How Natural Regeneration After Severe Disturbance Affects Ecosystem Services Provision of Andean Forest Soils at Contrasting Timescales

Abstract

Chile holds ~50% of temperate forests in the Southern Hemisphere, thus constituting a genetic–ecological heritage. However, intense anthropogenic pressures have been inducing distinct forest structural-regeneration patterns. Accordingly, we evaluated 22 soil properties at 0–5 and 5–20 cm depths in two protected sites, with similar perturbation records but contrasting post-disturbance regeneration stages: long-term secondary forest (~50 y) (SEC_FORST) (dominated by Chusquea sp.-understory) and a short-term forest after disturbance (~5 y) (FAD_IST) within a Nothofagus spp. forest to determine the potential of these soils to promote nutrient availability, water cycling, soil organic carbon (SOC) sequestration (CO₂→SOC), and microbiome. Results detected 93 correlations (r ≥ 0.80); however, no significant differences (p < 0.05) in physical or chemical properties, except for infiltration velocity (+27.97%), penetration resistance (−23%), SOC (+5.64%), and % Al saturation (+5.64%) relative to SEC_FORST, and a consistent trend of suitable values 0–5 > 5–20 cm were estimated. The SOC→CO₂ capacity reached 4.2 ± 0.5 (FAD_IST) and 2.7 ± 0.2 Mg C y⁻¹ (SEC_FORST) and only microbial abundance shifts were observed. These findings provide relevant insights on belowground resilience, evidenced by similar ecosystem services provision capacities over time, which may be influenced progressively by opportunistic Chusquea sp.

1. Introduction

Terrestrial organic carbon (C) stored in soil (SOC) controls multiple scalar natural processes as ecosystem functionality (water and nutrient cycling), which are primarily mediated by microorganisms [1]. However, historically, ~132 Pg of SOC pool (8.8% of total) have been released into the atmosphere in response to human actions (greenhouse gas emissions) (SOC→CO₂), causing land degradation [2] and climate change [3,4,5,6], and currently accounting for a net loss trend of 0.22–0.53 Pg y⁻¹ [7]. Notwithstanding the above, soils represent the greatest CO₂→SOC potential globally, and specific capacities are associated with climatic zones, individual native properties, and historical land use records [8], where deep soils (≥1 m) from temperate to cold zones and mineral fractions rich in Fe-oxides and active clays, probably fulfilling the best conditions [8,9]. Regarding most favorable ecosystems facilitating (SOC→CO₂), forests have a great sink potential, storing 861–967 Pg of SOC (0–1 m), encompassing up to 30% of total emissions within the pedosphere [10,11,12,13]. Particularly, temperate forests, defined as “highly-seasonal biomes, dominated by freeze-hardy woody species due to remarkable intervals of long growing and a cold winter period” [14], are distributed across 23.5–66.5° N and S latitudes, covering 19–26 % of global forest area, having a presence on all continents (except Africa), usually meet the criteria mentioned above [12,15].

The main aspects associated with the relevant CO₂→SOC capacity in temperate are (a) at ecosystem level: (i) constant inputs of SOC precursors (plant debris), (ii) permanent soil cover, (iii) deep-rooted systems, increasing the uptake-recycling nutrients, and (iv) suitable conditions for biological activity [16]; (b) at pedosphere level: (i) an increase in recalcitrance due to lignin-rich substrates, (ii) lowering soil temperature–moisture reducing SOC mineralization rates, (iii) the formation of stable aggregates, and (iv) mineral-associated C [9,17,18,19]; and (c) at microscopic level: (i) inputs of fresh organics (e.g., leaf/herbaceous strata litter), (ii) nutrient bioavailability processes such as organic re-synthesis, nutrient mining cycling (e.g., N, S. P, Fe, Cu) for plant development, and (iii) structural and functional microbial pools (e.g., active microorganisms and those able to restore the soil to a pre-disturbance condition), leading to re-aggregation processes mediated by biological byproducts [2,9,17,20,21].

However, forest degradation caused a decline of 3% of the global area (4128 M ha) over the last 35 y, reversing the CO₂→SOC processes and mechanisms previously described [22,23]. One of the most severely affected regions is South America, reaching a forest loss area of 88,803 K ha [10,24]. In addition, it is estimated that only less than 20% of temperate forests are present in the Southern Hemisphere, resulting in one of the highest levels of endemism (85% of woody species and 34% of genera, prevailing evergreen species) [12,15].

The previous scenario is critical for genetic preservation in Chile, which holds 50 and 79.5% of the total temperate forest area in the Southern Hemisphere and continental levels, respectively (33–55° S latitude, 14.41 M ha), dominated by Nothofagus spp. (pure or mixed) [12,25,26,27,28]. Local forest suppression has been subordinated to agricultural purposes, excessive logging for fuelwood, over-grazing–browsing, and the replacement by commercial plantations (e.g., Pinus radiata, Eucalyptus nitens) due to the high productivity and timber quality of these forest types (including soils) [28,29]. In response, nowadays, Chile holds 23% of the protected forest area in South America, due to extensive government and institutional efforts that have led to the implementation of land reclamation programs (including a sustainable agro-silvicultural policy framework) [12]. Nonetheless, vast derelict territories consequently undergo natural regeneration processes (e.g., secondary forest), many of them based on Andisols, which are of high agricultural importance, accounting for 50–60 of arable land [30]. The above suggests a need to evaluate the soil ecosystem services status in those areas since their inherent high CO₂→SOC potential is widely reported in the literature [7,31,32].

Accordingly, we aimed to compare different soil properties that are useful for establishing fertility status and the current capacity to provide ecosystem services of two contrasting time-lapse scenarios within an unmanaged post-disturbance Nothofagus sp. association forest; both scenarios are highly representative of the aforementioned context and dominant biome type across the south-central macro-zone of Chile.

We hypothesized that soil properties–ecosystem services capacities of long-term passive management have been influenced by the effects of emergent now dominant Chusquea spp. understory, particularly, generating (i) a robust microbiome and particularly decomposer communities; (ii) a greater fertility status; (iii) better physical aptitudes related to hydric conductivity; and iv) a lower annual CO₂→SOC capacity rate.

2. Material and Methods

2.1. Study Area Description

The study was carried out at the “Ranchillo Alto” state-owned property, a 635 ha native forest that was declared “National Heritage” and a “Protected National Asset” “by the Chilean Government; it is located in the Ñuble Region (Figure 1), nearby the town of Yungay, within the biological corridor “Nevados de Chillán—Laguna del Laja”, which was designated as a “Biosphere Reserve” by UNESCO [33]. The local climate is classified as temperate Mediterranean, with average temperatures of 13.5–25 °C during the summer and a pronounced winter season from May to September, with frequent snowfall–frost events and a mean annual rainfall of over 3.000 mm [33]. Through a recognition visit and preliminary site examination (e.g., interviews with local landowners, soil sampling, pers. commun.) [34], it was determined that the area possesses a historical record of over-management–utilization of native Nothofagus spp. involving diverse detrimental activities (such as those mentioned in the previous section), causing the interruption of biomass–litter production, which is the partial to total detachment of soil cover and even topsoil, evidenced by the presence of gullies, leading to consequent massive losses of soil organic matter and nutrients (Figure 2) [21,35].

However, those processes of land degradation occurred at different spatial–temporal scales that do not always agree with the government–social policies implemented to cease them, and (if this is the case), nor do the actions implemented for its reclamation (e.g., agroforestry systems, conservation silviculture) [29,35], or, in the absence of these, different timeline patterns of natural regeneration (e.g., passive management), leading to a complex mosaic of ecological–vegetational conditions [33].

Accordingly, we evaluated two temporal-contrasting sectors within the derived secondary forest, both meeting natural regeneration/no history of post-disturbance management and a similar land degradation record. A description of each condition is provided as follows (

Table 1. Description of study cases.

Condition	Location	Assessed Area (ha)	N° Plots and Area (ha)	Tree Density (No ha−1); Average Tree Height (m) and Diameter at Breast Height (DBH) (cm)	Forest Species	Understory-Herbaceous Strata Composition	Degradation Record
FADIST	37°03′31.87″ S, 71°38′21.76″ W 1367 m.a.s.l	4	3 × 1.33	60; 15 and 34	Roble (Nothofagus obliqua (Mirb.) Oerst. - coihue (Nothofagus dombeyi) (Mirb.) Oerst. (5:1)	Vetch (Fabaceae purpurea), clover (Trifolium incarnatum, T. subterraneum, and T. vesiculosum), Lolium multiflorum westerwoldicum, Phalaris acuatica, Lolium perenne, oats (Avena sativa), Festuca arundinacea, Dactylis glomerata L.), Chusquea sp., and there-sprouting of Radal (Lomatia hirsuta (Lam.) Diels)	Evidence of wildfire, agricultural burning, and intense logging. Testimony of use of intensive animal–mechanical loads; cattle grazing–browsing, prolonged commercial, and domestic timber harvesting
SECFORST	37°4′43.01″ S, 71°39′25.32″ W 1250 m.s.n.m	4	3 × 1.33	296; 33 and 46	Roble (Nothofagus obliqua)- coihue (Nothofagus dombeyi) (8:3)	Primarily dominated by Chusquea sp., a circumstance that has been reported in the literature during regenerative dynamics processes post-disturbance (e.g., Muñoz and Gonzalez, 2009; Donoso et al., 2022 [28,36])	Evidence of fire events. Testimony of use of intensive animal loads; cattle/goat/equine grazing and/or browsing, prolonged commercial and domestic timber harvesting

).

The soils were classified as medial, amorphic, and mesic Typic Haploxerands within the “Santa Barbara” series with an Ap/A/AB/Bw/BC/C sequence, according to the Soil Survey Staff (2022) [37] and Stolpe (2006) [38].

2.2. Soil Sampling and Characterization

To analyze soil properties under different post-disturbance conditions, we collected soil samples from two distinct forest sites: a long-term secondary forest (SEC_FORST) and a recently disturbed area (FAD_IST). These samples allow us to compare how soil chemistry, microbiology, and physics vary over time. In January 2020, three composite soil samples (2 kg each) were taken from three representative plots (1.33 ha each) at two depths: 0–5 cm in triplicate (n = 36), which represents the most biologically active soil layer, and 5–20 cm, where deeper-rooted microbial communities and chemical interactions occur. This depth selection ensures that we capture key soil processes occurring at different strata.

2.2.1. Physical Trials

To assess how soil structure changes over time, we measured different physical properties that influence water retention, root penetration, and overall soil stability. These properties help determine whether the soil is improving or degrading after disturbance, which is crucial for forest recovery. The following tests were performed: (i) Bulk density (BD) measures how compacted the soil is, which affects root growth and water movement. A denser soil can limit plant development and microbial activity. We determined BD by extracting soil cores (211 cm³) and drying them at 105 °C until reaching a constant weight; (ii) soil particle density (PD), estimated by the pycnometer method (Pobel, Gay-Lussac pycnometer with thermometer 10–35 °C/0.2 °C; frosted 10/19, 50 mL, Madrid, Spain) [39]; (iii) net pore space (POR), which was calculated from BD and PD values:

POR = [1 − (BD/PD)] × 100;

(1)

(iv) infiltration rate-velocity (I_NFV), unsatured (k), which was tested utilizing an infiltrometer model Mini Disk Infiltrometer S (Pullman, WA, USA), where the K value (cm day⁻¹) was estimated according to Zhang [40], through cumulative infiltration measurements; (v) water holding capacity (WHC) was estimated according to Zagal et al. [41], placing samples into cones previously sealed at the bottom at 1:2 soil–water ratio for 12 h. Therefore, the tape was cautiously perforated in order to drain the formed solution while it was collected in plastic bottles and its volume was then measured; (iv) water-stable aggregates (WSA) (%) were estimated by the Kemper and Rosenau [42] method, which consisted of sieving (0.250 mm) each sample, then immersing them into an aluminum capsule containing distilled water for 3 min (at 35 rep min⁻¹) by using an Eijkelkamp Agrisearch Equipment (Giesbeek, The Netherlands), and the resulting dispersed soil was dried at 105^◦C, while the remaining soil was placed into another aluminum capsule containing sodium hexametaphosphate (2 g L⁻¹) for 15 min (35 rep min⁻¹), and the dispersed soil was dried at 105 °C. Finally, samples for both procedures were weighed to calculate their ratio/percentage in comparison with that of the original sample, and (vii) penetration resistance (P_ENR) was estimated using a penetrometer model Soil Compaction Tester Dickey–John (Auburn, IL, USA), where a total of 60 measurement points were established by longitudinal transects to ensure representativeness.

2.2.2. Chemical Trials

Chemical analyses were performed according to methods compiled by Sadzawka et al. [43] at the Agricultural Research Institute of Chile (INIA-Quilamapu, Chillán, Chile). After the air drying procedure and extraction of a <2 mm fraction from the samples, particular analytical methods were conducted as follows: pH was determined potentiometrically in water (pH-H₂O) a 1:2.5 ratio (HI 4211, Hanna Instruments, Smithfield, RI, USA); NH₄⁺ and NO₃⁻ were analyzed by KCl extraction (2M) for 1 h, then filtered prior to determination by automated colorimetry by using an autoanalyzer (segmented flux spectrophotometer, Skalar, Breda, The Netherlands); P Olsen P was determined using 0.5 M NaHCO₃ at pH 8.5, and therefore estimated by the Murphy and Riley method and turbidimetry; K⁺, Ca²⁺, and Mg²⁺ by an extraction with 1 mol/L ammonium acetate solution at pH 7.0 and determination by spectrophotometry atomic absorption (969, Unicam, Ilminster, UK); effective cation exchange capacity (ECEC), which was determined by the sum of Ca+ Mg + K + Na+ Al; S by an extraction with dihydrogen solution calcium phosphate 0.01 mol/L and turbidimetric determination; and exchangeable aluminum (Al_EXCH) was analyzed by extraction with potassium chloride solution 1 mol/L and determination by spectrophotometry atomic absorption (969, Unicam, Ilminster, UK), whereas percent aluminum saturation (% Al_SAT) was determined by the following formula:

% Al_SAT = (Al_EXCH/ECEC)*100

(2)

where % Al_SAT is aluminum saturation, Al_EXCH is exchangeable aluminum and ECEC is effective cation exchange capacity.

The total N and SOC contents were determined by dry combustion (Leco CN-2000, macro-analyzer, LECO Corporation, Saint Joseph, MI, USA) [44] at the Soil and Natural Resources Laboratory (Faculty of Agronomy, University of Concepción).

In addition, SOC stock was estimated by

SOC stock = H × BD × SOC × 100

(3)

where H is the soil depth (cm); BD is bulk density (g cm⁻³); and SOC is total SOC content (g kg⁻¹).

2.2.3. Molecular Microbiome Analysis

Samples consisting of 10 g of soil were taken from each site at 0–5 and 5–20 cm, sealed in plastic bags, and transported within coolers for later analysis. To analyze the microbial communities in the soil, we extracted DNA from each sample. DNA extraction isolates genetic material from bacteria and fungi, allowing us to identify their diversity and functions. We used a Macherey Nagel soil DNA extraction kit (Macherey Nagel Labs, Germany) according to the manufacturer’s protocol, ensuring high-quality DNA for sequencing. The DNA was verified by electrophoresis in an agarose gel, according to Brody and Kern [45], and the concentration was evaluated using a Thermo Scientific NanoDrop™ 3300 Spectrophotometer. DNA samples were merged (from triplicate), lyophilized, and stored for further sequencing. To combine multiple primers from a single genomic DNA extraction, a Fluidigm technology technique was performed according Mallot et al. [46], in order to minimice PCR errors and cross. For that purpose, the following primers were selected: Bacteria F515 (5′-GTGYCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACNVGGGTWTCTAAT-3′); Eukarya: Euk1391f (5′-GTACACACCGCCCGTC-3′) and EukBr (5′-TGATCCTTCTGCAGGTTCACCTAC-3′); and Fungi: ITS3 (5′-GGACTACVSGGGTATCTAAT-3′)–ITS4 (5′-TCCTCCGCTTATTGATATGC-3′).

Sequencing was performed using a single 2 × 250 Illumina Hiseq2000 run. Every single sample was analyzed with the dada2 package (v1.11.3) in R (v3.4.4). To ensure high-quality sequence data, a rigorous quality control process was performed following the standard Mothur pipeline. Low-quality bases and ambiguous sequences were filtered using the screen.seqs and filter.seqs commands, applying strict thresholds to remove sequencing artifacts. Chimeric sequences were identified and removed using chimera.vsearch, ensuring only high-confidence reads were retained for downstream analysis. Paired-end reads were merged using the make.contigs command, which aligns overlapping sequences based on quality scores to generate accurate consensus sequences. This step minimizes sequencing errors and ensures the reliability of the assembled reads before taxonomic classification and community analysis. Taxonomic identification was carried out using the GTDB database (release 202) for bacteria, the PR2v5 database for Eukarya, and the UNITE ITS database for fungi (v209). For the subsequent analyses, the microeco package in the R environment was employed. Specifically, we calculated the relative abundance of ASVs in each sample and conducted a community analysis, encompassing observed composition and alpha-diversity and beta-diversity indices. Following this, functional profiles were generated using a Functional Annotation of Prokaryotic Taxa (FAPROTAX) [47] for bacterial and fungal traits for fungi [48]. This tool facilitates the prediction of the functional roles of microorganisms based on their taxonomic information and offers insights into the functional potential of the microbial community. As environmental factors and microbial functions can be generally applied to explain microbial community structure and assembly, we correlated the diversity indices and functional potential with the environmental variables using Spearman’s correlation method.

2.3. Statistical Analysis

To determine whether soil properties differed significantly between SEC_FORST and FAD_IST, we used statistical tests (‘ARTool’ package [49]) to compare their physical, chemical, and microbial characteristics. By applying non-parametric and correlation analyses, we identified patterns in how soil recovers after disturbance. In all cases, significant differences were considered with a p < 0.05 significance level. All statistical analyses and graphs were performed using RStudio Statistical Software (V4.1.0, the RCore Team 2021).

To compare microbial taxa, we used non-parametric statistical tests due to the non-normal distribution of abundance data. Specifically, taxa abundances were analyzed using the Kruskal–Wallis test, which allows for comparisons across multiple groups without assuming normality. When significant differences were found, pairwise comparisons were conducted using Dunn’s test with Bonferroni correction to control for multiple testing. This approach ensures robust statistical inference while accounting for the distributional characteristics of microbial abundance data.

3. Results and Discussion

3.1. Physical Properties

Physical results are presented in

Table 2. Results of physical characterization.

Property	FADIST 0–5	FADIST 5–20	SECFORST 0–5	SECFORST 5–20
A INFVk *	13.05 ± 0.3 A	13.05 ± 0.3 A	18.1 ± 0.66 B	18.1 ± 0.66 B
WHC **	40.42 ± 4.35 Aa	38.05 ± 4.42 Aa	39.30 ± 1.68 Aa	34.86 ± 6.07 Aa
BD ***	0.63 ± 0.02 Aa	0.59 ± 0.02 Aa	0.57 ± 0.02 Aa	0.56 ± 0.01 Aa
PD ***	1.94 ± 0.01 Aa	1.92 ± 0.02 Aa	1.95 ± 0.03 Aa	1.93 ± 0.03 Aa
B POR (%) **	67.19 ± 0.69 Aa	70.77 ± 1.18 Aa	69.59 ± 1.61 Aa	70.98 ± 0.89 Aa
WSA **	49.77 ± 0.36 Aa	49.80 ± 0.14 Aa	52.47 ± 4.77 Aa	50.83 ± 1.64 Aa
PENRES ****	350.0 ± 0.0 Aa	316.67 ± 57.7 Ab	250.0 ± 0.0 Bb	250.0 ± 0.0 Bb

^A Relative to total depth (0–20), ^B estimated through the Formula (1), *: (cm d⁻¹); **: (%); ***: (g cm⁻³); ****: (PSI). Disparity in capital letters indicates significant differences (p < 0.05) between PEN_RES and FAD_IST, while in the case of lowercase letters, it refers to differences between depths.

, showing no significant differences (p < 0.05), except for I_NFVk (SEC_FORST > FAD_IST) and PEN_RES (FAD_IST > SEC_FORST), both presumably related to greater SOC contents and the associated biological processes in SEC_FORST. The I_NFVk averaged +4.6 cm d⁻¹ in SEC_FORST condition, denoting possible differences in internal soil architecture by evolution over the time period of ~45 y since the last disruptive event.

With respect to PEN_RES, the high mechanical and animal loads in the recent past within FAD_IST have caused compaction exceeding critical levels for plant development (≥300 psi) [50], which can be evidenced in the area by bare soil patches and the presence of gullies. However, a decrease of 33.4 psi was observed in FAD_IST (0–5 cm) in relation to the 5–20 cm depth, suggesting a significant effect (p < 0.05) of bioturbation root developed to reverse compaction itself after 5 y, whilst the −100 psi difference between SEC_FORST and FAD_IST (0–5 cm) mirrors these bioactive effects.

However, a numeric positive trend was observed in the upper depth for the properties WHC, BD, and PD in both FAD_IST and SEC_FORST and P_OR and WSA in SEC_FORST and PEN_RES in FAD_IST, which is directly related to water retention–supply capacity. In relation to total depth (0–20 cm), we observed a net decrease in I_NFVk (26.52%), WSA (2.8%), PD (0.5%), and P_OR (2.4 %) (FAD_IST < SEC_FORST), contrary to BD (7.0), WHC (3.08), and PEN_RES (23.08 %) (FAD_IST > SEC_FORST).

The WSA decline in FAD_IST may also be influenced by the progressive depletion of Ca²⁺/native lack in Andisols, which is considered the most relevant inorganic binding agent (e.g., clay–polyvalent cation–SOC aggregation mechanisms) [22,51].

Both BD and PD showed typical values for volcanic soils (BD ≤ 1 g cm⁻³) [52], which is intrinsically linked to the physical high internal porosity of parent material (e.g., allophane) [31,53]. Consequently, the estimated P_OR values (Equation (1)) clearly exceeded the typical values (>40) [52], which is in compliance with the high WHC observed, globally promoting water cycling–bioavailability [54].

3.2. Chemical Properties

Chemical properties resume in

Table 3. Soil chemical properties.

		Condition
Property	FADIST 0–5	FADIST 5–20	SECFORST 0–5	SECFORST 5–20
pH	6.19 ± 0.05 Aa	6.15 ± 0.05 Aa	5.9 ± 0.06 Bb	5.86 ± 0.07 Bb
SOC *	10.10 ± 0.25 Aa	8.47 ± 0.66 Bb	11.03 ± 1.01 Cc	10.46 ± 0.96 Dd
N *	0.73 ± 0.06 Aa	0.63 ± 0.06 Aa	0.77 ± 0.06 Bb	0.77 ± 0.06 Bb
C:N	17.61 ± 1.34 Aa	16.71 ± 1.64 Aa	17.01 ± 1.67 Aa	16.39 ± 2.55 Aa
NH4+ **	5.94 ± 0.32 Aa	5.37 ± 0.66 Aa	6.11 ± 0.27 Aa	5.36 ± 0.41 Aa
NO3− **	5.4 ± 0.18 Aa	4.43 ± 0.39 Bb	3.72 ± 0.43 Bb	3.02 ± 0.28 Cc
P+ **	2.25 ± 0.29 Ab	1.91 ± 0.31 Aa	2.62 ± 0.11 Bb	1.91 ± 0.24 Aa
K **	0.34 ± 0.06 Aa	0.19 ± 0.03 Bb	0.71 ± 0.05 Cc	0.59 ± 0.02 Cc
Ca2+ **	8.06 ± 2.11 Aa	3.57 ± 0.91 Bb	3.78 ± 0.23 Bb	2.05 ± 0.61 Cc
Mg2+ **	0.84 ± 0.07 Aa	0.30 ± 0.07 Bb	0.68 ± 0.09 Aa	0.33 ± 0.02 Bb
S **	2.83 ± 0.71 Aa	5.86 ± 1.14 Bb	12.70 ± 0.44 Cc	11.97 ± 0.60 Cc
ECEC ***	5.92 ± 0.73 Aa	4.44 ± 0.68 Bb	6.40 ± 0.80 Aa	2.81 ± 0.29 Bb
AlEXCH ***	0.09 ± 0.05 Aa	0.08 ± 0.02 Aa	0.18 ± 0.03 Bb	0.22 ± 0.02 Bb
A AlSAT *	1.52 ± 0.07 Aa	1.80 ± 0.15 Bb	0.43 ± 0.60 Cc	7.83 ± 0.26 Dd

*: %; **: mg kg⁻¹; ***: cmol (+) kg⁻¹). Disparity in capital letters indicates significant differences (p < 0.05) between PEN_RES and FAD_IST, while in the case of lowercase letters, it refers to differences between depths. ^A Estimated through the Formula (2).

, showing significant variations (p < 0.05) only between SOC and Al_SAT among FAD_IST-SEC_FORST and depths comparisons. However, other significant differences were detected between the comparisons as follows: (i) FAD_IST-SEC_FORST:—pH, N, and Al_EXCH varied; (ii) similar depths—Mg²⁺ and ECEC; (iii) FAD_IST (0–5) and SEC_FORST (5–20) depths—NO₃⁻ and Ca²⁺⁺, (iv) depths in FAD_IST with respect to the SEC_FORST condition (which did not vary between depths K and S, (v) 0–5 depths of both FAD_IST and SEC_FORST—P⁺ and (vi), FAD_IST-SEC_FORST, or depths: C:N and NH₄⁺. In addition, as occurred with the physical properties, better chemical-fertility conditions were generally observed for the 0–5 cm depth, except for Al_EXCH, Al_SAT, N (only for FAD_IST), and S (only for SEC_FORST).

When comparing the main percentage variations from significantly different properties (p < 0.05) between FAD_IST and SEC_FORST (0–20 depth), the following was estimated: NO₃⁻ (+31.17% in FAD_IST), K⁺ (+62.91% in SEC_FORST), Ca²⁺ (+46.33% in FAD_IST), S (+58.03% in SEC_FORST), ECEC (+22.87% in FAD_IST), and potential toxicity with Al_EXCH (+61.91% in SEC_FORST) and Al_SAT (+83.65% in SEC_FORST).

The implications and effects of the concentrations observed are described as follows:

(i) pH values are adequate for plant development in all cases [55] and the ranges observed (5.86 to 6.1) could be attributable to the high precipitation local regime (≥3000 mm y⁻¹), native basic poor volcanic materials, and desilication processes [31], and the more acidic values in SEC_FORST are due to slight shifts in precipitation patterns (north > south) [54,56]; (ii) the high SOC concentrations may be due to the particular mechanisms of SOC preservation in volcanic soils described in Section 1 (e.g., Al-SOC complexation, extracellular enzymes sorption, occlusion of organics) [7,17,31,32,57,58]. However, specific differences in SOC contents cycling between FAD_IST and SEC_FORST can be related to the possible influence of Chusquea sp., contributing to more labile SOC and rich lignin substrates exerted by the high tree density for SEC_FORST [59,60], and the presence of pyrogenic C in FAD_IST as a remnant of site historical use, which was evidenced by charcoal fragments, which has been accounted up to 5% of total SOC in national forests [61], and these materials are also able to neutralize pH values due to their diverse functional (e.g., COO–, CaCO₃, PO₄³⁻) [53]; (iii) the N differences may be explained by the massive presence of Chusquea sp., leading to an increase in total-available N coupled with N-NO₃⁻ ratios, our results are in concordance with Pérez et al. (2019) [59], who studied a Laureliopsis philippiana (Monimiaceae) forest in Chiloé, Chile, which was logged for 10 y of having understory strata dominated by Chusquea sp. However, N uptake by Nothofagus individuals readily counteracted this effect, which could also explain the lower NO₃⁻ compared to FAD_IST, which in turn had a less dense understory. The lower numerical NH₄⁺ levels in FAD_IST indicate a greater demand by vigorous understory herbaceous strata, which uptake this bioavailable N species due to the minor uptake energetic cost compared to NO₃⁻ [62]. Regarding C:N ratios, numerical differences may be related to limiting N bioavailability and the presence of recalcitrant C and low substrate quality; (iv) the low critical P levels observed under all criteria correspond to typical volcanic soils due to the high PO₄³ fixation of at least 70% [63,64]; however, the significant differences (p < 0.05) between 0 and 5 cm depths may be related to greater nutrient cycling in SEC_FORST; and (v) the S values were under adequate levels under all criteria; however, the FAD_IST < SEC_FORST pattern observed, near to the lower fertile limit (<1 mg kg⁻¹) [65], suggests a deeper and well-established root system by both Nothofagus obliqua and Chusquea sp. in SEC_FORST and a greater uptake-mining by root systems in FAD_IST, along with a possible SO₄²-retention/leaching processes exerted by charcoal fragments having anion exchange capacity [66], and percolation by competition with NO₃⁻ for interchange sites [67].

(vi) Regarding basic cations, the critical deficiencies of K, Ca²⁺⁺, and Mg²⁺ could be related to three main mechanisms: (a) soil solution percolating due to high local precipitation regime and suitable native hydraulic–structural properties [31,54], (b) uptaking for overall metabolic processes, (c) desilication, leading to a progressive lowering on cation concentrations [17], and/or (d) higher soil organic matter content and nutrient cycling [55], as in the case of K (SEC_FORST > FAD_IST ), where, however, the higher Ca²⁺ values in FAD_IST may be related to the high specific area of the present pyrogenic materials (50–≥500 m² g⁻¹) [68]; (vii) the ECEC values ranged from moderate in all criteria, except for SEC_FORST (0–5 cm) corresponding to an adequate level (>6.27 cmol kg⁻¹) [65], which could be related to the more developed deep root systems in SEC_FORST of both Chusquea sp. and Nothofagus obliqua mining nutrients to depths greater than 5 cm, whether they can compete for nutrients or not; and (viii) the variables Al_EXCH and Al_SAT, expressing potential toxic conditions for plants, showed adequate values in FAD_IST, indicating the effect of liming practices along the total depth studied (0–20 cm), whereas moderate Al_EXCH and moderate Al_SAT levels were determined in all conditions (except for SEC_FORST 5–20 cm depth, critical). However, in all cases, Al_EXCH varied significantly between the depths of both FAD_IST and SEC_FORST, expressing a lesser capacity at lower depths to form Al–organic complexes due to the decreasing SOC content [69]. Finally, the higher absolute values observed at depth 0–5 (except for S in FAD_IST and desirably in the Al-related variables), suggest a propensity for the development/thickening of A horizon–SOC stock enlargement and the associated direct effects in terms of fertility status and ecosystem services provision.

3.3. Soil Organic C Sequestration (CO_q→SOC)

According to earlier research in areas near our study site (Dube, pers. Commun.) [34], temporal variations in C content and stocks could be estimated at total depth (0–20 cm) and the relative temporal increment or decrease rates (

Table 4. Temporal variation in SOC concentration, SOC stocks, and CO₂→SOC.

	FADIST	SECFORST
Previous SOC (%)	8.2 *	10.1 *
SOC2020	8.9	10.6
A Previous SOC stock (Mg ha−1)	81.18 *	103.53 *
ASOC stock 2020 (Mg ha−1)	106.53	120.07
Theoretical annual CO2→SOC	4.23	2.75

The data listed correspond to the weighted values for the depths 0–5 and 5–20 as follows: * for the year 2014 (source: Dube, pers. Commun.) [34], as the nearest referential records to our study conditions. ^A Estimated through the Formula (3).

). The carbon concentration and SOC stock were greater at SEC_FORST; however, larger CO₂→SOC was estimated at FAD_IST. Total variations on SOC stocks from the 6 y time lapse were 26.1 and 15.96 Mg SOC ha⁻¹ for FAD_IST and SEC_FORST, respectively.

These positive differences on SOC stocks reaching +21.3 and +16.5 Mg ha⁻¹ y⁻¹, (2014–2020 period) for FAD_IST and SEC_FORST, respectively, resulted in the CO₂→SOC potentials expressed in Table 4. In consequence, a +1.48 Mg SOC ha y⁻¹ was estimated for FAD_IST, compared to SEC_FORST, despite similar SOC % concentrations. This can be explained by the following: In SEC_FORST, the particular role of Chusquea sp. on soil fertility is of major relevance, since its chemical characteristics result in up to 67.2% of alpha cellulose + hemicellulose and only 13.7% of lignin content, reaching a height of 2.0 m and a density of 290 kg m³ [70], implying a potential massive input of labile SOC (glucose-based), facilitating its rapid mineralization, not only by their chemical nature per se, but also by their effect in reactivating nutrient mining processes by decomposers of the anterior organic matter (priming effect), resulting in limited CO₂→SOC rates [20].

In the particular scenario of FAD_IST, and despite the significantly lower tree density, this condition showed higher CO₂→SOC capacity, which can be related to the following factors: (i) C inputs derived from the debris of emergent vegetation (understory and herbaceous strata), (ii) presumably higher SOC–metal (Fe³⁺, Al³⁺) complexation potentials than SEC_FORST, (iii) an appreciable enrichment of energy-rich reduced SOC residues (aromatics) derived from previous ignition events and practices (wildfires and agricultural slash-and-burn) resulting in the most stable recalcitrant SOC pool (centennial–millennial mean residence time), and (iv) because it is located in an area of accumulation (Section 2.1).

These findings are comparable with data reported by Ortiz et al. [71], who estimated annual SOC increases of 7.5, 4.8, and 1.6 Mg ha⁻¹ y⁻¹ in nearby silvopastoral systems based on Nothofagus obliqua to our study sites at 3 different tree densities of 134, 60, and 258 stems ha⁻¹, respectively. Similarly, Ortiz et al. [72] reported CO₂→SOC of 5.50 and 5.48 Mg C y⁻¹ in two conterminous agroforestry systems corresponding to silvopasture under N. obliqua and N. dombeyi of a tree density of 173.3 stems ha⁻¹ (133.3:40, respectively) and an agroforest growing oats (Avena sativa)-vetch (Vicia atropurpurea) associations below N. obliqua at a tree density of 446.6 stems ha⁻¹. The authors concluded that these systems promote nutrient cycling, SOC accumulation, and a progressive desirable increase/decrease in physical and chemical properties, except those that are closely related to the intrinsic characteristics of volcanic soils (for example, P) and at the ecosystem level (e.g., total and bioavailable N). In addition, Muñoz et al. [73] estimated SOC increases of 33.1 to 35.5 Mg ha⁻¹ in a monoculture agro-system under no-tillage after 16 y in the same soil series as our study sites. In different managed forests (e.g., plantations) developing over Ultisols from Sothern China, Li et al. [74] did not find variations in SOC associated with minerals, but critical changes in SOC related to stable aggregates. The above suggests that well-planned and operational productive management may be arguably better alternatives for land reclamation than the mere cessation of human activity in degraded native forests of south-central Chile, not only as a better CO₂→SOC option, but in order to control the occurrence and dominance of opportunistic species that represent a potential threat, such as (i) underbrush for farming proposes, (ii) a habitat of Oligoryzomys longicaudatus, a vector of the hantavirus Bunyaviridae of major medical importance, (iii) a causal agent of wildfires during summer seasons as considerable dry biomass aboveground stores, and (iv) impeding the re-consolidation of the previous understory and therefore of the original ecosystem biocomplexity and/or long-term sustainable management of forests such as agroforestry (for example, [35,75,76,77,78]) and planted forests [79].

3.4. Interactions Among the Analyzed Soil Properties

A total of 93 relations (r ≥ 0.80) were observed among multiple variables (including inverse), of which 37, 11, and 45 occurred between chemical (4 inverse), physical (5 inverse), and 45 involving physicochemical properties (12 inverse), respectively (Figure 3). pH was the most influential property correlated with the other nine, explaining, for instance, their inverse relationship with Al_EXCH and Al_SAT (r = 98 and r = 89, respectively). SOC was correlated with N availability (r = 98) and exerted an influence on other nutrients (for example, P⁺, K, Mg²⁺); however, it did not meet the r ≥ 0.80 criteria.

The C:N ratio clearly influenced both N occurrence and availability (r = 97, r = 75, and r = 86) for total N, NH₄⁺, and NO₃⁻, respectively, as well as the nutrient parameters Ca²⁺, Mg²⁺, and ECEC. Interactions between physical properties, including I_NFVk, were well correlated with P_OR and WSA (r = 98 and 85), respectively, and inversely correlated with BD (r = −98) and PEN_RES (r = −96); P_OR was correlated with WSA and PEN_RES (r = 98 and r = 85), and WSA had an inverse correlation with PEN_RES (r = −82). With respect to physicochemical interactions, WHC was well correlated with the C:N ratio (r = 92) and nutrient levels (r = 83, r = 81, r = 95) for Ca²⁺, Mg²⁺, and ECEC, respectively.

Our physical results are consistent with those reported by Ortiz et al. [71], who reported in conterminous plots under silvopastoral management correlations of (r ≥ 0.8) for %WSA, POR, and INFV, indicating suitable hydraulic conditions in these soils. In relation to chemical properties, our results are contrary to those reported by Ortiz et al. [71] and Ortiz et al. [72] (both under agroforestry), where SOC was the core property, which correlated to nutrient availability; in our case, the presumably higher pH values tend to drive such a role.

3.5. Soil Microbial Communities

The metabarcoding results showed that changes in soil microbial communities in degraded (FAD) and non-degraded (SEC) soils at two depths (20 and 5 cm) revealed distinct patterns in the relative abundance and diversity of microbial classes (Figure 4), affecting both prokaryotic and eukaryotic microorganisms. However, the eukaryotic 18S rRNA sequences obtained from the study fields were predominantly composed of fungi. Therefore, we focused on the analysis of bacterial communities based on 16S rRNA sequences, and fungal communities based on ITS-specific sequences.

For bacterial communities, the data showed that Alphaproteobacteria and Gammaproteobacteria were dominant in bacterial communities across all samples, with notable variations between depths and soil types. In deeper soils (20 cm), there was a higher relative abundance of Acidobacteria, especially in non-degraded soils (SEC). Conversely, degraded soils (FAD) exhibited a higher presence of Bacilli and Thermoleophilia, indicating a potential adaptation to altered soil conditions due to degradation. Several authors have reported that the presence of Bacilli, known for their ability to form endospores, can help them survive harsh conditions, including soil degradation [80,81]. Thermoleophilia, a class of bacteria within the phylum Actinobacteria, is typically found in hot and nutrient-poor environments, indicating its adaptability to stress conditions [82]. The bar plots (Figure 4A) showed that the dominant class in both SEC_FORST and FAD_IST was Alphaproteobacteria, followed by Gammaproteobacteria, which was more represented in SEC_Frost 5–20. Thermoleophilia was particularly abundant at 0–5 m in both FAD_IST and SEC_FORST (13.6 and 14.2%, respectively), while Acidobacterieae was more abundant at depth, representing 7.7% of the community at FAD_ist and 15.7% at SEC_Frost, a similar distribution pattern was observed for Dormibacteria. Viciniamibacteria was more abundant at the FAD_ist site at both depths.

When analyzing the bacterial community structure, we observed that the representation of Dormibacteriaceae and Steroidobacteriaceae negatively correlates with the Mg²⁺ concentration, while other families as Solirubrobacteriaceae strongly correlated with PD, P, and C:N. Pyrimonadaceae was positively correlated with pH but negatively correlated with total S, Al, INFV, and K. Finally, UBA2999 negatively correlates with total N (Figure A1).

The bacterial community composition displayed a similar representation as those reported by Navarrete et al. [21], who observed within the bacterial community a prevailing occurrence of Proteobacteria (45.35 ± 0.89%), Acidobacteria (20.73 ± 1.48%), Actinobacteria (12.59 ± 0.34%), and Bacteroidetes (7.32 ± 0.36%). Among Proteobacteria, the most abundant Alphaproteobacteria corresponded to Bradyrhizobium. Bacteria of this genus penetrate the roots of many legume species, producing root nodules in which N fixation occurs. Symbiotic N fixation is most important in forestry, where trees are closely associated with many wild, herbaceous legumes [83]. It has also been demonstrated that soil pH and soil C:N ratios significantly vary with soil microbial composition [84]. In our work, Pyrinomonadaceae was strongly correlated with pH, while Solirubrobacteraceae was strongly correlated with C:N, as well as with total P. Both pH and C:N were heavily influenced by symbiotic N₂ fixation. Recent research has revealed a link between the phosphate solubilization capacity and increased potential for polysaccharide hydrolysis and carbohydrate metabolism. This unique link strongly corresponds to the positive relationship between the population density of PSB and available dissolved organic carbon in the soil [85]. In our work, Pyrinomonadaceae strongly correlated with pH, while Solirubrobacteraceae strongly correlated with C:N, but also with total P. Both the pH and the C:N are heavily influenced by symbiotic N2 fixation. Recent research has revealed a link between the phosphate solubilization capacity and increased potential for polysaccharide hydrolysis and carbohydrate metabolism (Figure A1). This unique link corresponds strongly to the positive relationship between the population density of PSB and available dissolved SOC [85].

For fungal communities, Figure 4B displays a significant shift in composition between depths and soil types. The phyla Ascomycota and Basidiomycota were dominant in both FAD_IST and SEC_FORST. The topsoil layer (0–5 cm) was dominated by Ascomycota (49% and 50% of the fungal community) and Basidiomycota (50–49% of the fungal community). In the second soil layer (5–20 cm), both FAD_IST and SEC_FORST were also dominated by Ascomycota, accounting for 54% and 74% of the fungal community, respectively, along with Basidiomycota (45% and 25% of the fungal community, respectively). In non-degraded soils, classes such as Agaricomycetes and Leoiomycetes were more prevalent, suggesting a rich saprotrophic community. In degraded soils, there was a marked increase in Dothideomycetes in the second soil layer (5–20 cm), which could be associated with altered organic matter composition and nutrient availability. Interestingly, the soils of the second layer exhibited a lower diversity of fungi than the non-degraded soils. This could indicate the influence of agronomic management practices and the negative effects of degradation on soil nutrient cycling [86,87,88].

The functional analysis using FAPROTAX (Figure 5) highlights key in silico processes occurring in these soils. Soil depth and degradation status significantly influence the structure and functional potential of microbial communities. Non-degraded soils, particularly at greater depths, support more diverse and functionally rich microbial communities, which are crucial for maintaining soil health and ecosystem stability. Degraded soils, on the other hand, exhibit a shift towards less diverse and more pathogenic microbial communities, along with a potential reduction in key nutrient cycling processes. This study underscores the importance of soil management practices in preserving soil microbial diversity and functionality, which are essential for sustainable ecosystem services. On the other hand, non-degraded soils exhibit a higher diversity of saprotrophic fungi, which play a crucial role in decomposing complex organic matter. This is consistent with the higher organic matter content typically found in non-degraded soils. However, degraded soils have a higher representation of fungal pathogens and mycorrhizal fungi, possibly reflecting a stressed environment in which plants are more susceptible to diseases and rely heavily on symbiotic relationships for nutrient acquisition.

We also observed a strong partitioning of the functions related to the soil properties (Figure 5C,D, Appendix A). The sulfur cycle was strongly represented when PEN_RES, BD, pH, and NO₃⁻ were higher. In addition, N-related processes, such as denitrification and aerobic ammonia oxidation, were more represented when the concentrations of Ca, Mg, and WHC were higher. On the other hand, methanol oxidation, metanothrophy, ureolysis, and photoheterotrophy were more represented when available K, K total, and SOC were higher.

In bacterial communities, deeper, non-degraded soils showed enhanced capacities for nitrogen fixation and sulfur cycle processes, such as sulfite respiration and sulfate reduction. This indicates a robust biogeochemical cycling capacity, possibly supporting a more stable ecosystem [89]. Degraded soils, particularly at 5 cm depth, show increased potential for fermentation and aerobic chemoheterotrophy, suggesting a shift towards more anaerobic and less efficient energy-yielding processes, likely due to soil compaction and reduced oxygen availability. Fungal functional profiles (Figure 5D) also differ significantly between soil types and depths. The aerobic chemoheterothrophy was the most represented energy source. The N fixation appears to be more represented at the FAD_IST, as well as the dark hydrogen oxidation. Other metabolisms represented corresponded to metanothrophy sulfite respiration and sulfate oxidation, iron respiration, and aerobic ammonia oxidation.

Microbial communities play a fundamental role in soil health, influencing nutrient cycling, organic matter decomposition, and overall ecosystem resilience [2,90]. The observed differences in bacterial and fungal composition between FAD_IST and SEC_FORST soils suggest that post-disturbance microbial dynamics may drive long-term soil recovery [91]. The dominance of Alphaproteobacteria and Gammaproteobacteria, known for their roles in organic matter decomposition and nitrogen cycling, indicates potential microbial contributions to nutrient turnover [92,93]. Conversely, the higher presence of Acidobacteria in deeper soils suggests microbial adaptation to resource-limited conditions, which may affect carbon stabilization processes over time [94].

Additionally, the shifts in fungal communities, particularly the balance between Ascomycota and Basidiomycota, highlight potential changes in decomposition pathways that could influence soil organic matter persistence [95]. These microbial trends align with broader ecological resilience frameworks, where community stability and functional redundancy are key indicators of a system’s ability to recover from disturbance [96]. The functional analysis further supports this perspective, as degraded soils showed increased microbial metabolic activity related to stress adaptation, while non-degraded soils maintained a more diverse functional profile, reinforcing their potential for long-term stability [97]. Understanding these microbial shifts is critical for assessing the trajectory of soil health in post-disturbance ecosystems. Future studies should further explore the long-term implications of these microbial dynamics in soil carbon storage and nutrient cycling, providing a more comprehensive view of their ecological impact [98].

Further research is necessary in terms of similar tests on different biomes-passive management, temporal variations in tested properties in other secondary Nothofagus sp. forests, as well as greater depths of present studies sites in order to elucidate the effects of both systems on the soil profile, as well as the variations in their physical, chemical, and biological properties, taking into account that this study seeks to be an approximation to these conditions widely distributed at national level.

4. Conclusions

Our study highlights that post-disturbance passive management of native Nothofagus forests in south-central Chile had similar effects on nutrient status, physical aptitudes for water cycling–plant utilization, a decline of soil carbon sequestration rates (1.48 Mg SOC ha⁻¹ y⁻¹), and shifts only on abundance of soil microbiome for a ~45 y period. Such edaphic proximity could be attributed to three concurrent factors, the dominance of Chusquea sp., which is a widespread regional opportunistic genre, able to inhibit understory biocomplexity, transitioning SEC_FORST into to a more dynamic SOC system, the SOC and nutritional residual effects of pyrogenic materias and recent agricultural activities in FAD_IST and the overall influence of disctintive mineral fraction of volcanic soils on nutrient availability-limitations and hydraulic properties. In addition, these soils exhibit remarkable resilience traits, since a generalized upper desirable level (at 0–5 cm depth) of most variables was analyzed, despite being statistically significant (p < 0.05), a trend also observed in the SEC_FORST > FAD_IST comparison. Consequently, an a priori scarification processes that promote ecosystem recovery (>25% of surface area) is highly recommended, because, although Chusquea sp. promotes soil protection (e.g., soil cover, pedogenesis), it impedes ecosystem recovery by setting severe limitations to original regeneration patterns. In the case of productive management, agroforestry techniques are very suitable, since in a direct comparison between natural regeneration vs. agroforestry systems established under similar conditions, the latter also has great social and ecological advantages, potentially serving as ecotones and/or major hillside production systems. Knowledge gaps involve seasonal and forest evolution in relation to robust metagenomics and belowground carbon dynamics studies (including pyrogenic) coupled with Chusquea sp. occurrence in order to discriminate the effective C stabilization (CO₂→SOC).