS at the local and regional scales, since there are differences in the stores of these nutrients. ^{Ávila et al. (2004)} noted that the concentration of Nt is reduced as the slope increases.

Depth Influence

The depth factor had the expected behavior for the content of Nt, since in most cases, regardless of the slope and coverage, it was reduced with depth in a similar way to what happened with the COS content. The above agrees with the results of ^{Ávila et al. (2004)}, who pointed out in their study that the concentration of Nt decreased as the depth of the soil increased. This decrease, according to ^{Schargel and Marvaez (2009)}, is due to the fact that nitrogen is linked to organic matter.

The three largest magnitudes of Nt content were for the depth of 0-10 cm with 14.19 Mg ha^-1, followed by 11.61 Mg ha^-1 and finally 10.57 Mg ha^-1. The three lowest contents of Nt were given at the depth of 20 to 30 cm with 3.37 Mg ha^-1, secondly to the 10 to 20 cm with 3.81 Mg ha^-1 and thirdly that of 0 to 10 with 4.62 Mg ha^-1.

Total nitrogen content from 0 to 30 cm depth

For Nt, the coverages with the three highest values were those of pine-alder with the lowest slope, followed by that of fir with the greater and lesser slope. The coverage with the three lowest values of Nt were those of pine with all slopes. These results were higher than those reported by ^{Sacramento et al. (2013)} of 10.15 Mg ha^-1 in a natural system of the Sobral municipality, Ceará, Brazil and by ^{Carvajal et al. (2009)} of 10.41 Mg ha^-1 in forest and 6.65 Mg ha^-1 in coffee plantations.

The data obtained for the pine-alder coverage, and the content of Nt, agree with the fact that the areas covered by species of the genus Alnus (Alnus firmifolia) in association with pine, are of ecological value for the storage and fixation of N₂; therefore, this hardwood species represents an alternative for the management of forest areas and recovery of nitrogen in the soil (^{Gutiérrez-Guzmán et al., 2005}).

When intervention exists in the conserved areas, the storage of carbon and nitrogen decreased; the pine cover revealed the lowest values and the largest reservoirs were always for the pine-alder and fir forests. This fact could be explained by the composition and the greater magnitude of its plant covers, as they are better preserved.