Evaluation of Topsoil Carbon Content and Quality in a Peatland and Reforested Soil after 50 Years of Soil Restoration in the Sierra de Guadarrama National Park (Spain)

: The increase in atmospheric CO 2 levels and the advance of deserti ﬁ cation due to soil degradation across our planet is becoming one of humanity’s most serious concerns. The restoration and development of soil management techniques are becoming widespread tools to protect soils. The Sierra de Guadarrama National Park (Spain) is an area that has su ﬀ ered historically severe de-forestation, but it was reforested in an extensive program 50 years ago. In this study, an evaluation of the soils in the restored area was carried out. For this purpose, the chemical composition of the di ﬀ erent soil organic ma tt er fractions was characterized using infrared and UV-vis spectroscopies. The results showed a large increase in carbon stocks in the topsoil (0–10 cm) (about 30 Mg·ha − 1 more than the area not reforested) after reforestation 50 years ago. There was also an increased level of transformation of organic carbon into resilient humic structures, which are resistant to degradation. Reforestation activities within the National Park have greatly increased the humi ﬁ cation rates of organic ma tt er, resulting in the accumulation of high-quality organic carbon.


Introduction
The Earth's environmental situation is becoming very worrying, with the increase in greenhouse gases, especially CO2, causing global warming of more than 1.2 °C in the last century [1].In this whole problem, the soil has an important role, because soils are among the largest carbon reservoirs on the planet after the oceans [2,3].For this reason, preventing soil degradation is crucial for mitigating environmental issues resulting from excessive emissions of these types of gases with the loss of organic matter.
Soil conservation and its protection against degradation is becoming one of the most important objectives worldwide.This has led to strong international initiatives, such as the "4per1000" initiative [4].The advance of desertification is a global world problem due to several aspects [5].On the one hand, there is the loss of fertile soils, both natural and cultivated, along with the associated environmental and socio-economic problems that this fact generates, and, on the other hand, there is the emission of CO2 to the atmosphere due to the degradation of the soil organic matter (SOM).These facts have increased interest in this problem, and some researchers have focused on SOM stabilization pathways, organo-mineral interactions, and chemical stability [6][7][8][9], while others have focused on evaluating the impact of different soil management practices on SOM content and quality [10][11][12][13].Finally, all of them have the common objective of shedding light on stabilization pathways to prevent the loss of this carbon and to better understand the transformation mechanisms.
The restoration of degraded soils is one of the potential tools to combat the advance of desertification while enhancing carbon sequestration and fertility [3].It is necessary to evaluate the impact of reforestation on the soil, particularly regarding carbon sequestration and the transformation of SOM.In this respect, the Sierra de Guadarrama National Park in Spain, where extensive reforestation was undertaken in the early 1970s, provides an ideal scenario to compare the chemical evolution and carbon sequestration in reforested and non-reforested areas.In addition, this area is of special interest due to the peatlands that are also found in it, which are important due to their carbon storage capacity and their vulnerability to prolonged drought scenarios that can degrade these areas.
The SOM has a complex chemical composition and a chaotic structure that remains obscure.However, its structure retains fingerprints of the processes and conditions that have affected its evolution [14,15].Different SOM fractions can be isolated based on their solubility at varying pH values: fulvic acid (FA) is soluble in alkaline and acidic solutions, humic acid (HA) is only soluble in alkaline pH, and humin is insoluble.HA is of special interest in all of these fractions because it is one of the most transformed fractions and usually the most abundant and easy to isolate and purify.This fraction may provide information about the humification processes reflected in its chemical structure [10,11].The study of the chemical composition of SOM and its fractions is important to understand its degree of transformation and stability, for which several techniques have been developed [11,12].Among all of the analytical techniques, there are two that have given us a good idea about the degree of transformation of organic carbon: UV-visible spectroscopy and infrared (IR) spectroscopy.In UV-visible spectroscopy, the wavelengths 465 nm (E4) and 665 nm (E6) have been widely used to provide a global idea about the transformation degree of the HA fraction [8].The ratio E4/E6 is accepted as an index that indicates the evolution and condensation degree of SOM [8,16,17]; more transformed organic matter gives low values in this ratio, while more labile and fresh organic matter shows a high value of E4/E6.An analysis using IR spectroscopy gives a more detailed view of the chemical structure of SOM and can help to assess structural variability under different forms of soil management or environmental conditions.Several bands studied using IR can give information about functional groups in the HA chemical structure, i.e., important bands, such as 1720 and 1620 cm −1 , which are characteristic of carboxylic groups and aromatic carbons, respectively [18].
The aim of this study is to evaluate the potential of reforestation to accumulate organic carbon in the soil and to study the evolution and transformation degree of this organic carbon in the Sierra de Guadarrama National Park (Spain).This work will provide a good approximation of the effect of this reforestation after 50 years on the quality and quantity of organic matter on the surface of the area under study, in addition to evaluating the quality of the organic matter of an adjacent peatland due to its great interest as a potential carbon reservoir.

Study Area and Sampling
The study area is located in the Sierra de Guadarrama National Park (Spain), and sampling was carried out in the specific area called "Puerto de la Morcuera" (40°49′36′'N; 3°50′7′'W) at 1770 m a.s.l (Figure 1).Gneiss is the dominant geological substrate through-out the area, and two main soils have been sampled in the study area-Gleyic Histic Umbrisol and Fibric Histosol-according to WRB [19].The study area was selected because the reforested and non-reforested soils were similar and the profile was not altered in the reforestation process, while in other areas with steeper slopes, it was modified by terracing.Following Koppen's climate classification, the area has a Dsc climate type, where the temperature of the coldest month is below −3 °C and that of the warmest month is above 10 °C, with short and dry summers.Three soils were sampled: one soil under forest vegetation (F) resulting from reforestation that took place more than 50 years ago, with a planting density of about 7 m between trees, mainly with Pinus sylvestris; another adjacent soil, which is still deforested (D), with only herbaceous vegetation and some isolated mountain scrub; and, finally, a peatland (P) located in the same area.Natural vegetation in the area is typical of high-altitude scrubland, with species like Cytisus purgans or Juniperus communis and herbaceous vegetation adapted to cold winters and high-speed winds (Amaryllidaceae and Poaceae).The percentage of course elements in the area is 2.5%, except in the case of the Histosol, where coarse elements were lacking in the topsoil.The herbaceous vegetation in the reforested area was the same, but with a reduced population.Three spatial replicates were sampled at each site, with sampling points 20 m apart, and all soil samples were taken from the topsoil (depth of 0-10 cm).Five separate subsamples were collected at every point (four at the corners of a 2 m side square and the fifth in the center).Finally, all subsamples were combined and homogenized with a total weight of 1 kg.The sites were selected in an attempt to achieve approximately 4% uniformity of slope across the area.The soil samples were transported to the laboratory, air-dried, and then sieved to 2 mm for further analysis.

Soil Characterization
Soil pH was determined using a 1:2.5 soil:water suspension.To determine the electrical conductivity (EC), a 1:5 suspension of soil:water was used.For the peatland samples, the determination of pH and EC was carried out directly in the field.Soil organic carbon (SOC) was analyzed using the Walkley and Black method based on wet oxidation using potassium dichromate in acid medium and then using Mohr's salt solution to quantify the reacted dichromate [20].Soil texture was determined for the surface horizon of the Umbrisol (horizon A) following the Bouyoucos method [21].However, it was not possible to determine the soil texture for the Histosol as it consisted of organic soil.The water holding capacity (WHC) was determined for each soil by wetting by capillarity for 24 h, followed by a 2 h drainage period at atmospheric pressure.Finally, the water content was determined by weighing.The bulk density was also determined using the core method for all of the soils sampled.Cylinders of 7 cm in diameter were introduced into the soil to obtain a sample of the unaltered soil structure.The soil weight was obtained after drying them in the laboratory at 105 °C for 24 h.The SOC stock per hectare was calculated based on information on the SOC content, bulk density, and depth (10 cm) considered in this investigation.

Isolation and Characterization of Humic Substances
To extract soil humic substances (FAs and HAs), we began the process by using 0.1 M Na4P2O7 and stirring in a reducing atmosphere.The mixture was centrifuged after stirring, and the supernatant extract was collected.The procedure was repeated once more with a 0.1 M Na4P2O7 solution, and then with 0.1 M NaOH until a clear extract was obtained.The obtained extract was acidified using 6 M HCl, resulting in the precipitation of HAs, which were later separated from FAs through centrifugation.Once the HAs were isolated, they were redissolved in 0.1 M NaOH and centrifuged at high speed (18,000 × g) to sediment the clays and separate them from the HAs.Finally, to complete the purification of HAs, they were precipitated again with 6 M HCl and dialyzed (Visking tube size 6, Medicell International Ltd., London, UK) to eliminate all soluble salts.The quantification of organic carbon for each fraction was performed by taking aliquots, drying them, and subsequently analyzing them using the Walkley and Black method, as described above [20].
The optical density E4 and E6 of the humic extract (HE), composed of FAs and HAs, was determined at 465 and 665 nm in HE solutions in 0.01 M NaOH at 0.05 mg C•cm −3 .The analysis was conducted using a spectrophotometer GENESYS 150 (Thermo Fisher Scientific Inc.).Optical density at 272 nm (OD272) was also measured, as previous research has indicated a correlation with the degree of condensation of humic substances [22,23].
The infrared spectra were obtained using attenuated total reflectance (ATR) in a Cary 630 FTIR (Agilent) spectrometer.The wavelength range was 4000-400 cm −1 with a resolution of 4 cm −1 .The spectra baseline was corrected, and a resolution enhancement of the spectra was carried out following the protocol described by Almendros et al. [24].This was performed to enable a clearer visualization of the bands.Finally, a semiquantitative analysis of the characteristic bands was conducted using the second derivative of the spectra.Several bands were selected for this study due to their importance in the chemical structure of organic matter [18,25,26]: the band at 1720 cm −1 is associated with carboxyl groups, the aromatic structures are represented by the bands at 1620 and 1510 cm −1 and the group of bands related to the structure of lignin (1510, 1420, 1380, 1270, 1230, and 1130 cm −1 ) and, finally, the band at 1030 cm −1 is related to polysaccharide structures [18].

Statistical Analysis
A statistical test of analysis of variance (ANOVA) between groups was performed using Statistica ver.7.1 software to evaluate significant differences.For an early assessment of the potential connections between soil properties, such as the SOC content, and the chemical properties of the SOM fractions, we conducted a principal component analysis (PCA).The soil properties and the normalized intensities of the principal IR bands measured in the second derivative were used for the PCA using Statistica ver.7.1 software.
Finally, a third statistical analysis was carried out using cluster analysis.This analysis used the 1-Pearson correlation index to measure linkage distances.The proximity of the variables is related to the correlations between them, and those located closely together reveal a high correlation.

Soil Properties
The general results of the soil properties analysis are shown in Table 1.All soils present acid pH, but a difference has been observed in the case of peatland, where the lowest acidity is present, which is possibly due to the washing effect of the water in this humid location.This effect can also be observed in the EC, with the lowest EC of 11.3 µS•cm −1 in the peatland and higher values in the reforested area (80.6 µS•cm −1 ) and deforested area (52.6 µS•cm −1 ).The highest value of WHC was observed in the peatland, followed by the forest and the deforested soil, with values of 722, 95 and 72%, respectively.The soil in the forested and deforested areas is classified as loamy sand based on texture.The highest SOC content was found in the peatland (30.5%), followed by reforested soil (8.4%) and, finally, deforested soil (7.0%).In the results of the estimated SOC stock per hectare, we can see that the sequence is consistent.However, in this case, the stock in the reforested area shows values close to the peatland stock, and it is 35.5 Mg•ha −1 higher than the SOC stock of deforested soil.
Table 1.General characteristics of samples from reforested (F), deforested (D), and peatland (P) soils (0-10 cm).The standard deviation is shown next to the mean value of the variables.The percentages are expressed with respect to the weight of the soil, except for the particle size fraction (clay, silt, and sand) percentages, which are presented in relation to the content of the total mineral particle.Different letters indicate significant differences, p < 0.05.

Organic Carbon Characterization
The values of the different fractions isolated from the SOM are presented in Table 2.We see a notable difference between the amount of HE in peatlands (26.7 g C•100 g SOC −1 ) and other soils, but no significant differences are observed between reforested and deforested soils, for which the values are 52.1 and 55.0 g C•100 g SOC −1 , respectively.A similar trend is observed for the content of FA and HA fractions.The characterization of humic substances through UV-visible spectroscopy revealed significant differences between the three soil types.The highest OD272 value is found in the reforested soil (2.658 absorbance units, AU), indicating the high degree of condensation of the chemical structure of humic substances (Figure 2) [22,23].A less condensed HE structure is found in the peatland (1.651 AU), while the HE extracted from the topsoil of deforested area shows a moderate degree of condensation (2.029 AU).The E4/E6 values behave inversely to the values at 272 nm mentioned above.The highest value, which corresponds to a lower degree of organic matter transformation, is found in the peatland (5.828).The reforested soil presents the lowest E4/E6 value (4.385), and the next is the deforested soil, with a value for the ratio of 5.207 (Figure 2).Table 2. General characteristics of humic extract (HE) isolated from the different soils samples (0-10 cm): reforested (F), deforested (D), and peatland (P).The standard deviation is shown next to the mean value of the variables.Different letters indicate significant differences, p < 0.05.   Figure 3 shows the IR spectra of the HA isolated from the different soil samples.Figure 3a displays the original IR spectra, while Figure 3b depicts the spectra obtained after the resolution enhancement.The HA replicates from each soil demonstrate a notably comparable spectrum.On the one hand, it is apparent after visual inspection that the bands relating to carboxyl groups and aromatic carbon (1720 and 1620 cm −1 , respectively) are more prominent in the HA extracted from the reforested and deforested soils, whereas in the peatland, these bands are less notable.On the other hand, the lignin-associated bands are more pronounced in peatland HA compared to the other soils, particularly the band at 1030 cm −1 , which is linked to polysaccharides.The PCA biplot is presented in Figure 4, and it shows that the first two factors account for 83.78% of the overall inertia of the variables in the plane.Two well-defined clusters formed by the variables can be observed.Two distinct variable clusters are evident, with the first comprising bands 1510, 1420, 1380, 1270, 1230, 1130, and 1030 cm −1 , linked to the lignin structure, SOC content, and the E4/E6 ratio.The second group comprises the HE content, the OD272, and the bands 1720 and 1620 cm −1 , which are associated with carboxylic groups and aromatic carbon, respectively.The latter group exhibits clear indications of significant SOM transformation and level of humification.By examining the sample positions in the biplot, it is evident that the peatland samples are significantly influenced by variables related to fresh organic matter, while both the reforested and deforested soil samples are situated adjacent to variables indicating increased humification; in particular, scores for samples from the reforested area are located near the eigenvector for the 1720 cm −1 band and the values of OD272.).The biplot suggests two main clusters: the first embraces variables related to well-transformed organic matter (yellow cluster), while the second (blue cluster) includes variables associated with relatively more fresh organic matter.
The cluster analysis results are consistent with those obtained through the PCA (Figure 5).Two main clusters are distinguished by their IR bands; one comprises the bands associated with the lignin structure, while the other is characterized by carboxyl and aromatic structures.The SOC content shows an outstanding correlation with both WHC and pH, and all of these variables are correlated with the IR bands at 1510, 1420, 1380, 1270, 1230, 1130 and 1030 cm −1 (bands associated with the lignin pattern) and the E4/E6 ratio.The EC, OD272, and HE content correlate well with the 1720 and 1620 cm −1 bands, which are associated with the more transformed SOM.

Discussion
The reforestation process has exhibited positive progress in organic carbon sequestration compared to the deforested regions.After 50 years, the SOC stocks in the topsoil have increased by 35.5 Mg•ha −1 compared to the area that was not reforested.Although peatland is the largest organic carbon reservoir, it is also highly labile, making it extremely susceptible to environmental changes, such as marked declines in the water table or high temperatures that may stimulate mineralization processes [27].These facts have been evident in several variables analyzed during the chemical characterization of the SOM.The HE content is widely used to explore the humification rate [28].It has been seen that the peatland showed much lower humification rates compared to the other soils.This is consistent with the waterlogged conditions, which favor anaerobic environments that inhibit microbial activity and thus the rapid transformation of organic matter, resulting in decreased FA and HA contents [29,30].These results are supported by both the E4/E6 data and the OD272.The peatland's HE exhibits the highest E4/E6 values, which indicates a lower level of transformation and limited polymerization or molecular weight, and it is accompanied by low values of OD272, which shows a low degree of aromaticity [22,23].The organic matter of the reforested soil showed the lowest E4/E6 values and the highest OD272.This suggests that carbon sequestration has increased in the topsoil, and the degree of humification in the area has also improved after 50 years of reforestation.In brief, this area of the National Park stores more carbon, and the carbon is more chemically stable.The reforestation of the area has had a positive effect on the improvement of the SOM content and its quality, at least within the first 10 cm.This confirms earlier research that showed a correlation between enhanced diversity in the natural environment and increased organic matter transformation [31,32].This range of diversity presents a challenge in agricultural soils, which are often monocultures, and carbon sequestration processes typically store less humified and more labile carbon.This issue can be addressed by implementing vegetation covers that improve carbon storage and increase biodiversity while also enhancing the humification process in relation to the mineralization of organic matter [31][32][33].Therefore, the restoration of natural ecosystems is crucial, as it allows for the manipulation of biodiversity, thus leading to a more stable increase in organic carbon levels in the soil.
The analysis of the chemical structure of HA using IR spectroscopy revealed variations in the organic matter evolution among the different soils.HA isolated from the peatland exhibited a conspicuous pattern of lignin structures and an abundance of carbohydrate-derived structures.These findings suggest a low level of humification reflected by the similarity of the IR patterns to those typical in fresh vegetation [25,26].On the contrary, the structures observed in the other soils display more aromatic structures, and they are enriched with carboxyl groups, indicating a more advanced state of transformation.All of these facts suggest that SOC storage is strongly linked to peatland conditions, but this accumulation is primarily due to a very labile carbon.In comparison, other soils have less carbon but are more evolved and stable.Consequently, certain properties, such as WHC, are highly correlated with SOC content due to the importance of organic matter for soil structure and water retention.
All of these findings illustrate the importance of restoring damaged natural habitats.The advancement of innovative soil management techniques facilitates the increase in carbon levels in soils and improves their properties [34,35]; however, it is crucial to consider accumulating resilient carbon structures that can endure extreme climatic conditions.Not only should we consider the amount of organic carbon, but also its quality and stability, as they are important factors.
It is worth noting that this study has focused on the topsoil, because it is where the first changes in organic matter can be observed.However, changes could be taking place in deeper layers.This fact could be an object of study for future research because the carbon stock in the deep layers is of great interest, and its chemistry could vary in different scenarios.

Conclusions
After 50 years, the reforestation initiative of Sierra de Guadarrama National Park has yielded exceptional SOC sequestration outcomes in the topsoil.The reforestation of the study region with Pinus sylvestris resulted in an estimated increase of 35.5 Mg•ha −1 in the initial 10 cm compared to the non-reforested area.Furthermore, the process of reforestation had a significant impact on the transformation and stabilization of organic carbon in the soil, resulting in a more aromatic and polymerized structure.
These processes for restoring soil have been deemed of great importance in soil conservation, not only due to their high level of carbon storage but also because of their ability to stabilize carbon, which can be difficult to achieve in other scenarios, such as with agricultural soils.This can open a range of possibilities in the study of new soil management not only in forest soils, but also in agricultural soils, where the beginning of the use of cover crops can allow for achieving greater stabilization in addition to increasing the organic carbon stock.

Figure 1 .
Figure 1.Study area in the Sierra de Guadarrama National Park.Sampling points are marked at each site.Green circles represent reforested soils (F), blue triangles represent deforested soils (D), and red squares represent peatlands (P).

Figure 3 .
Figure 3. Infrared spectra of purified humic acid (HA) extracted from reforested soil (F in green), deforested soil (D, blue), and peatland (P, red).(a) Original spectra.(b) Resolution-enhanced spectra.The box with a dashed line indicates the region with bands characteristic of lignin.

Figure 4 .
Figure 4. Biplot obtained through principal component analysis; soil samples are represented with green circles: reforested soil (F), deforested soil (D), and peatland (P).Variables used to perform the PCA are represented with red circles: SOC, E4/E6, HE, OD272, and the wavenumbers in the infrared spectra (1720, 1620, 1510, 1420, 1380, 1270, 1230, 1130 and 1030 cm −1).The biplot suggests two main clusters: the first embraces variables related to well-transformed organic matter (yellow cluster), while the second (blue cluster) includes variables associated with relatively more fresh organic matter.

Figure 5 .
Figure 5. Cluster analysis of the soil variables, including the intensities of the infrared bands of the HAs using the 1-Pearson correlation index to determine linkage distances.