Recovery of Soil-Based Ecosystem Services in Abandoned Ski Resorts: The Valcanale Case Study (Bergamo, Italian Alps)

Abstract

Climate change and declining economic revenues are driving the closure of many ski resorts in mountainous regions worldwide, particularly at lower elevations, where winter snow cover is becoming increasingly sporadic. This abandonment is impacting wide areas of the Alps, previously managed to reduce erosion and to control trees/shrubs encroachment. As result, natural rewilding processes may lead either to the environmental degradation or to the restoration of pre-disturbance conditions, each with different implications for sustainability. Our aim was to assess the rewilding state and the drivers of sustainability at an abandoned ski resort in the Italian Alps (Valcanale, Bergamo), where the ecosystem has been evolving under minimal human pressure since the ski facilities were decommissioned in 1993. The assessment focused on pedological/vegetational perspectives, with particular attention to soil-based ecosystem services (SBESs). The results show that the interventions made during ski run construction significantly influenced the recovery of SBESs (and thus their long-term sustainability). Areas with minimal disturbance (e.g., forest vegetation removal without soil movement) now support SBESs at levels comparable to nearby undisturbed areas. Conversely, ski runs that underwent slope reshaping/grading support poorly developed soils and significant sheet/gully erosion, rendering them hazardous for pedestrians. Nevertheless, plant biodiversity has benefited in some cases, as many rare/endemic protected species colonize stony/eroded ski runs soils, extending their distribution beyond their original habitat.

1. Introduction

Soil is essential for vegetation growth and supports terrestrial ecosystems. It forms a thin layer, susceptible to disturbance from both human activities and natural events (such as fires, storms, floods, earthquakes, and landslides). Soil formation is a slow process, often taken thousands of years, and is the result of pedogenic processes influenced by geological, climatic, biological, and anthropogenic factors [1]. These processes gradually transform parent rock material into fertile soil through a complex interplay of physical, chemical, and biological interactions [2]. In mountain environments, soil formation is even slower than in lowland areas, due to harsher conditions [3] including low temperatures, freeze-thaw cycles, snow accumulation, snow gliding, and intense rainfall, all of which contribute to erosion and nutrient leaching [4]. Human interventions, especially the construction of ski resorts, can significantly disrupt the delicate balance of mountain soils, leading to soil disturbance, habitat loss, and broader climatic impacts. During ski run construction, deforestation removes the protective tree cover, reducing water interception and increasing surface runoff. The construction of service roads further accelerates erosion by compacting surfaces, which limits water infiltration and contributes to sediment transport. Machine grading, used to reshape slopes, strips away topsoil and vegetation, disrupting soil structure, depleting seed banks, and altering the hydrological cycle. These processes leaves behind compacted surfaces that are poor in carbon and nutrients and highly vulnerable to erosion [5,6]. Artificial snowmaking, although it does not physically alter the terrain, affects local hydrology by varying snowmelt patterns, modifying water availability, and altering the temperature regime of the underlying soil [7]. Additionally, the chemicals sometimes added to improve snow quality can pollute the environment and permanently disturb microbial and invertebrate communities in the soil [8]. Altogether, these changes lead to substantial loss of soil-based ecosystem services (SBESs) [9,10]. The impacts are often long-lasting, posing significant challenges to ecosystem restoration on former ski runs [11,12,13]. The extent of these alterations depends on construction techniques, ski run preparation, maintenance, and specific local environmental conditions [14,15,16,17,18,19].

The definition and assessment of ecosystem services in mountain areas have been widely discussed in the scientific literature, emphasizing their critical role in supporting human well-being [20] and contributing to climate change mitigation [21]. The management of ski slopes can have both negative [22] and positive impacts on ecosystem services (e.g., [12,23]), making it essential to understand how these services evolve over time [24]. Several studies examined the effects of land cover changes on ecosystem services in mountainous regions [25]. In this context, soil properties play a key role in ecosystem functioning and the trade-offs between provisioning and regulating services [26]. Therefore, evaluating soil functions and their associated ecosystem services is an essential step towards understanding and mitigating negative impacts.

Due to climate change and social changes, the ski market is substantially stable, despite the number of skiers slowly decreasing worldwide [27,28,29]. Additionally, abandonment of ski resorts has been increasing since the late 1990s [30]. Many small ski resorts located below 1600–1800 m a.s.l. have been abandoned across Europe in recent decades, largely because of reduced snow cover [31]. Looking ahead, global warming poses a serious threat to the viability of many European ski resorts [32]. If global temperatures rise by 2 °C, 53% of the 2234 European resorts studied (from Iceland to Turkey) could experience poor snow conditions. With a 4 °C increase, nearly 75% of them may face snow scarcity every other year, even with the use of artificial snowmaking, which itself contributes to climate change [32].

In the absence of proper management, abandoned ski runs follow different ecological trajectories: they can either be colonized by pre-existing vegetation, moving toward the restoration of original conditions, or undergo land degradation due to erosion (e.g., [33]). Recolonization by forest species can have positive effects, such as increasing biomass and carbon storage in forest vegetation and improving soil hydrological properties. However, this process may also lead to a reduction in grassland or habitats for other rare species [34]. Abandoned ski resorts therefore offer a valuable opportunity to study the effects of human activities, and their sudden cessation, on mountain ecosystems, as well as the process of autonomous recovery, viewed through the lens of ecosystem services [24].

The primary objective of our study is to assess ecological recovery on abandoned ski runs, in terms of soil properties, vegetation, carbon stocks, and erosion processes, using nearby undisturbed sites as control. By analyzing both ski runs and their surrounding areas, the present study aims to evaluate how the ecosystem has evolved over 30 years of abandonment, taking into account the initial disturbance caused by ski run construction.

2. Material and Methods

2.1. Study Site

This study was conducted in Valcanale, a hamlet within the municipality of Ardesio (BG), located on the northern slopes of the Mount Arera (Orobic Alps, South-Eastern Alps, Figure 1). This area, formerly a ski resort, open between 1972 and 1993, included access roads, accommodations, parking lots, ski slopes, and lifts. These infrastructures significantly altered the landscape and affected local ecology and hydrogeology through activities such as clear-cutting, excavation, and terrain leveling. This study site is a representative example of ski resort abandonment throughout the Italian Alps, primarily due to climate change reducing snow cover duration and evolving social trends that led to declining interest in smaller resorts.

Located on the right side of the Acqualina stream valley, Valcanale exhibits a “dolomitic” landscape shaped by Quaternary glaciers, wind-driven deposits outside glaciated areas, and intense water erosion. The valley’s geology comprises “Permo-Mesozoic Cover”, with red Verrucano Lombardo conglomerates on the northern side and limestones and dolostones belonging to the Esino Limestone formation and other carbonate lithologies on the southern side. The area’s complex geological structure, including thrust faults and other tectonic features, disrupts the continuity of the stratigraphic sequences and influences hydrogeology by directing underground water flows along the fractures [33,35].

Valcanale experiences intense rainfall due to its proximity to the southernmost fringe of the Alps, which occasionally cause floods. Meteorological data from the Environmental Protection Agency of Lombardy (ARPA-Lombardia; https://idro.arpalombardia.it/en/#/it/, accessed on 19 May 2025) recorded by a rain gauge in the village of Valcanale (2004–2023) show an average annual precipitation of ca. 2400 mm. Extreme rainfall events, often exceeding 200 mm in a single day, are associated with erosion processes and debris flows. The elevation of the former ski resort and of the sampling sites ranges between approximately 1200 m and 1700 m a.s.l. Average temperatures vary around −3 °C in January and +17 °C in July, decreasing with increasing elevation.

In the montane belt, the potential natural vegetation is primarily composed of beech forests (Fagus sylvatica L.). With increasing elevation, coniferous species such as Norway spruce (Picea abies L.), silver fir (Abies alba L.), and European larch (Larix decidua Mill.) become more dominant [36]. In the study area, forest communities interface with anthropogenic grasslands (meadows and pastures), as well as habitats on limestone cliffs and screes. These latter environments in the Orobic Alps have a high number of endemic species of the southern Alps, including Galium montis-arerae Merxm. et Ehrend., Campanula raineri Perp., Xerolekia speciosissima (L.) Anderb., Aquilegia confusa Rota, Linaria tonzigii Lona, and Allium insubricum Boiss. et Reut. [37,38,39]. Common soil types in the Eastern Italian Alps on carbonate bedrock include Leptosols, Phaeozems, Cambisols, and Luvisols (Regional Agency for Services to Agriculture and Forestry-ERSAF Lombardia: https://www.ersaf.lombardia.it/territorio/i-suoli-della-lombardia/, accessed on 19 May 2025; classified according to IUSS Working Group WRB [40]).

The natural vegetation was significantly altered during operation of the ski resort. In particular, construction of the ski resort, such as deforestation, excavation, and terrain levelling, caused substantial environmental changes, leading to long-lasting instability and degradation. Many affected areas have not undergone spontaneous re-naturalization [33]. Since the ski resort’s closure in 1993, the lack of maintenance has further worsened these conditions, increasing the risk of hazards such as erosion, avalanches, and rockfalls.

Moreover, this area is part of the “Orobie Bergamasche” Regional Park, making identification of re-naturalization pathways and assessment of habitat degradation levels particularly important.

2.2. Field and Laboratory Methods

Field measurements and sampling were conducted between July and August 2022. Soil profiles were excavated along five ski runs (“Scala”, “Canalino”, “Vallone”, “Erika”, and “Collino”; Figure 1;

Table 1. Location, main environmental factors, soil [40], and humus form classification [41] of the study sites.

Profile	P01	P02	P03	P04	P05	P06	P07	P08	P09	P10	P11	P12	P13	P14	P15	P16
Location	Canalino	Canalino	Canalino	Collino	Vallone	Vallone	Vallone	Muro/Vallone	Muro/Vallone	Muro/Vallone/Erika	Erika	Erika	Erika	Scala	Scala	Scala
Elevation (m a.s.l.)	1550	1400	1400	1520	1500	1550	1550	1670	1670	1650	1580	1580	1530	1330	1400	1400
Aspect (°)	262	30	315	105	60	80	120	100	160	50	330	330	345	0	40	48
Steepness (°)	26	12	20	22	15	23	33	27	16	27	25	20	30	43	16	16
Land cover	Scree	Forest renovation	Beech forest	Pasture	Pasture	Eroded area	Grassland, heath, beech trees	Beech forest	Grassland	Heath	Eroded area	Beech forest	Spruce forest	Eroded area	Spruce forest	Grassland
Tree cover (%)	5	40	80	0	0	5	50	85	0	50	2	70	85	5	80
Herbaceous cover (%)	10	80	40	95	99	95	50	15	100	50	20	80		85
Stoniness (%)	90	20	15	5	10	80	10	20	50	15	70	20		90
Parental material	Limestone and dolostone	Limestone and dolostone	Limestone and dolostone	Limestone	Dolostone and dolomitic limestone	Dolomitic limestone with quartz	Limestone and dolostone	Dolostone	Dolostone and limestone	Limestone and dolostone	Limestone and dolostone	Limestone and dolostone	Limestone and dolostone	Dolostone	Limestone and dolostone	Limestone and dolostone
WRB	Skeletic Dolomitic Regosol (Ochric)	Skeletic Relocatic Dolomitic Regosol (Ochric)	Dolomitic Rendzic Phaeozem (Arenic)	Haplic Luvisol (Siltic)	Calcaric Skeletic Leptic Cambisol (Loamic)	Skeletic Rendzic Leptosol	Haplic Luvisol (Loamic)	Haplic Luvisol (Loamic)	Skeletic Rendzic Leptosol	Eutric Cambisol (Loamic)	Dolomitic Skeletic Leptosol (Ochric)	Skeletic Eutric Cambisol (Loamic)	Eutric Cambisol (Loamic)	Calcaric Transportic Hyperskeletic Regosol (Ochric)	Calcaric Transportic Hyperskeletic Regosol (Ochric)	Calcaric Hyperskeletic Leptosol (Ochric)
Humus form	Eumull	Hemimoder	Amphi	Eumull	Hemimoder	Eumull	Amphi	Amphi	Hemimoder	Amphi	Eumull	Amphi	Amphi	Eumull	Hemimoder	Hemimoder

), as well as in nearby natural plots without visible signs of human disturbance, to observe impacts on soil and vegetation by ski run construction and their subsequent abandonment. Some ski runs underwent significant artificial earth movements, while others, such as the “Collino” ski run (P04, Figure 1), remained largely unaltered. The soil profiles were described and classified according to IUSS Working Group WRB [40], and humus types were identified following the classification proposed by Zanella et al. (2018) [41].

Floristic surveys were carried out on 10 m x 10 m surfaces adjacent to the soil profiles. Plant species, chorological type, and state of protection were identified [42,43,44]. The number of species/taxa was used as proxy for plant diversity, while ecological indicator values developed by Landolt (L: light needs; F: soil moisture; N: nutrients in the soil; H: humus content) were used to characterize the main environmental conditions of each site [45]. The scientific names of the species follow the taxonomy of Pignatti et al. (2017) [42].

Soil samples were collected from all genetic horizons, air dried in the laboratory, sieved at 2 mm, and analyzed following standard procedures [46]. Organic horizons were sampled using a 20 cm × 20 cm square frame down to the boundary with the A horizon (with a sampled area equal to 18 cm × 18 cm), then dried and weighed before grinding and analysis. Soil pH was measured in a 1:2.5 soil-to-water suspension. Total carbon (C) and nitrogen (N) concentrations were measured by dry combustion using a Flash EA 1112 NC-Soil elemental analyzer (Thermo Fisher Scientific CN, Pittsburgh, PA, USA). Carbonate content was measured through volumetric analysis of the carbon dioxide liberated by a 6 M HCl solution. Organic carbon (OC) was then calculated as the difference between total C measured by dry combustion and carbonate-C. Particle size distribution was measured using the pipette method.

Carbon stock (Cstock, in kg m⁻²) was calculated for entire soil profiles using the following equation:

$$Cstock=\sum _{n=0}^i\left(\mathrm{TOC}\left[\mathrm{\%}\right]·0.01·\mathrm{B}\mathrm{D}·\mathrm{T}\mathrm{H}·\mathrm{V}\mathrm{F}\right)$$

(1)

where i is the number of pedogenic horizons in each profile, TOC is the soil organic carbon concentration (in %) of the mineral horizons, BD is the bulk density (in kg m⁻³), TH is the horizon thickness (in m), and VF is the volume of fine earth (in m³) excluding the coarse mineral fraction (>2 mm, visually estimated in the field for each horizons). BD was either determined from undisturbed samples taken using 100 cm³ cores (where feasible) or estimated using a pedotransfer function based on 671 samples from natural or seminatural soils in the Alps (R² = 0.667, D’Amico, unpublished):

$$\mathrm{B}\mathrm{D}=-0.238\times ln\left(\mathrm{TOC}\right[g·k{g}^{-1}\left]\right)+1.667$$

(2)

In addition to Cstock for the entire soil profile, Cstock was also calculated for the top 30 cm of mineral soils (Cstock30 in kg m⁻²) and for the organic horizons (CstockO in kg m⁻²), according to the formula

$$CstockO=\mathrm{TOC}\left[\mathrm{\%}\right]·{\displaystyle \frac{M_s}{A_s}}·0.01$$

(3)

where $M_s$ is the mass of the sample (in kg), whereas $A_s$ is the sampled area (in m²).

To calculate the soil erodibility we used the equation proposed by Wischmeier and Smith (1978) [47] and commonly used in the scientific literature:

$$K=\left\{{\displaystyle \frac{\left[2.1·{10}^{-4}·M^{1.14}·\left(12-OM\right)+3.25·\left(s-2\right)+2.5·\left(p-3\right)\right]}{100}}\right\}·0.1317$$

(4)

where

• M (dimensionless) is the textural factor, as follows:

$$M=\left(m_{silt}+m_{vfs}\right)·\left(100-m_c\right);$$

(5)

• $m_{silt}$ is the silt fraction content (0.002–0.05 mm) in %;
• $m_{vfs}$ is the very fine sand fraction content (0.05–0.1 mm) in %;
• $m_c$ is the clay fraction content <0.002 mm) in %;
• OM is the organic matter content in %;
• s is the soil structure class (s = 1 corresponds to very fine granular soils, s = 2 to fine granular soils, s = 3 to medium or coarse granular soils, and s = 4 to blocky, platy, or massive soils);
• p is the permeability class (p = 1 for very fast permeability, >61 mm h⁻¹, p = 2 for moderately fast permeability, 20.3–61.0 mm h⁻¹, p = 3 for moderate permeability, 5.1–20.3 mm h⁻¹, p = 4 for moderately low permeability, 2.0–5.1 mm h⁻¹, p = 5 for slow permeability, 1.0–2.0 mm h⁻¹, p = 6 for very slow permeability, <1.0 mm h⁻¹).

2.3. Soil-Based Ecosystem Service Evaluation

Assessment of soil-based ecosystem services (SBESs) at the former Valcanale ski resort was conducted using a semi-quantitative approach aimed at providing a preliminary overview of the state of ecological recovery (or re-naturalization). Given the exploratory nature of the present study, the analysis relied on the environmental characteristics of the area and was supported by evidence from the scientific literature on SBESs in mountain contexts [20,24]. The selected services were chosen for their ecological relevance in alpine environments and for their sensitivity to land use changes following the abandonment of ski-related activities and their importance to human well-being at both local and global scales [48,49].

SBESs considered in the analysis include production of agricultural and forest biomass (provisioning services); water retention, soil erosion control, and local and global climate regulation (regulation services); water filtration and purification, biodiversity support, and nutrient cycling regulation (supporting services); and recreational and spiritual services and conservation of cultural and natural heritage (cultural services).

This selection reflects the unique features of remote mountain soils, where functions such as water retention, nutrient cycling, and habitat provision are essential for ecological stability [10,26,50]. As a result, some ecosystem services such as the transformation and retention of organic contaminants and groundwater recharge were excluded from the evaluation. In particular, the important service of groundwater recharge was excluded, because all soils were developed on hard, almost impermeable, bedrock. Given the remoteness of the area and the absence of important sources of pollution nearby, we did not consider retention and transformation of organic or inorganic contaminants.

The assessment was based on integrated analysis of the site’s physical and biological characteristics (see Section 2.1), interpretation of soil and vegetation data collected during fieldwork, and comparison with findings from other studies on ecosystem services in mountain environments.

Each SBES was evaluated on a 0 to 1 scale, where higher scores indicate better service provision, based on specific measured or observed indicators (

Table 2. Soil-based ecosystem services (SBESs) and the input data used for their calculation.

Code	SBES	Function	Input Data
ABP	Agricultural Biomass Production	Capability of soils to support production of high-quality forage	Soil depth (cm); fine earth (%, 100—stoniness); N content (%); C/N ratio; humus form; pasture quality (plant species composition)
FBP	Forest Biomass Production	Capability of soils to support highly productive forests	Soil depth (cm); fine earth (%, 100—stoniness); N content (%); C/N ratio; humus form; forest quality (tree height, density, qualitative)
WRE	Water Retention	Capability of soils to absorb and retain rainwater	Soil depth (cm); fine earth (%, 100—stoniness); clay content (%); structural development; bulk density (g cm−3); porosity (% volume)
WFP	Water Filtration and Purification	Capability of soils to filter and purify water thanks to exchange properties and pore dimension	Soil depth (cm); fine earth (%, 100—stoniness); clay content (%); organic matter content (%); structural development; bulk density (g cm−3); porosity (% volume)
LCR	Local Climate Regulation	Related to capability to retain sufficient water to allow evapotranspiration	Soil depth (cm); fine earth (%, 100—stoniness); clay content (%); organic matter content (%); structural development; bulk density (g cm−3); porosity (% volume)
SEC	Soil Erosion Control	Soil resistance to erosion	Soil erodibility (K in the RUSLE Model [47]; t ha h ha−1 MJ−1 mm−1); erosion evidence in the field
GCR	Global Climate Regulation	Capacity of soils to store carbon and support productive ecosystems	Cstock (kg m−2); Cstock in O horizons (kg m−2); forest biomass (not measured, qualitative)
HBI	Habitat and Biodiversity	Capacity of soils to support diverse ecosystems, rare plant species, high soil biodiversity	Total number of plant species; number of protected, endemic, and stenoendemic species; humus form
CAN	Cultural and Natural Archive	Natural and historical legacies in soils	Endemic plant species; specific soil horizons indicating past environmental conditions.
RSS	Recreation and Spiritual Services	Capacity of soils to support/create attractive environments	Landforms with geoheritage value [53]; high-quality ecosystems; stenoendemic species
NCR	Nutrient Cycle Regulation	Dynamics of organic matter decomposition influencing nutrient availability	C/N ratio; humus form; presence of earthworms; Landolt N indicator (reduced according to fine earth content)

). In addition to measured soil physical and chemical properties, several qualitative indicators were assessed using an expert-based approach. For instance, the presence of buried charcoal horizons in the soil was considered to enhance the Cultural and Natural Archive (CAN) service, as these horizons often indicate historical forest clearings for pasture creation—an established practice in mountainous regions [51]. Additionally, deep cryoturbated horizons were assigned high significance, as they represent legacies of cold periods during the Late Quaternary [52]. Recreation and spiritual services (RSSs) were mainly evaluated based on the geomorphological features and landscapes of geoheritage value [53], as well as high-quality ecosystems that attract hikers (e.g., a gorge with distinctive karst formations).

Each property was normalized (min-max) for the studied sites, following the approach suggested by Calzolari et al. (2016) [54]. Although the assessment was only semi-quantitative, it was grounded in three key principles from the scientific literature: (i) recognition of interconnections among SBESs, including potential synergies and trade-offs (e.g., increased forest biomass may enhance carbon sequestration but reduce local water availability [50,55]); (ii) the importance of biotic and abiotic ecological factors in determining the capacity to provide services; and (iii) an awareness of the inherent limitations of qualitative assessments, which offer general insights rather than precise quantifications.

2.4. Statistical Analysis

Statistical analyses were conducted to evaluate differences in soil physical and chemical properties, Cstock measures, soil erodibility, plant species richness, and Landolt’s indices, by considering the recovering ski run sites and control (undisturbed) sites. The Shapiro–Wilk test and Levene’s test were used to verify the normality and homogeneity of the variances, respectively. When data violated both assumptions, the nonparametric Kruskal–Wallis test was applied. If, instead, the residuals were normally distributed but showed unequal variance of samples, Welch t-test was used to compare two independent groups. When both assumptions of normality and homoscedasticity were met, one-way analysis of variance (ANOVA) was performed to compare the means of two or more groups for each dependent variable. A confidence level of 95% (p < 0.05) was used for all statistical tests. Furthermore, principal component analysis (PCA) was used to explore similarities and differences among grouped data. To reduce redundancy, Pearson’s correlation coefficient ($\mathit{\rho }$) was calculated, and variables with high correlation ($\left|\mathit{\rho }\right|$ > 0.80) were excluded from the PCA to avoid multicollinearity. All statistical analyses were performed using R 4.0.3 software (https://www.r-project.org, accessed on 19 May 2025).

3. Results and Discussion

3.1. Soil Properties on Ski Runs and at Control Sites

Various soil types developed outside the ski runs, in virtually undisturbed areas, including Leptosols, Regosols, Luvisols, and Cambisols. Their formation reflects the influence of underlying lithology, steepness, and resulting geomorphic processes (Table 1). Consequently, a wide range of soil horizon types with distinct morphological (

Table 3. Morphological properties of the observed soil profiles. Codes for structure and consistency are taken from the IUSS Working Group WRB (2022). Aggregate shapes can be granular (GR), subangular blocky (BS), angular blocky (BA), platy (PL), or prismatic (PR) structureless with single grain aggregation derived from the parent material (SR). Aggregate dimension can be fine (FI), medium (ME), or coarse (CO). Aggregation degree can be weak (W), medium (M), or strong (S). Moist consistency can be loose (LO), very friable (VF), friable (FR), firm (FI), or very firm (VI).

Profile	Horizon	Depth (cm)	Sand (%)	Silt (%)	Clay (%)	Coarse Fraction (%)	Munsell Color (Moist)	Structure (Shape, Dimension, Degree)	Consistence (Moist)
P01	C	0–20				90	10YR 5/4		LO
	ACb	20–40	72	22	7	60	10YR 6/2	GR, ME, W	VF
P02	OH	0–2					10YR 2/1		LO
	A1	2–6.				40	10YR 2/2	GR, FI, M	FR
	A2	6–30/45	70	25	5	50	10YR 4/2 upper part, 10YR 4/3 lower part	GR, FI, M	FR
	C	30/45–80				85		SR	LO
P03	OF	0–7							LO
	OH	7–18							LO
	A	18–35/45	65	27	8	50	7.5YR 3/2	GR, CO, W	VF
	C	35/45–85/95				60	7.5YR 5/4	BS, ME, W	LO
	AB	85/95–90/105				20	7.5YR 4/3	GR, CO, S	VF
	CB	105–112				85	7.5YR 5/4	BS, ME, W	LO
P04	A1	4–28.				10	10YR 3/4	BS, ME, S	FR
	A2	28–48				10	10YR 4/3	PL, ME, M	FR
	A3	48–53				5	10YR 2/2	GR, CO, S	FR
	A4	53–58				5	Black	PL, ME, M/GR, CO, M	FR
	Bt1	58–65	21	54	25	5	10R 4/4	BA, ME, S	FR
	Bt2	65–110				10	10YR 5/8	BS, ME, S	FR
	BC	110–140				15	10YR 5/6	BS, ME, S	FI
	2Bt3	140–180+				0	10YR 5/8	PR, CO, S	FI
P05	OF	0–2							LO
	A	2–15	80	15	5	60	10YR 2/2	GR, FI, M	VF
	Bw	15–35				40	10YR 4/4	BD, ME, M	FR
P06	A	0–8				80	2.5YR 4/2	GR, ME, W	VF
	BC	8–35	61	30	9	90	10YR 5/4	GR, ME, W	VF
P07	OL	0–2
	OH	2–4.							LO
	A	4–13.	48	31	21	50	7.5YR 3/2	GR, CO, S	FR
	BA	13–30				30	7.5YR 3/4	BS, ME, M/GR, CO, M	FR
	Bw	30–55				30	7.5YR 4/6	BS, ME, M	FR
	C	55–80				90	10YR 6/4	SR	LO
P08	OL	0–3
	OF	3–5							LO
	OH	5–12							LO
	A	12–20/25				50	10YR 4/3	GR, ME, M	FR
	Bw	20/25–85	26	43	31	30	10YR 5/6	BS/BA, ME, M	FR
	Bt	85–				70	10YR 4/6	BA, ME, S	FR
P09	OH	0–1							LO
	A	1–3	58	35	8	50	10YR 3/2	GR, ME, W	VF
	C	3–25				50	2.5Y 5/4	SR	LO
P10	OL	0–1
	OF	1–2.							LO
	OH	2–4.							LO
	A1	4–14.				10	7.5YR 4/6	GR, ME, S	VF
	A2	14–23	41	30	29	10	7.5YR 4/3	GR, ME, M	FR
	Bw1	23–30				10	7.5YR 5/6	BS, ME, M/GR, CO, M	FR
	Bw2	30–45				10	7.5YR 5/8	BS, ME, M	FR
	B@g	45–60				2	10YR 5/6	PL, ME, M	VI
	CB	60–85				3	10YR 6/8	PL, ME, W	FI
P11	A	0–10	48	28	24	80	10YR 5/6	GR, ME, M	VF
	CB	10–20				80	10YR 4/4	BS, ME, W	VF
P12	OH	0–3
	A	3–15	31	43	26	40	10YR 3/4	GR, FI-CO, S	FR
	Bw	15–27				5	10YR 5/8	BS, ME, W/GR, CO, M	FR
	BC	27–35				80	10YR 4/6	SR	LO
	CB	35–70				90	10YR 5/4	SR	LO
P13	OL	0–0.5
	OF	0.5–1
	AE	1–17.	36	43	21	20	10YR 4/2	BS, ME, W/GR, ME, M	VF
	Bw1	17–30				0	10YR 4/6	BS, CO, S	FR
	Bw2	30–55				20	10YR 4.5/6	BS, CO, S	FR
P14	BC	0–50	60	29	11	90		BS, ME, W/GR, ME, M	FR
P15	OL	0–1
	A	1–7	81	17	2	90	10YR 3/6	GR, FI, W	VF
	CB	7–30				90	10YR 3/6	SR	VF
P16	OL	0–1
	A	1–20	75	22	3	20	10YR 2/2	BS, ME, W/GR, fi, M	VF
	CB	20–35				90	10YR 5/6	SR	LO

), physical, and chemical properties (

Table 4. Main chemical properties of the observed soil profiles.

Profile	Hor.	pH	CaCO₃ (%)	BD (kg/m3)	C org (%)	N tot (%)	C/N
P01	C	/	78	1275	0.52	0.05	11.3
	ACb	8.25	70	1094	1.11	0.09	12.0
P02	A1	7.4	68	544	11.19	0.95	11.8
	A2	8.0	70	802	3.79	0.22	17.2
	C		78	1490	0.21	0.02	10.5
P03	A	7.6	64	662	8.089	0.58	13.9
	C	8.4	77	1502	0.20	0.04	5.0
	AB	7.7	43	668	6.64	0.55	12.1
	CB	8.4	74	1502	0.20	0.04	5.5
P04	A1	7.2	0	939	2.13	0.19	11.2
	A2	7.9	12	882	2.71	0.41	6.6
	A3	7.4	2	436	17.60	-	-
	A4	7.5	0	909	2.40	0.23	10.4
	Bt1	7.6	1	1368	0.35	0.01	37.8
	Bt2	7.6	0	1357	0.37	0.01	39.9
	BC	7.7	0	896	2.55	0.10	26.8
	2Bt3	7.6	0	1188	0.75	0.05	13.8
P05	A	7.3	5	866	2.90	0.35	8.3
	Bw	7.5	1	1669	0.10	0.01	10.0
P06	A	7.4	78	1009	1.59	0.10	16.4
	BC	8.1	87	1502	0.20	1.07	0.2
P07	A	6.9	0	556	10.63	0.96	11.1
	BA	7.6	1	609	8.51	0.99	8.6
	Bw	7.5	1	768	4.36	0.48	9.1
P08	A	6.2	0	747	4.78	0.39	12.3
	Bw	7.3	0	901	2.50	0.28	9.0
	Bt	7.2	0	978	1.81	0.14	12.7
P09	A	7.1	17	1502	0.20	0.08	2.4
	C	8.1	34	1335	0.40	0.39	1.0
P10	A1	6.1	0	702	5.77	0.40	14.3
	A2	7.1	0	576	9.80	0.75	13.1
	Bw1	7.8	8.1	1220	0.65	0.01	71.7
	Bw2	6.2	0	1204	0.70	0.04	17.5
	B@g	7.6	7.6	1667	0.74	0.03	24.7
	CB	7.6	12.1	1667	0.41	0.02	20.5
P11	A	7.8	13.2	818	3.5	0.40	9.0
	CB	8.1	25.8	1092	1.1	0.10	11.7
P12	A	5.7	0	731	5.11	0.41	12.4
	Bw	6.1	0	977	1.82	0.13	14.2
	BC	7.5	29	1502	0.20	0.1	2.4
	CB	10.5	44	1009	1.59	0.14	11.2
P13	AE	5.7	0	758	4.56	0.44	10.3
	Bw1	7.2	0	573	9.91	0.04	267.3
	Bw2	7	1.1	1035	1.42	0.11	13.2
P14	BC	7.3	70	1502	0.20	0.12	1.6
P15	A	7.2	21	440	17.32	1.07	16.2
	CB	7.6	84	1502	0.20	-	-
P16	A	7.3	0	497	13.60	1.41	9.6
	CB	8.2	86	1125	0.97	0.08	12.9

) was observed. Soil formed on gentler slopes was generally thick and rich in clay and silt (P04, P08, P10, P12, P13; Figure 2). In some cases, the clay and silt content was notably high. This fine texture is typical where carbonate weathering has effectively dissolved the parent material, concentrating fine particles in the soil profile [56]. The thick yellowish-brown silty layers observed in the Bt horizons of profiles P04, P10, and P12 closely resemble loess deposits commonly associated with the Last Glacial Maximum (e.g., [57]). Evidence suggesting a pre-Holocene origin for these soils includes the presence of cryogenic platy structures in the deeper horizons (e.g., P04, P10), which are unlikely to have formed under Holocene climatic conditions. These features are often considered indicators of past permafrost or deep seasonal freezing [52,58].

Soils within the ski runs were generally shallow and poorly developed, typically classified as Leptosols or Regosols, with a few exceptions (P04 and P10) where slope reshaping was not necessary. As a result, the total soil profile thickness varied, but due to high variability among control soils, the difference was not statistically significant (p = 0.073, Figure 3b). Such variability in thickness was mainly attributed to grading and erosion during ski run construction, rather than to natural topographic features (Figure 3c). Ski run soils also showed a higher stone content compared to undisturbed areas (p = 0.063, Figure 3d), primarily due to slope reworking and mixing with the underlying fragmented rock layers during construction. This alteration strongly influenced pH values in the A horizons, which were consistently higher and more uniform in ski run soils (p = 0.047, Figure 3e). In contrast, natural soils showed more variability in pH due to differing leaching conditions across soil types. The alkaline values of all ski runs were linked to the presence of carbonates and the high stone content, both results of soil rejuvenation processes. The alkaline pH values of ski run soils were associated with reduced biological activity, despite the absence of tree species. This was reflected in slightly higher C/N ratios (although not significant, p = 0.919, Figure 3f). In fact, ski runs are dominated by short-cycle easily degradable grasses, which tend to produce organic matter with a lower C/N ratio than tree-dominated systems (e.g., [59]). Therefore, the observed higher C/N ratio likely indicates suppressed biological activity rather than vegetation effects. Alkaline conditions are generally less favorable for soil fauna, earthworms, and microbial communities, leading to slower organic matter decomposition [60,61]. The weak biogenic granular structure observed in many A horizons of ski run soil profiles further supports this interpretation. Exceptions included P02 and P03. P03 showed lower pH values for the ski run soil than for its natural counterpart. This was likely due to the low slope steepness of P02, which reduced erosion, combined with complete herbaceous cover and abundant spruce regeneration, contrasting with the much steeper and less vegetated natural site P03.

Four of the ski run soil profiles exhibited hemimoder humus forms, characterized by arthropod activity and an absence of earthworms (Table 1). This is somewhat unexpected, as organic matter from herbaceous plant communities is generally more easily degradable and more palatable to decomposer organisms than that from forest vegetation. Hemimoder is typically associated with acidic forest soils or grassland soils under harsh climatic conditions. This suggests that the decomposer communities have not yet fully recolonized soils disturbed by the construction and operation of the ski tracks. The persistence of limiting factors, such as excess of carbonates or other disturbances introduced during ski run construction and management, appears to have inhibited soil recovery for decades. Additional constraints, possibly active during the ski runs’ operational period, may include lower soil temperatures under the snowpack due to higher density of artificially compacted snow (e.g., [62]). In contrast, the remaining four ski run profiles had eumull forms, indicating that equilibrium has been reached between edaphic conditions and faunal activity under non-acidic grassland environments. In comparison, all control sites, except one with a hemimoder form, showed amphi humus forms. Amphi is the most commonly observed humus form in beech forests in Italy [63,64].

3.2. Carbon Stocks on Ski Runs and at Control Sites

Cstock analysis revealed that soils under natural vegetation generally contain more organic carbon, except for two very deep, undisturbed soil profiles (P04 and P10) excavated from ungraded, cleared ski runs. The much greater thickness of these profiles contributed to the variations in Cstocks across the entire soil profile (Figure 4a). When focusing on the top 30 cm of soil, the difference between ski run and control soils remained significant (Figure 4b). Excluding P04 and P10, ski run soils contained, on average, only 16% of the organic C stocked by control soils. This provides further evidence that human disturbances associated with ski run construction and use can have long-lasting impacts, even after abandonment. The findings align with results from other alpine sites, where organic Cstocks in ski run soils were reduced by approximately 80% compared to nearby undisturbed soils [9,10], and in the Sierra Nevada in Spain, where reductions of 33% were observed [65]. Additionally, the thickness of the O horizon, and consequently its Cstock, was significantly lower in ski run soils compared to control soils (Figure 4c). This difference is likely due to most of the ski run soils being covered by pioneer or grassland communities, with very young trees observed only occasionally.

3.3. Soil Erodibility and Erosion

Applying the equation by Wischmeier and Smith (1978) [47], ski run soils were found to be more erodible than natural control soils (p < 0.007, Figure 5). Soil erodibility is influenced by factors such as texture, organic matter content, structure, and permeability. As discussed in earlier sections, most ski run sites showed lower organic matter content and weaker biological activity compared to natural soils, which also affected soil structure. Additionally, while not statistically significant, ski run soils had a lower clay content, a key factor in soil cohesion that strongly reduces erodibility. Although the permeability of topsoils was generally good, very shallow soils on graded slopes posed challenges. During intense rainfall, the limited total pore volume of these soils can quickly fill down to the almost impermeable bedrock, leading to runoff. This behavior contrasts with the typically permeable mountain soils characterized by low BD (Table 4). A similar reduction in the water-holding capacity of ski run soils has been detected in other studies [6].

Graded ski runs often lacked soil on their steepest sections (e.g., P06, P11, P14, Figure 6a), where active erosion occurs during the intense rainfall events characterizing the climate of Valcanale. Additionally, the removal of complete vegetation cover on silt-rich loess soils during ski run grading and managing (e.g., P11) has created ideal conditions for the development of gullies. These gullies continue to strip soils entirely from the affected ski runs (Figure 6b,c). Unlike clay, which enhances aggregate stability and increases resistance to erosion processes, silt inherently increases soil erodibility, exacerbating these processes [66,67].

3.4. Floristic Features

A total of 110 species/taxa were identified across the 16 sampling sites (Supplementary Table S1), including 36 species protected at both international and national levels [43,44]. This underscores the area’s remarkable floristic and conservation importance.

While ski runs often hosted a greater number of species compared to control sites, the difference was not significant (p = 0.290, Figure 7).

The main chorological types identified at the sites include endemics (8.2%), mid- or south-European orophiles (29.0%), and Eurasian and Eurosiberian (29.9%), Atlantic (4%), Circumboreal (13.6%), Arctic-alpine (5.5%), and Cosmopolitan and Paleotemperate (8.2%) species. Notably, no exotic species were recorded, indicating that the study area has avoided the floristic pollution often observed in other mountain regions of the Alps affected by anthropogenic disturbance [68].

Nine endemic species from the Alps or Italy were identified, including some with a particularly restricted geographical distribution (stenoendemic species). Examples include Aquilegia confusa, Campanula raineri, Carex baldensis, and Xerolekia speciosissima, which are confined to a few valleys in the Italian Alps between Lake Como and the Venetian Prealps (Insubrian endemic species, [37,42,69]). Of these, seven endemics (Aquilegia confusa, Cerastium latifolium, and all the stenoendemics) were observed exclusively in highly disturbed ski run soils (Supplementary Table S1). In contrast, three endemic alpine species (Rhododendron hirsutum, Phyteuma betonicifolium, and Phyteuma scheuchzeri) were found only at control sites, while one species (Centaurea transalpina) was present in both habitats. Cosmopolitan species were more frequently observed in ski run soils.

In agreement with the lack of forest cover and with the shallower and stony soils, plant species on ski runs were generally more heliophilous than those at control sites. On average, L values were 3.74 on ski runs and 2.96 at control sites (Figure 8a). This aligns with findings from other studies showing that ski runs facilitate habitat expansion for grassland species (e.g., [23]). Moreover, control sites supported a greater number of species requiring higher levels of soil organic matter, with an average H value of 3.42 at the control sites compared to 2.75 on the ski runs (Figure 8b). These two indices were significantly different (p < 0.002). The other Landolt indices did not show differences between ski runs and control sites (p > 0.12). Plant communities on the ski runs were indicative of drier soil conditions, with an average F value of 2.62 compared to 2.85 at control sites (Figure 8c), and slightly lower nutrient content, with F values averaging 2.60 on ski runs versus 2.71 at control sites (Figure 8d). These environmental characteristics are similar to those observed in limestone screes [70].

Interestingly, the creation of ski runs has resulted in the development of skeletal and shallow soils, providing suitable habitats for certain endemic species that thrive on limestone rocks and screes. This has enabled the range expansion of these species in profiles P01 and P06. Although human interventions have caused severe erosion, particularly on the “Canalino” ski run, they have also facilitated the establishment of species with high natural value. This paradox highlights how anthropogenic activities can simultaneously enhance biodiversity, through the range expansion of endemic species, and intensify ecosystem degradation, including the loss of carbon storage.

3.5. Plant and Soil Correlations

The PCA analysis, performed after removing intercorrelated variables ($\left|\mathit{\rho }\right|>0$.80, Figure 9), provided additional insights. Notably, Cstock showed a strong positive correlation with soil thickness ($\mathit{\rho }$ = +0.82) and a strong negative correlation with stoniness ($\mathit{\rho }$ = −0.85), which itself was highly correlated with Cstock30 ($\mathit{\rho }$ = −0.86). Consequently, both Cstock and stoniness were excluded from the analysis. The resulting PCA biplot (Figure 10) separated two large groups along the first axis, primarily representing a gradient of increasing Cstock in mineral horizons, soil thickness, biodiversity, and Landolt indices. Most ski runs were on the left side of the biplot, while control sites aligned on the right, along with ski runs P04 and P10. These two sites significantly differed from the other ski runs, as they were not subjected to grading, slope reshaping, and soil movements. Instead, P10 was only cleared of scattered trees, and P04 was already a pasture. Consequently, their soil and ecosystem properties were more similar to those of the control sites than to the graded ski runs. This highlights the important and long-lasting effects of the construction methods of ski runs [71]. Based on these findings, we distinguished the two cleared ski runs from the graded ones when evaluating SBESs.

3.6. Soil-Based Ecosystem Service

The assessment of SBESs, illustrated through radar charts (Figure 11), reveals substantial heterogeneity within the former ski resort at Valcanale. This variability is closely tied to differences in the intensity and type of anthropogenic disturbance, demonstrating how the degree and nature of human activities have left enduring imprints on the mountain environment.

Areas where human disturbance was limited to forest clearing—such as the “Collino” ski run (P04) and the upper section of the “Muro” ski run (P10)—show a more advanced trajectory of ecological recovery. These zones demonstrate favorable soil conditions and relatively intact ecosystem functionality, suggesting that vegetation removal alone, although a disturbance, leads to less persistent impacts than interventions involving significant geomorphological modifications.

In contrast, areas subject to intense earthworks, grading, and infilling during the construction and reshaping of ski runs exhibit highly heterogeneous soil conditions. These are often characterized by shallow soil and signs of soil degradation and instability [33]. Such heterogeneity likely results from altered physical and chemical soil properties, including the loss of organic and organo-mineral horizons, soil compaction, reduced biological activity, and disrupted drainage regimes. In these cases, geomorphological changes have not naturally stabilized over time, and without targeted restoration works, these areas are unlikely to return to equilibrium in the foreseeable future. Infrastructures such as paths, ski runs, and roads further exacerbate the situation of channeling water flow, contributing to erosion and damaging both infrastructure and adjacent slopes. Overall, human-induced alterations have significantly compromised the soil’s capacity to support vegetation and sustain ecological processes. The ongoing erosive dynamics and the challenges in re-establishing a stable vegetative cover highlight the complexity of ecological recovery in these settings [13]. Addressing these persistent challenges requires active management to restore ecological balance and prevent further environmental degradation [33].

A compelling example of this duality was observed along the “Canalino” ski run, where striking differences emerge among profiles P01, P06, and P02. At P01, human activity intensified the geomorphic processes within a pre-existing gorge, leading to severe soil erosion, low organic carbon content, unfavorable C/N ratios, and suboptimal pH. Despite these degraded conditions, the site has gained unexpected ecological value due to the appearance of the endemic species Campanula raineri, indicating an increase in biodiversity and the formation of novel habitats. At P06, located in the upper sector of the Vallone ski run, increased erosion led to near complete removal of the soil layer. However, this disturbance created favorable conditions for several endemic species, including C. raineri, Xerolekia speciosissima, and others, suggesting that certain highly disturbed areas may offer refuge for rare and endemic flora. This has contributed to the HBI for ski runs. In this sense, ski run construction can, in specific cases, provide a sort of positive ecological effect, i.e., the (re-)creation of habitats for rare grassland species [34].

In contrast, at P02, the soil has been mixed, disrupting the original horizon sequence, but it has not been removed. Here, the conditions are more favorable for forest species regeneration, yet no endemic species have been observed.

These cases reflect the ambivalent ecological outcomes of human disturbance. On one hand, altered environments can foster increased floristic diversity and enhance both the natural and cultural heritage. On the other hand, they often exhibit critical ecological deficits, such as reduced water retention capacity, compromised erosion control, impaired local microclimate regulation, and disrupted carbon cycling. This tension underscores the need for a nuanced understanding of post-disturbance ecological trajectories.

Our findings align with existing literature that highlights the long-term ecological consequences of ski infrastructure, particularly in terms of changes in soil structure, fertility, and vegetation dynamics [62]. Furthermore, the spatial variability observed at Valcanale reinforces the importance of considering the specific history of disturbance, as well as the underlying geological and geomorphological context, when assessing re-naturalization processes. As also noted in studies on soil instability in the area, anthropogenic modifications often interact with pre-existing terrain fragilities, thereby amplifying degradation and instability processes [33].

The observed heterogeneity in SBES provision reflects not only current ecological conditions, but also the lasting legacy of past management practices and disturbance regimes. In many cases, spontaneous morphological and functional recovery of these ecosystems is unlikely due to ongoing erosion, hydrological channeling caused by ski runs and service roads, and the absence of natural pathways for vegetation recolonization. These findings highlight the urgent need for targeted, site-specific, and proactive management strategies to reverse degradation trends and support long-term ecological restoration in alpine ski landscapes.

Regarding specific ecological properties [71], the recovery of SBESs on abandoned graded ski runs has not reached the levels observed at nearby undisturbed control sites. In contrast, cleared ski runs, where soil layers were not disturbed during ski run construction, exhibit ecosystem service provision comparable to that of control sites.

4. Conclusions

Following abandonment, ski slopes show differing recovery dynamics depending on the extent of initial soil disturbance. Graded slopes subjected to significant earthworks remain ecologically degraded, characterized by poor soil structure, ongoing erosion, and low carbon stocks. However, these areas may support the local expansion of rare and endemic species, particularly those adapted to scree environments, resulting in a localized increase in biodiversity. This ecological gain, however, comes at the cost of reduced carbon sequestration capacity and diminished resistance to erosion.

In contrast, cleared slopes, where interventions were limited to forest removal, undergo faster and more predictable recovery. These areas more readily regain vegetation cover, with species such as willows contributing to slope stabilization and soil regeneration. Their ecosystem functions increasingly resemble those of adjacent undisturbed forests and do not typically require active restoration.

This study assessed soil re-naturalization and ecosystem functionality in the former Valcanale ski area, revealing a clear contrast in ecological recovery across different types of ski runs.

The key finding is that ski runs generally exhibit lower carbon stocks and higher erodibility compared to nearby natural undisturbed areas. Nonetheless, these disturbed habitats have also facilitated the emergence and expansion of rare and endemic species, particularly those suited to scree and stony substrates. Among these, cleared slopes, where soil structure was largely preserved, show a faster and more consistent recovery trajectory, with soil and vegetation characteristics closely aligning with those of nearby undisturbed ecosystems. These areas typically do not require intervention to restore ecological function.

Conversely, ski runs, affected by intensive earthworks and grading, remain in a degraded state, with poorly developed soils, persistent erosion, and limited ecological functionality. In these cases, successful re-naturalization will require targeted restoration measures, particularly soil accumulation techniques aimed at reconstructing lost pedogenic horizons and improving substrate conditions.