Tracking Mountain Degradation for the United Nations (UN) Sustainable Development Goals (SDGs) Using the State of Colorado (USA) as an Example

Abstract

Mountain ecosystems, strongly affected by climate-related variability and human impact, are degrading faster than other terrestrial ecosystems. Currently, the United Nations (UN) utilizes Sustainable Development Goal (SDG) 15: Life on Land (Target 15.4 and Sub-indicators 15.4.2a and 15.4.2b), along with the System for Earth Observation Data Access, Processing and Analysis for Land Monitoring, commonly referred to as SEPAL, to track mountain degradation. This SEPAL analysis does not include soil data, which is critical to understanding mountain degradation. The present research focuses on improving the tracking and evaluation of mountain land degradation (LD) utilizing soil data in the state of Colorado (CO) in the United States of America (USA) as an example. Total anthropogenic LD affects an estimated 19% of Colorado’s territory as of 2024, driven mainly by agricultural activities (80%). Between 2001 and 2024, overall LD in CO decreased (−0.4%), but LD from development increased by 23.3%. For mountain areas in CO, the mountain green cover index (MGCI) was 96% for 2024, and it decreased (−0.4%) between 2001 and 2024. The mountain LD proportion was 2.5% as determined by the SEPAL method compared to 4.4% by LULC analysis. Incorporation of soil data into LULC analysis found that between 2001 and 2024 LD increased to 6.6%. All soil types in the mountains exhibited anthropogenic LD due to development with a total developed area of 1385.1 km². Current total mountain LD (inherent + anthropogenic) in CO may be as high as 38.9%. Future estimates of total mountain LD should include both inherent and anthropogenic LD.

1. Introduction

Mountains encompass vast regions across many countries worldwide and provide ecosystem services to people who live within and beyond them, which are crucial for sustaining life. Based on updated estimates from the United Nations Environment Program World Conservation Monitoring Centre (UNEP-WCMC), mountainous regions account for approximately 27% of the global terrestrial surface [1]. Mountain ecosystems harbor nearly one-third of global biodiversity and include approximately half of the world’s biodiversity hotspots, a pattern shaped by steep gradients and complex topography [2,3,4]. However, ecosystems found in mountains face growing threats due to their sensitivity to climate change and human-induced impacts. These fragile landscapes are deteriorating faster than any other terrestrial habitats [5], highlighting the urgent need for their monitoring, protection, and restoration.

Considering this, adopting Chapter 13 of Agenda 21 in Rio (1992) was a pivotal step in acknowledging mountain conservation [6]. Later efforts included the 1995 establishment of the Mountain Forum, the declaration of 2002 as the International Year of Mountains (IYM2002), as well as the 2002 World Summit on Sustainable Development (WSSD2002) [6,7]. Similarly, the 2015 United Nations (UN) 2030 Agenda for Sustainable Development listed Target 15.4 (

Table 1. Mountain land degradation target, indicator, and sub-indicators relevant to the United Nations (UN) Sustainable Development Goal (SDG) from the “Global Indicator Framework for the Sustainable Development Goals and Targets of the 2030 Agenda for Sustainable Development” (adapted from Assembly, U.G., 2017) [11] ¹.

United Nations (UN) Sustainable Development Goal (SDG), Target and Indicator
SDG 15. Protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss.
Target 15.4: By 2030, ensure the conservation of mountain ecosystems, including their biodiversity, in order to enhance their capacity to provide benefits that are essential for sustainable development.	Indicator 15.4.2: - Sub-indicator 15.4.2a: Mountain green cover index. Units: %. - Sub-indicator 15.4.2b: Proportion of degraded mountain land. Units: %.

¹ Sustainable Development Goal indicators should be disaggregated, where relevant, by income, sex, age, race, ethnicity, migratory status, disability and geographic location, or other characteristics, in accordance with the Fundamental Principles of Official Statistics, United Nations (UN) Resolution 68/261 [12].

), which focuses on the conservation of mountain ecosystems, which are critical for sustainable development [8]. The sub-indicator “mountain green cover index” (MGCI) (15.4.2a) serves as the metric for assessing this target, measuring changes in green cover (cropland, grassland, wetland, tree-covered, and shrub-covered areas) in mountainous areas (Table 1). Furthermore, at the 54th session of the Statistical Commission, the Inter-Agency and Expert Group on Sustainable Development Goals (IAEG-SDG) report presented the refined SDGs (target 15.4) by adopting a new sub-indicator “proportion of degraded mountain land” (15.4.2b) [9], to assess country-level patterns of mountain land degradation (LD) associated with changes in land use and land cover (LULC) (Table 1) [10].

The Food and Agriculture Organization (FAO) is a key organization that monitors SDG sub-indicators 15.4.2a and 15.4.2b through an open-source, cloud-based platform (SEPAL) [13] (details of SEPAL are provided in Section 2). This platform assesses both the mountain green cover index (Sub-indicator 15.4.2a) and land cover change to evaluate mountain land degradation (Sub-indicator 15.4.2b) and is designed to evaluate changes in degradation from a starting or base year [14]. Multi-Resolution Land Characteristics Consortium (MRLC) [15]-classified satellite imagery when reclassified into standard System of Environmental-Economic Accounting (SEEA) classes of land cover, including artificial surfaces, cropland, grassland, tree-covered areas, shrub-covered areas, wetland, barren land, snow/glacier and water bodies, can be used to evaluate the trajectory of LD using a standardized land cover change matrix [14] (Figure 1). Land cover changes are rated as either stable, improved, or degraded, with no change (even from a degraded state) being designated as stable [14] (Figure 1). This method’s change-matrix approach and sub-indicator offer advantages over the MGCI, as they identify land cover transitions that represent degradation but may also increase greenness levels in mountain regions. Limitations of this approach include the use of low-resolution land cover data (e.g., 300 m) that may not accurately represent mountain land covers, and the system’s exclusion of soil type and soil quality information, which can vary widely and limit plant productivity.

Unfortunately, analysis at this low resolution cannot effectively capture fine-scale changes [16]. Therefore, in recent years, the emergence of medium-resolution remote-sensing datasets has become a key approach for tracking land resources and plays a crucial role in evaluating the SDGs [17,18]. Mapping MGCI and LD using medium-resolution imagery is important for sustainable mountain protection and development at the state and county levels. The current methods for monitoring MGCI use land cover change [14] and greening trends over the last several decades, which are based on the net primary productivity (NPP) [19]; however, increase in greenness may be due to land conversions, longer growing seasons due to climate change, intensive agricultural practices and losses in soil fertility [20]. Human-induced land management (cropland) is a key driver of Earth’s greening, accounting for 33% of total global greening since 2000 [21]. Existing studies often focus exclusively on MGCI at a global scale [22], overlooking equally important degradation studies. The state of CO can serve as a model for studying LD in mountainous regions, because it contains the largest portion of the Rocky Mountains, which have degraded between 2000 and 2010 due to growing population pressure [23]. Detecting degradation at finer scales is crucial due to its subtle spatial details and slow progression over time [24], both of which are essential for effectively achieving mountain conservation goals. Many national or global assessments may show overall greening, but they often overlook significant areas of degradation. This emphasizes the need for higher-resolution satellite data, which is crucial to accurately tracking land cover changes and effectively reporting progress toward achieving SDGs [24]. Therefore, this study focuses on monitoring both sub-indicators, namely MGCI (15.4.2a) and estimates of the proportion of mountain land degradation (15.4.2b), derived from medium-resolution land cover data.

This study incorporates soil types found in Colorado (CO) to improve understanding of the MGCI and mountain degradation within each soil type (Figure 2).

Table 2. Soil diversity and soil carbon (C) stocks with corresponding social costs of C (SC-CO₂) (using the Environmental Protection Agency (EPA) valuation of $50 per metric ton of CO₂, valid until 2030, expressed in constant 2007 U.S. dollars and discounted at a mean annual rate of 3% [27]) in Colorado (CO), USA, reported only for areas with SSURGO [26] soil data available.

Soil Order	General Characteristics and Constraints	Area with Soil Data Available (km2)	Midpoint Total Soil Carbon (kg)	Midpoint Social Costs of C ($, USD)
Slightly Weathered Soils
Entisols	Embryonic soils with an ochric epipedon	36,455.1	4.7 × 1011	$85.3B
Inceptisols	Young soils with an ochric or umbric epipedon	10,645.9	1.5 × 1011	$27.4B
Histosols	Organic soils with ≥20% organic carbon	504.4	7.2 × 1011	$13.2B
Moderately Weathered Soils
Aridisols	Dry soil. Common in desert areas	30,262.1	6.0 × 1011	$110.5B
Vertisols	Soils with swelling clays	1554.9	5.9 × 1010	$10.8B
Alfisols	Clay-enriched B horizon with B.S. ≥35%	31,601.5	3.7 × 1011	$68.3B
Mollisols	Carbon-enriched soils with B.S. ≥50%	60,857.5	1.5 × 1012	$278.7B
Total		171,881.4	3.2 × 1012	$594.1B

Note: B.S. = base saturation. Entisols, Inceptisols, Aridisols, Vertisols, Alfisols, and Mollisols are mineral soils. Histosols are organic soils. B = billion = 10⁹. $ = United States dollars (USD). See Supplementary Materials Tables S1 and S2 for minimum and maximum values.

presents soil diversity, expressed as grouped soil orders, in two categories: slightly weathered and moderately weathered. Each soil order plays a distinct role in sustaining the biodiversity it hosts. For instance, Mollisols—among the most productive soils in Colorado—are closely linked to grassland ecosystems, whereas Aridisols are typically characteristic of arid and desert environments [25].

This study hypothesized that disaggregating land cover data analysis for monitoring the MGCI and mountain LD by soil type provides information to better understand which parts of the mountain ecosystem require conservation and management activities. The objectives of this study were to: (1) examine overall LD in CO and its temporal trends; (2) compare the assessment of the mountain green cover index (sub-indicator 15.4.2a) and the proportion of mountain land that is degraded (sub-indicator 15.4.2b) and their trends between 2001 and 2024 using SEPAL [13] and with a separate Geographic Information Systems (GIS) land cover change (LULC) analysis; (3) disaggregate these sub-indicators by soil type and by LD type (inherent, anthropogenic).

2. Materials and Methods

2.1. Study Area

This research was undertaken in Colorado (CO), a mountainous state known for the highest mountains in the contiguous United States (US) (Figure 3). More than half of the counties in CO have at least 60% of their area covered by mountains, with 16 counties having 90% of their area covered by mountains (Figure 3). The state lies within the elevation range from 2134 m to 4267 m, with an average altitude of approximately 2073 m above sea level [28]. It has 830 mountains between 3353 m and 4267 m, and 59 mountains higher than 4267 m in elevation [28]. The high mountains of CO rise gradually from the Kansas and Nebraska plains, stretching 200 miles to the foothills of the Rockies [29]. The eastern part of the state features flat and undulating prairies, with occasional hills and bluffs, creating a diverse landscape across the state [29]. This study determined that in 2024, the landscape of CO was largely covered by herbaceous grasslands (35%) and cultivated crops (14%), primarily concentrated in the eastern region, where mountains are scarce. In contrast, shrublands (23%) and evergreen forests (16%) dominate the western part of the state, which is characterized by extensive mountain ranges (Figure 3). Soil types found in CO are slightly weathered soils (Inceptisols, Entisols, and Histosols) and moderately weathered soils (Alfisols, Aridisols, Mollisols, and Vertisols) (Figure 2; Table 2).

The state is mainly dominated by Mollisols (35.4%) and Entisols (21.2%). Aridisols are dry, coarse soils found at mid-elevations across the state, which support drought-resistant vegetation. Another common soil type found in the state is Alfisols (18.4%), which are typically found in forested areas and mainly beneath coniferous forests at higher elevations. Mollisols are the most fertile soils, while Histosols, which occur in environments where organic material builds up beneath prairie vegetation and within forested areas characterized by limited drainage [31] and are relatively scarce (0.3%). In CO, Alfisols and Mollisols are considered the most productive and important agricultural soils and are scattered at mid-elevational ranges where moisture is more abundant [32]. Entisols are young soils that lack distinct horizons and are typically found on recent erosional surfaces along rivers and streams. Inceptisols represent slightly weathered soils that occur widely across mountainous terrain and along steep slopes. Vertisols (0.9%), which are clayey soils that shrink and form wide cracks at the surface during dry periods, are very rare in CO (Table 2, Figure 2).

2.2. Current Methods for UN SDGs Indicator 15.4.2

The UN SDGs Indicator 15.4.2 consists of two sub-indicators. This study employs the current UN methods for computing sub-indicator 15.4.2a, MGCI, and sub-indicator 15.4.2b, Proportion of degraded mountain area, using SDG indicator 15.4.2 metadata https://unstats.un.org/sdgs/metadata/files/Metadata-15-04-02.pdf (accessed on 14 January 2026) [14]. Land cover data is the primary input for this analysis; therefore, this study uses a medium-resolution national-level land cover dataset for the period of 2001 to 2024.

Sub-indicator 15.4.2a: Mountain Green Cover Index (MGCI) is defined as the total area of “green cover” (cropland, grassland, wetland, tree-covered and shrub-covered areas) within a mountain region in the reporting period n (e.g., 2024) divided by the total mountain area (km²) and then multiplied by 100 (Equation (1)) [14]. The total mountain area used here was defined according to UNEP-WCMC (2002) K1 mountain classification data [30], and green cover areas were determined from land cover maps [15].

$$\mathrm{MGCI}=\left({\displaystyle \frac{\mathrm{Mountain}\mathrm{Green}\mathrm{Cover}{\mathrm{Area}}_{\mathrm{n}}}{\mathrm{Total}\mathrm{Mountain}\mathrm{Area}}}\right)\times 100$$

(1)

Sub-indicator 15.4.2b: The proportion of degraded mountain area is defined as a binary classification (degraded versus non-degraded) representing the share of degraded land relative to the total mountain area [14]. The degraded mountain area represents the total area (km²) of mountain land identified as degraded during the reporting period n, derived from land cover changes deemed to indicate degradation when compared with baseline conditions set at an initial time period (e.g., 2001) (Equation (2)). Land cover changes that constitute degradation (as well as improvement and stable transitions) are defined using a land cover change matrix using the FAO global default estimates. The total mountain area (km²) was defined for this study according to UNEP-WCMC (2002) K1 mountain classification data [30].

$$\mathrm{Proportion}\mathrm{of}\mathrm{degraded}\mathrm{mountain}\mathrm{area}=\left({\displaystyle \frac{\mathrm{Total}\mathrm{degraded}\mathrm{mountain}{\mathrm{area}}_{\mathrm{n}}}{\mathrm{Total}\mathrm{mountain}\mathrm{area}}}\right)\times 100$$

(2)

SEPAL for calculating SDGs indicator 15.4.2: The SEPAL-SDG 15.4.2_beta tool was created by the United Nations Food and Agriculture Organization (UN-FAO) to support national authorities to compute and report SDG Indicator 15.4.2. SEPAL-SDG 15.4.2_beta was built on SEPAL, an open-source cloud-based platform that enables users to query and process geospatial data and perform sophisticated geospatial analyses [13]. This tool can be accessed at https://sepal.io (accessed 26 November 2025) [13]. Details about SEPAL can be found in the article by Ghosh et al. (2024) [33]. The SEPAL-SDG 15.4.2_beta tool requires a mountain area map, an administrative boundary map, and a land cover map for the area of interest to compute the SDG indicator 15.4.2. This tool uses several global data sources when similar datasets for the region of interest are unavailable. The global datasets, which include administrative boundaries, land cover, and digital elevation models, are available through FAO GAUL: Global Administrative Unit Layers 2015 (Google Earth Engine Data FAO GAUL) [34], European Space Agency (ESA) Climate Change Initiative (CCI) Land Cover (ESA-CCI land cover) [35] and the Shuttle Radar Topography Mission (SRTM) (Google Earth Engine Data SRTM) [36], respectively.

This study utilizes national-level land cover data (NLCD) (https://www.mrlc.gov/) [15] that has an overall accuracy of at least 82.75% [37], which is likely higher because of the use of deep-learning techniques in the most recent NLCD releases, which likely improved data accuracy [38]. Administrative boundary datasets were obtained from the National Map (https://apps.nationalmap.gov/) [39] for CO, while the mountain layer used was the global mountain dataset provided by SEPAL [13] based on the UNEP-WCMC definition [40]. Under the UNEP-WCMC framework, total global mountain area is calculated as the aggregate of seven elevation-based categories—commonly referred to as the Kapos mountain classes—defined using elevation, slope, and local relief metrics. These mountain areas are further partitioned into bioclimatic zones (nival, alpine, montane, and remaining mountain areas) based on mean temperature thresholds [41]. Both sub-indicators are disaggregated by mountain bioclimatic belts [41] using the SEPAL tools’ default settings. Additionally, sub-indicator 15.4.2a is disaggregated according to the ten SEEA categories defined by the UN Statistical Division (2014) [42]. For this study, the US NLCD layers were reclassified to align with the SEEA land cover classes. The baseline and reporting years are 2001 and 2024, respectively. Calculated values are expressed as both proportions (percent) and total area (km²).

2.3. Innovation of This Study

This study’s innovation is the combination of LD analysis using land cover data with soil spatial data. For LD analysis, land cover data were used with 2001 as the baseline year and 2024 as the reporting year. The LD was computed as the sum of degraded land from anthropogenic sources, including developed, agricultural, and barren land. Developed land is categorized into the following types: developed, low intensity; developed, medium intensity; developed, high intensity; developed, open space. The agriculture category includes land cover types such as cultivated crops and hay/pasture. Potential land for nature-based solutions (NBS) was computed as the sum of the land cover classes of barren land, shrub/scrub, and herbaceous areas to identify areas for NBS without altering current land uses. Additionally, the area change was calculated using Equation (3):

((2024 Area − 2001 Area)/2001 Area) × 100%.

(3)

The detailed methodology for calculating the area is described in the geospatial analysis section. A major innovation introduced in this study is the disaggregation of SDG indicator 15.4.2 by soil order. Although the global UN indicator reports degradation for mountain regions only, degradation is not uniform across ecological or pedological environments. Therefore, disaggregation by soil type is necessary to determine which soil order is most degraded and the extent of the potential area for NBS, both at the CO state level and in the mountain areas of the state.

2.4. Innovative Geospatial Analysis to Track Mountain Land Degradation

This study utilizes four input datasets: land cover (2001 and 2024) [15], Soil, Kapos (K1) mountain [30], and the CO administrative boundary [39] (Figure 4). The K1 mountain layer was chosen because it is the standard layer used in SEPAL analysis. The classified land cover dataset has a 30 m resolution and is derived from the MRLC (https://www.mrlc.gov/) [15]. The K1-mountain raster data has a resolution of approximately 1 km and was obtained from (https://apps.usgs.gov/glbeco/gme.html (accessed on 7 March 2024)) [30], which is resampled to match the 30 m land cover resolution, and soil data was obtained from the SSURGO database (https://www.nrcs.usda.gov/) [26]. The boundary shapefile was obtained from (https://apps.nationalmap.gov/) [39]. The obtained raster land cover layers were converted to vector format and subsequently unioned with the soil, utilizing the union tool to obtain the overall soil–land cover combined area. Results were verified in multiple locations to confirm that geoprocessing errors did not occur. As the target of this project is the mountain region, the vectorized mountain layer was utilized to clip the land cover and soil data within the mountain region. Then, using the union tool, the combined area of mountain land cover and soil was calculated. To quantify the area at a county scale, the unioned mountain land cover and soil combined layer was disaggregated by county boundaries using the intersect tool.

2.5. Soil Carbon Content and Social Cost of Carbon Analysis

The total soil carbon content and the social cost of carbon dioxide emissions (SC-CO₂) associated with land-use conversion (e.g., development) in the state and mountains in CO were analyzed. The total soil carbon (TSC) (kg m⁻²) is calculated by combining the spatial analysis of soil (total area of degraded soil from developments) and the C values obtained using soil order from Guo et al. (2006) [44]. For the economic analysis, the EPA-estimated SC-CO₂ of $50 per metric ton of CO₂ [27] (Table S1) was used to calculate C losses in monetary damages. The EPA’s SC-CO₂ estimate of $50 per metric ton, which is applicable for 2030, using 2007 US dollars and an average discount rate of 3% [27], provides a baseline valuation of damages from CO₂ emissions. However, this estimated value is likely conservative and therefore underestimates the full economic costs and broader impacts of climate-related damages. Monetary values ($ m⁻²) were obtained for each area with Equation (4), while totals were calculated by summing within polygon boundaries (with SC = soil carbon and a metric ton equal to 1 megagram (Mg) or 1000 kilogram (kg)):

$${\displaystyle \frac{\$\mathrm{USD}}{{\mathrm{m}}^2}}=\left(\mathrm{SC}\mathrm{Content},{\displaystyle \frac{\mathrm{kg}}{{\mathrm{m}}^2}}\right)\times {\displaystyle \frac{1\mathrm{Mg}}{{10}^3\mathrm{kg}}}\times {\displaystyle \frac{44\mathrm{Mg}{\mathrm{CO}}_2}{12\mathrm{Mg}\mathrm{SC}}}\times {\displaystyle \frac{\$50\mathrm{USD}}{\mathrm{Mg}{\mathrm{CO}}_2}}$$

(4)

Using Mollisols as an illustrative case, Guo et al. (2006) [44] reported a mid-range estimate of total soil carbon (TSC) of 25.0 kg m⁻² to a depth of 2 m (Table S1). This soil carbon stock was converted to an area-normalized TSC value using Equation (4), resulting in an estimate of $4.58 m⁻² (Table S1). The resulting TSC content and corresponding area-normalized value were then multiplied by the total area of Mollisols within the Colorado mountain region, for example (27,870.0 km²), to develop an estimated TSC stock of 7.0 × 10¹¹ kg of C with an associated $130B monetary value of SC-CO₂ (Table S5).

3. Results

3.1. SEPAL Analysis of the Sub-Indicator 15.4.2a: Mountain Green Cover Index (MGCI)

The total mountain area resulting from the SEPAL [13] was 123,471.05 km², of which mountain categories, such as nival, alpine, montane, and the remaining mountain areas, occupy 1078.18 km², 13,406.12 km², 108,553.6 km^2, and 433.15 km², respectively (

Table 3. Areas of the different System of Environmental-Economic Accounting (SEEA) [45] land cover classes along with the different bioclimatic belts for the mountains of Colorado (CO), USA, for the reporting period 2024 from the SEPAL tool [13]. Area changes (km²) from 2001 to 2024 are shown in parentheses. Reported values have been rounded; therefore, calculated sums and percentages may exhibit minor discrepancies.

System of Environmental-Economic Accounting (SEEA) Land Cover Classes	Bioclimatic Belts				Total
	Alpine	Montane	Nival	Remaining Mountain Area
2024 Area, km2 (Change 2001–2024, km2)
Artificial surfaces	43.9 (+3.4)	2094.2 (+333)	0.4 (+0.1)	64.1 (+9.9)	2202.6 (+346.5)
Croplands	0.2 (0.0)	1075.6 (−172.8)	0.0 (0.0)	21.3 (−5.2)	1097.1 (−179.9)
Grassland	2430.3 (+146.5)	6892.9 (+1724.1)	317.2 (+1.0)	0.1 (0.0)	9640.5 (+1871.7)
Inland water bodies	33.4 (+0.5)	320 (+8.8)	2.2 (+0.1)	9.8 (+0.3)	365.4 (+9.7)
Permanent snow and glaciers	7.5 (0.0)	0.4 (0.0)	0.3 (0.0)	0 (0.0)	8.2 (0.0)
Shrub-covered areas	2762.4 (+2.0)	38,267.9 (+109.5)	165.5 (−6.1)	309.3 (−4.5)	41,505.1 (+101.0)
Shrubs and/or herbaceous vegetation, aquatic or regularly flooded	556.5 (+1.6)	2342.3 (+3.7)	35.2 (+0.1)	22.7 (−0.6)	2956.8 (+4.8)
Terrestrial barren land	1664.2 (+8.1)	521.3 (+10.2)	366.6 (+0.8)	1.2 (0.0)	2553.1 (+18.9)
Tree-covered areas	5907.7 (−162.2)	57,039 (−2014.6)	190.8 (+4.0)	4.8 (+0.2)	63,142.3 (−2172.6)
Total	13,406.1	108,553.6	1078.2	433.1	123,471.1

Note: Nival: areas with a mean temperature < 3.5 °C during the growing season and a length of growing season < 10 days. Alpine: areas located above the tree line with mean growing season temperatures < 6.4 °C and growing season length between 10 and 54 days. Montane: areas below the tree line with mean growing season temperatures between 6.4 °C and 15 °C. The remaining mountain areas represent mountain regions with mean growing-season temperatures above 15 °C, including both frost and non-frost areas [41].

). The results from the SEPAL land cover classes across Colorado’s mountain region showed that several measurable changes occurred between 2001 and 2024. Artificial surfaces have shown a clear increase over the past two decades, rising from 1856.1 km² in 2001 to 2202.6 km² in 2024, with the Montane belt contributing most of this growth. The grassland area increased from 7768.8 km² to 9640.5 km²; similarly, shrub-covered areas also increased slightly (Table 3).

However, tree-covered areas decreased from 65,314.9 km² in 2001 to 63,142.3 km² in 2024. Croplands showed a small reduction, shifting from 1277.0 km² to 1097.1 km², while permanent snow and glaciers area (8.2 km²) remained unchanged across both periods (Table 3). Inland water bodies and wetland-associated classes showed small fluctuations. Across all land cover categories, the Montane belt consistently represented the greatest share of mountain area in both years. The MGCI remained high, with total, green-covered mountain area changing only slightly from 118,716.8 km² (96.2%) in 2001 to 118,341.7 km² (95.8%) in 2024

Table 4. Area and percent of Mountain Green Cover Index (MGCI) [14] for the different bioclimatic belts of the mountains of the state of Colorado (CO), USA, for the reporting period 2024 and the baseline period 2001, using SEPAL [13]. Values in the parentheses are the proportion of MGCI for each bioclimatic belt.

Mountain Green Cover Index (MGCI)	Bioclimatic Belts				Total
	Alpine	Montane	Nival	Remaining Mountain Area
2024 (km2, %)	11,657.1 (87.0)	105,617.7 (97.3)	708.7 (65.7)	358.3 (82.7)	118,341.7 (95.8)
2001 (km2, %)	11,669.2 (87.0)	105,969.7 (97.6)	709.7 (65.8)	368.3 (85.0)	118,716.8 (96.2)
Change 2001–2024 (km2)	−12.1	−352.0	−1.0	−10.0	−375.1

Note: Mountain Green Cover Index (MGCI) is defined as the total area of “green cover” (cropland, grassland, forest, shrubland, and wetlands) within a mountain region in the reporting period n divided by the total mountain area (km²) and then multiplied by 100. Nival: areas with a growing season mean temperature < 3.5 °C and a growing-season length of < 10 days. Alpine: areas above the tree line with growing season mean temperatures < 6.4 °C and growing season length between 10 and 54 days. Montane: areas below the tree line with growing season mean temperatures between 6.4 °C and 15 °C. Remaining mountain areas: mountain regions with growing-season mean temperatures above 15 °C, including both frost and non-frost areas [41].

). Overall green cover declined by 0.4% between 2001 and 2024 (Table 4). Most of the decline in green cover was found in the remaining mountain area bioclimatic belt, which declined 2.7% Table 4). However, there was little change in the green cover for other bioclimatic regions. Overall, the results show broad stability in major land-cover categories alongside a measurable increase in artificial surfaces within the mountain region.

3.2. SEPAL Analysis of the Sub-Indicator 15.4.2b: Proportion of Degraded Mountain Land

The transition matrix between 2001 and 2024 was created based on the results of the SEPAL tool analysis [13]. Most of Colorado’s mountain region remained stable. Approximately 59,803 km² of tree-covered areas, 37,292 km² of shrub-covered areas, and 6793 km² of grassland remained unchanged. A smaller, but ecologically meaningful portion of the landscape exhibited improvement, represented by transitions from more disturbed classes to greener or more natural vegetation, such as 194.8 km² of cropland converting to shrubland and 9.1 km² of barren land transitioning to grassland, along with additional small increases from disturbed to vegetated classes, which together represent only a small portion of the total mountain area (Figure 5).

In contrast, degradation—loss of greener classes to more disturbed or unnatural classes—accounts for the most consequential changes. Major degradation includes 3620.4 km² of forest converted to shrub cover, along with significant conversions of natural vegetation to artificial surfaces, including 66.7 km² of tree-covered areas and 270.2 km² of shrub-covered areas converted to artificial surfaces (Figure 5). Overall, the degraded lands from 2001 to 2024 represent the most significant category of land cover change, indicating that although the area is predominantly stable, a significant portion of land has shifted to human-modified and structurally altered land cover types.

From 2001 to 2024, SEPAL [13] reported that 7061.2 km² (5.7%) of mountain land underwent degradation, for instance, forests being converted to development. After considering areas that recovered, such as shrubland converted to forest, the net degraded area was 3111.1 km², which is 2.5% of the total mountain area. Although there was little degradation in the Colorado mountain region, it is concentrated primarily in the Montane belt, where development pressure and land cover transitions are most significant.

3.3. GIS Analysis of the Overall Land and Soil Degradation Status of Colorado (CO), USA, Using Soil Spatial Data

As of 2024, CO has undergone an estimated 19% level of anthropogenic LD, primarily due to agriculture (80%) (Figure 6,

Table 5. Geospatial assessment of anthropogenic land degradation and potential land for nature-based solutions across soil orders in the state of Colorado (CO), USA, in 2024, reported only for areas with SSURGO [26] soil data available. Percent changes in area between 2001 and 2024 are reported in parentheses. Reported values have been rounded; therefore, calculated sums and percentages may exhibit minor discrepancies.

Soil Order	Area with Soil Data Available		Anthropogenically Degraded Land	Types of Anthropogenic Degradation			Potential Land for Nature-Based Solutions
				Barren	Developed	Agriculture
	(km2)	(%)	(km2)	(km2)	(km2)	(km2)	(km2)
Slightly Weathered Soils
	47,605.4	27.7	6222 (−2.7)	256 (+0.4)	1335 (+23.5)	4632 (−8.4)	28,854 (+2.4)
Entisols	36,455.1	21.2	5749 (−3.3)	78 (−0.5)	1235 (+24.1)	4437 (−9.0)	25,021 (+1.1)
Inceptisols	10,645.9	6.2	460 (+6.0)	177 (+0.8)	89 (+17.9)	195 (+6.1)	3659 (+11.2)
Histosols	504.4	0.3	13 (+3.5)	1(+0.8)	11 (+4.7)	1 (−8.7)	174 (+40.3)
Moderately Weathered Soils
	124,275.9	72.3	26,290 (+0.1)	48 (+13.9)	4992 (+23.3)	21,018 (−4.1)	69,889 (+1.4)
Alfisols	31,601.5	18.4	5551 (+3.1)	13 (+38.4)	825 (+21.7)	4713 (+0.3)	14,751 (+7.0)
Mollisols	60,857.5	35.4	16,183 (+0.8)	23 (+7.3)	2939 (+22.3)	13,220 (−3.0)	29,484 (−0.8)
Aridisols	30,262.1	17.6	4271 (−0.6)	11 (+5.4)	1174 (+26.9)	3085 (−14.5)	24,520 (+0.9)
Vertisols	1554.9	0.9	285 (+5.0)	0 (−5.1)	53 (+23.0)	232 (+1.6)	1134 (−2.3)
All Soils
Totals	171,881.3	100.0	32,513 (−0.4)	304 (+2.3)	6327 (+23.3)	25,882 (−4.9)	98,743 (+1.7)

Note: Entisols, Inceptisols, Alfisols, Mollisols, Vertisols, and Aridisols are mineral soils. Histosols are organic soils. Land that is considered anthropogenically degraded was calculated as the sum of degraded land from developed, agriculture, and barren land. Developed land includes categories: developed, open space; developed, low intensity; developed, medium intensity; developed, high intensity. Agriculture includes categories: cultivated crops; hay/pasture. Potential land for nature-based solutions (NBS) is assumed to include only the land cover classes of shrub/scrub, barren land, and herbaceous, to identify potential areas for NBS without altering current land uses. The change in land area was determined by the following equation: ((2024 Area − 2001 Area)/2001 Area) × 100%.

,

Table 6. The soil quality continuum was represented by the land use/land cover (LULC) and soil order areas within the state of Colorado (CO), USA, in 2024, reported only for areas with SSURGO [26] soil data available.

NLCD Land Cover Classes (LULC)	2024 Total Area by LULC (km2)	Degree of Weathering and Soil Development
		Slightly Weathered			Moderately Weathered
		Entisols	Inceptisols	Histosols	Aridisols	Vertisols	Alfisols	Mollisols
		2024 Area by Soil Order (km2)
Woody wetlands	1178.2	197.4	134.2	58.5	79.5	5.9	131.5	571.3
Shrub/Scrub	39,005.9	10,504.6	1873.2	122.0	8517.4	814.5	3291.8	13,882.4
Mixed forest	829.1	25.9	102.3	0.8	0.3	0.0	475.3	224.4
Deciduous forest	8618.4	210.0	383.1	6.8	41.8	42.3	1408.1	6526.3
Herbaceous	59,433.6	14,438.4	1608.7	51.3	15,991.2	319.6	11,446.0	15,578.4
Evergreen forest	28,041.3	4625.0	5966.8	209.9	1125.2	77.4	9232.7	6804.3
Emergent herbaceous wetlands	2262.1	704.8	117.5	41.8	236.1	9.9	64.6	1087.4
Hay/Pasture	1433.4	238.2	21.3	0.7	429.0	22.3	70.5	651.5
Cultivated crops	24,448.7	4198.5	173.3	0.0	2656.2	209.6	4642.3	12,568.9
Developed, open space	2899.7	487.8	43.4	5.4	429.7	25.1	514.2	1394.0
Developed, low intensity	2380.3	540.4	38.7	5.8	548.6	21.5	233.1	992.3
Developed, medium intensity	921.1	180.5	6.0	0.3	172.9	6.0	69.7	485.8
Developed, high intensity	125.8	26.0	0.5	0.0	23.0	0.8	8.3	67.1
Barren land	304.0	78.0	177.0	1.0	11.0	0.0	13.0	23.0
Totals	171,881.3	36,455.1	10,645.9	504.4	30,262.1	1554.9	31,601.5	60,857.5

Note: Entisols, Inceptisols, Aridisols, Vertisols, Alfisols, and Mollisols are mineral soils. Histosols are organic soils.

). Each of the seven soil orders was subject to different levels of anthropogenic LD as indicated in parentheses: Entisols (16%), Inceptisols (4%), Histosols (3%), Alfisols (18%), Mollisols (27%), Aridisols (14%), Vertisols (18%). Between 2001 and 2024, LD in CO decreased slightly (−0.4%); however, overall degradation remains a concern (Table 5). When considering only the LC types (barren land, shrub/scrub and herbaceous), 57% of CO has the potential for NBS (Table 5). However, 45% of the state soils are inherently low in soil quality (Entisols, Inceptisols and Aridisols), and 35% of the land has Mollisols, which are fertile agricultural soils, of which 22% are already degraded due to agricultural activities (Table 5) [32].

Almost 96% of CO is occupied by green cover (forest, shrubland, wetlands, herbaceous, hay/pasture and cultivated crops), which is often found on Mollisols (Table 6, Figure 6). From 2001 to 2024, anthropogenic LD across soil types led to decreases in land cover types, including deciduous forest, evergreen forest, and mixed forest, and a rapid increase in development in CO (

Table 7. Land use/land cover (LULC) changes between 2001 and 2024, disaggregated by soil order, within the state of Colorado (CO), USA, reported only for areas with SSURGO [26] soil data available.

NLCD Land Cover Classes (LULC)	Change in Area, 2001–2024 (%)	Degree of Weathering and Soil Development
		Slightly Weathered			Moderately Weathered
		Entisols	Inceptisols	Histosols	Aridisols	Vertisols	Alfisols	Mollisols
		Change in Area, 2001–2024 (%)
Woody wetlands	−2.0	−5.4	−0.8	−2.9	−0.4	−4.3	−1.7	−1.2
Shrub/Scrub	+0.6	−0.6	+2.7	+13.6	+0.2	−2.0	+21.4	−2.4
Mixed forest	−18.2	−11.3	−16.6	−25.9	−40.3	−44.4	−20.1	−15.1
Deciduous forest	−2.1	−7.4	−5.2	+2.6	+3.2	+18.2	−5.3	−1.1
Herbaceous	+2.4	+2.4	+24.5	+225.3	+1.3	−3.1	+3.5	+0.7
Evergreen forest	−4.0	−1.1	−5.6	−19.2	+4.7	+9.8	−9.2	+2.7
Emergent herbaceous wetlands	+2.6	+2.1	+2.6	+3.5	−0.1	+3.7	+9.9	+3.2
Hay/Pasture	−10.4	−5.4	−14.4	−8.9	−7.7	−18.5	+1.1	−14.4
Cultivated crops	−4.6	−9.2	+9.4	0.0	−15.5	+4.4	+0.3	−2.3
Developed, open space	+6.5	+4.6	+12.7	+4.5	+7.2	+11.7	+8.4	+6.0
Developed, low intensity	+36.2	+34.2	+23.4	+8.0	+36.5	+38.5	+45.2	+35.8
Developed, medium intensity	+57.8	+66.1	+19.9	−35.3	+58.6	+23.3	+82.5	+53.0
Developed, high intensity	+69.0	+53.9	+76.4	−42.8	+74.8	+44.7	+74.1	+73.3
Barren land	+2.3	−0.5	+0.8	+0.8	+5.4	−5.1	+38.4	+7.3

Note: Entisols, Inceptisols, Aridisols, Vertisols, Alfisols, and Mollisols are mineral soils. Histosols are organic soils. Change in the area was calculated using the following equation: ((2024 Area − 2001 Area)/2001 Area) × 100%.

).

A total of 6327 km² of the land area was converted to developments before and through 2024, resulting in midpoint C loss of 1.3 × 10¹¹ kg of C and associated midpoint social cost of CO₂ (SC-CO₂) of $23B ($ = USD = US dollars) (Table S3). Among the affected soil types, Mollisols were the most impacted. These soils, known for their fertile topsoil and high organic matter content [32], have significant potential for carbon storage (Table 6). Table 7 shows an increase in development across almost all soil types between 2001 and 2024, resulting in a midpoint C loss of 2.5 × 10¹⁰ kg of C and an associated midpoint social cost of CO₂ (SC-CO₂) of $4.4B (USD) (Table S4).

3.4. GIS Analysis of Sub-Indicator 15.4.2a: Mountain Green Cover Index as Applied to the State of Colorado (CO), USA

Sub-indicator 15.4.2a is based on monitoring changes in the extent of green cover in mountainous regions. Land cover is used to categorize land into areas with green and non-green cover. Green cover includes areas covered by both natural vegetation and vegetation resulting from anthropogenic activities, such as woody wetlands, shrub/scrub, mixed forest, deciduous forest, herbaceous vegetation, evergreen forest, emergent herbaceous wetlands, hay/pasture, and cultivated crops. Non-green areas include non-vegetated areas such as barren land, developed open spaces, and areas with low, medium, and high development intensities. For the year 2024, the results from this study showed that green cover occupied 120,746.8 km² (95.9%) of the area (

Table 8. Land use/land cover (LULC) for 2024 and changes between 2001 and 2024 for the mountains in the state of Colorado (CO), USA.

NLCD Land Cover Classes (LULC)	Total Mountain Area
	2024 Area by LULC (km2)	Change (2001–2024) Area (km2)	Change (2001–2024) (%)
Open water	390.0	+16.4	+4.4
Snow	8.2	0.0	0.0
Woody wetlands	1698.7	−19.6	−1.1
Shrub/Scrub	42,850.6	+67.1	+0.2
Mixed forest	1438.3	−316.3	−18.0
Deciduous forest	14,096.3	−412.9	−2.8
Herbaceous	10,094.7	+2319.1	+29.8
Evergreen forest	48,200.6	−1925.1	−3.8
Emergent herbaceous wetlands	1304.9	+26.3	+2.1
Hay/Pasture	874.8	−142.4	−14.0
Cultivated crops	187.9	−66.8	−26.2
Developed, open space	1142.0	+156.6	+15.9
Developed, low intensity	934.9	+223.6	+31.4
Developed, medium intensity	155.0	+45.8	+41.9
Developed, high intensity	13.0	+4.7	+57.1
Barren land	2554.8	+23.4	+0.9
Totals	125,944.8	-	-

). The largest area covered was evergreen forest, i.e., 48,200.6 km², and the most significant loss between 2001 and 2024 was also from evergreen forest, i.e., 1925.1 km² (Table 8). The results for the change between 2001 and 2024 showed that the MGCI decreased by −0.4% and almost all green cover types faced significant losses, such as woody wetlands (−1.1%), mixed forest (−18.0%), deciduous forest (−2.8%), evergreen forest (−3.8%), hay/pasture (−14%) and cultivated crops (−26.2%) (Table 8). Conversely, the area occupied by non-green spaces increased rapidly, with the largest increases observed in developed high-intensity (+57.1%), developed medium-intensity (+41.9%), developed low-intensity (+31.4%), and developed open-space (+15.9%) areas, simultaneously (Table 8).

3.5. GIS Analysis of Sub-Indicator 15.4.2b: Proportion of Degraded Mountain Land as Applied to the State of Colorado (CO), USA

The land cover change matrix helps identify areas that have degraded, improved, or remained stable, which is important for monitoring biodiversity. This study adopted the approach of Helfenstein et al. (2022) [46] for the land cover transition matrix. They used a Swiss case study to track agricultural-related development.

Our study adapted Helfenstein et al.’s (2022) [46] land cover change matrix for the mountainous region of CO between 2001 and 2024 to understand the area of different land cover transitions from one type to another (Figure 7). Areas exhibiting no land cover change between 2001 and 2024 are represented in gray, whereas land cover changes are displayed using an orange gradient scaled to the relative magnitude of change. Conversion to agricultural or developed land cover classes is classified as a biodiversity loss [47].

Sub-indicator 15.4.2b, which is the proportion of mountain land that is degraded, was used to assess the spatial extent of land-cover-driven degradation in mountainous regions of CO. The total extent of anthropogenically degraded land was computed as the combined area of degraded land resulting from agriculture, development, and barren land. Land considered as developed is categorized by the following land cover types: developed, open space; developed, medium intensity; developed, low intensity; and developed, high intensity. Agriculture encompasses two primary categories: cultivated crops and hay/pasture. Up through 2024, the CO mountainous region has experienced 4.7% anthropogenic LD, primarily from development expansion (38.3%) (

Table 9. Assessment of anthropogenic mountain land degradation and potential land for nature-based solutions in the state of Colorado (CO), USA, in 2024. Percent changes in area between 2001 and 2024 are reported in parentheses. Reported values have been rounded; therefore, calculated sums and percentages may exhibit minor discrepancies.

Total Mountain Area	Anthropogenically Degraded Land	Types of Anthropogenic Degradation			Potential Land for Nature-Based Solutions
		Barren	Developed	Agriculture
(km2)	(km2)	(km2)	(km2)	(km2)	(km2)
125,944.8	5862.4 (+4.4)	2554.8 (+0.9)	2245.0 (+23.7)	1062.6 (−16.4)	55,500.1 (+4.5)

Note: Land that is considered anthropogenically degraded was calculated as the sum of degraded land from developed, agriculture, and barren land. Developed land includes categories: developed, open space; developed, low intensity; developed, medium intensity; developed, high intensity. Agriculture includes categories: cultivated crops; hay/pasture. Potential land for nature-based solutions (NBS) is assumed to include only the land cover classes of shrub/scrub, barren land, and herbaceous, to identify potential areas for NBS without altering current land uses. The change in land area was determined by the following equation: ((2024 Area − 2001 Area)/2001 Area) × 100%.

). In the period from 2001 to 2024, the proportion of LD increased by 4.4% (Table 9). Among the LC types (barren land, shrub/scrub and herbaceous), 44.1% of the mountainous regions have the potential for NBS.

Transition matrices for land cover provide a quantitative measurement of change between different classes, and to do this, medium resolution MRLC land cover maps [15] are able to account for this change in land cover with a high relative accuracy, which can also be further visualized through high-resolution aerial imagery. For instance, in the mountain region of CO (USA), Figure 8 shows a location where shrub/forest transitioned to new housing developments. The high-resolution aerial images (Figure 8a,b) show where this land was changed to housing developments sometime between 2009 and 2023. For the period 2024, most development occurred in Mesa County, CO, which has 7123.1 km² of mountain area (82.3%), and has the largest development area, i.e., 141.03 km², with the highest development occurring in Entisols (63.93 km²), Aridisols (52.23 km²), and Mollisols (19.0 km²). In second place, Jefferson County, with 1458.4 km² of mountain area (72.7%), has a development area of 118.81 km², with the highest development occurring in the Mollisols (i.e., 84.61 km² area). Similarly, for the period between 2001 and 2024, Garfield County showed the greatest increase in development (53.2%), followed by Chaffee County (42.8%).

3.6. Enhancing Sub-Indicator 15.4.2a: Mountain Green Cover Index Using the State of Colorado (CO), USA as an Example

Sub-indicator 15.4.2a is based on mountain-area monitoring changes in the extent of green cover (cropland, grassland, wetland, tree-covered and shrub-covered areas). Land cover data is used to classify land areas into green and non-green categories. Green cover encompasses areas that are both naturally vegetated and those resulting from anthropogenic activities. Non-green areas include non-vegetated areas such as barren land and development areas.

For the year 2024, the total area of the mountainous region of CO disaggregated by soil type was 66,930 km², with a diverse mix of land uses and soil types. Mollisols (27,870.3 km²) and Alfisols (14,246.0 km²) were the most widespread soils, together comprising more than 60% of the total area. Entisols (9873.0 km²) and Inceptisols (9579.3 km²) were also common, while Aridisols (4319.9 km²), Vertisols (540.1 km²), and Histosols (502.0 km²) had smaller extents (

Table 10. The soil quality continuum in the mountains of the state of Colorado (CO), USA, by land use/land cover (LULC) and soil order areas in 2024, reported only for areas with SSURGO [26] soil data available.

NLCD Land Cover Classes (LULC)	2024 Total Area by LULC (km2)	Degree of Soil Development and Weathering
		Slightly Weathered			Moderately Weathered
		Entisols	Inceptisols	Histosols	Aridisols	Vertisols	Alfisols	Mollisols
		2024 Area by Soil Order (km2)
Woody wetlands	948.2	104.1	131.7	58.4	26.8	2.8	124.4	499.9
Shrub/Scrub	23,413.5	4908.1	1687.6	120.5	2815.9	391.2	2453.7	11,036.6
Mixed forest	824.8	25.4	102.1	0.8	0.3	0.0	473.0	223.1
Deciduous forest	8258.9	195.9	379.9	6.8	39.4	35.0	1382.8	6219.1
Herbaceous	4827.7	438.4	1137.0	51.1	177.2	17.6	905.2	2101.2
Evergreen forest	25,607.4	3758.9	5808.7	209.7	907.8	64.8	8612.4	6245.1
Emergent herbaceous wetlands	781.2	63.5	89.9	41.6	17.0	1.7	28.2	539.3
Hay/Pasture	488.5	66.2	5.0	0.6	121.2	9.6	17.3	268.5
Cultivated crops	130.2	39.3	0.4	0.0	42.6	3.4	27.2	17.4
Developed, open space	710.7	84.4	29.0	5.4	66.5	7.6	130.5	387.3
Developed, low intensity	566.2	106.0	28.6	5.6	83.5	5.3	74.2	263.1
Developed, medium intensity	99.1	24.8	2.9	0.3	18.5	0.8	5.1	46.7
Developed, high intensity	9.1	3.0	0.1	0.0	2.0	0.1	0.1	3.7
Barren land	264.9	55.0	176.5	1.1	1.2	0.0	11.9	19.3
Totals	66,930.4	9873.0	9579.3	502.0	4319.9	540.1	14,246.0	27,870.3

Note: Entisols, Inceptisols, Aridisols, Vertisols, Alfisols, and Mollisols are mineral soils. Histosols are organic soils.

).

The main land cover types were evergreen forest (25,607.4 km²) and shrub/scrub (24,313.5 km²), both of which were mainly distributed on Entisols and Mollisols, respectively, indicating that these soils support dense and stable vegetation (Table 10). In contrast, developed areas covered smaller extents and were mostly concentrated on Mollisols. However, these areas increased rapidly across all soil types during the period from 2001 to 2024. Similarly, all forest types declined within that period, and for deciduous forests, all soil types show a decrease in area. Moreover, cultivated and pasture land declined in area across all soil types, particularly in Mollisols (

Table 11. Land use/land cover (LULC) changes between 2001 and 2024 by soil order for the mountainous region in the state of Colorado (CO), USA, reported only for areas with SSURGO [26] soil data available.

NLCD Land Cover Classes (LULC)	Change in Area, 2001–2024, (%)	Degree of Soil Development and Weathering
		Slightly Weathered			Moderately Weathered
		Entisols	Inceptisols	Histosols	Aridisols	Vertisols	Alfisols	Mollisols
		Change in Area, 2001–2024 (%)
Woody wetlands	−1.3	−1.0	−0.7	−2.8	−2.1	−1.2	−1.2	−1.3
Shrub/Scrub	+0.8	−0.5	+3.2	+13.7	−1.1	−4.5	+28.0	−3.0
Mixed forest	−18.1	−11.2	−16.5	−25.9	−40.5	−44.4	−20.1	−15.1
Deciduous forest	−2.0	−6.9	−4.9	+2.9	+3.1	+16.6	−5.0	−1.1
Herbaceous	+34.0	+23.6	+43.3	+223.7	−6.1	+80.4	+156.1	+11.3
Evergreen forest	−4.5	−1.5	−5.9	−19.2	+3.7	+8.2	−9.7	+2.5
Emergent herbaceous wetlands	+2.3	+0.8	+1.3	+3.5	+4.4	+0.1	+11.8	+2.0
Hay/Pasture	−12.9	−8.3	−4.7	−9.7	−9.9	−15.7	−2.3	−15.8
Cultivated crops	−29.5	−26.6	−16.8	0.0	−27.2	−27.9	−29.2	−40.3
Developed, open space	+15.6	+15.6	+15.7	+4.7	+18.7	+20.8	+9.6	+17.3
Developed, low intensity	+30.5	+21.0	+19.2	+7.9	+31.1	+72.3	+30.3	+36.0
Developed, medium intensity	+43.6	+54.5	−0.4	−35.0	+61.1	+47.4	+63.3	+35.5
Developed, high intensity	+51.6	+49.2	+47.4	−42.8	+45.4	+67.9	+101.2	+55.8
Barren land	+1.9	−1.6	+0.9	+0.8	+3.0	+34.0	+37.5	+5.2

Note: Entisols, Inceptisols, Aridisols, Vertisols, Alfisols, and Mollisols are mineral soils. Histosols are organic soils. Change in the area was calculated using the following equation: ((2024 Area − 2001 Area)/2001 Area) × 100%.

).

3.7. Enhancing Sub-Indicator 15.4.2b: Proportion of Degraded Mountain Land Using the State of Colorado (CO), USA as an Example

For effective biodiversity management, incorporating soil adds important context and information. For example, soil data can be used to calculate soil C-regulating services in the CO mountains (Table S5). Analysis of land cover change over time helps identify areas with degraded soil resources. This study adapted the land cover change matrix developed by Helfenstein et al. (2022) [46] and applied it across all seven soil orders (Figures S1–S6). Alfisols, a soil order of high agricultural relevance in the mountainous region of Colorado, USA, were used as a case study to illustrate the effects of land cover transitions on soil resources between 2001 and 2024 (Figure 9). Gray cells represent land cover classes that remained stable over the study period, whereas land cover changes are depicted in orange, with increasing color intensity corresponding to greater proportional area change. Transition to development or agricultural land cover is considered an overall loss of biodiversity in both above-ground and below-ground areas [47]. Examining land cover changes by soil type (i.e., soil order) provides additional information on how different soils are affected, as soil properties and carbon levels can vary significantly (Table S5). This also helps identify which soils are more susceptible to losing their health or carbon when the land is converted to agriculture, development, or other intensive uses (Tables S6 and S7).

The proportion of mountain degradation is relevant to sub-indicator 15.4.2b and target 15.4: Ensure the conservation of mountain ecosystems, including their biodiversity, in order to enhance their capacity to provide benefits that are essential for sustainable development. According to the SDG indicator metadata [14], mountain degradation was assessed based on anthropogenic LC changes, where degradation includes developed areas (developed, low intensity; developed, medium intensity; developed, high intensity; developed, open space), agricultural (cultivated crops and hay/pasture), and barren lands. This study first analyzed the extent of degradation across the entire state to understand the general trends in LC status by soil type, indicating the state’s overall degradation status (Table 5). Although the overall state showed a decline in LD, when disaggregated by mountainous areas, it revealed an increase in LD (

Table 12. Assessment of anthropogenic land degradation and potential land for nature-based solutions across soil orders in the mountains of the state of Colorado, USA, in 2024, reported only for areas with SSURGO [26] soil data available. Percent changes in area between 2001 and 2024 are reported in parentheses. Reported values have been rounded; therefore, calculated sums and percentages may exhibit minor discrepancies.

Soil Order	Area with Soil Data Available		Anthropogenically Degraded Land	Types of Anthropogenic Degradation			Potential Land for Nature-Based Solutions
				Barren	Developed	Agriculture
	(km2)	(%)	(km2)	(km2)	(km2)	(km2)	(km2)
Slightly Weathered Soils
	19,954.2	29.8	634 (+4.7)	233 (+0.3)	290 (+20.1)	112 (−15.6)	8575 (+6.2)
Entisols	9873.0	14.8	379 (+5.1)	55 (−1.6)	218 (+22.2)	105 (−16.1)	5401 (+1.0)
Inceptisols	9579.3	14.3	243 (+4.2)	176 (+0.9)	16 (+16.5)	5 (−5.7)	3001 (+15.2)
Histosols	502.0	0.8	13 (+3.6)	1 (+0.8)	11 (+4.7)	1 (−9.5)	173 (+40.6)
Moderately Weathered Soils
	46,976.1	70.2	1634 (+7.3)	32 (+15.1)	1095 (+24.1)	507 (−19.4)	19,931 (+4.8)
Alfisols	14,246.0	21.3	266 (+9.2)	12 (+37.5)	210 (+17.1)	44 (−20.7)	3371 (+47.9)
Mollisols	27,870.3	41.6	1006 (+8.5)	19 (+5.2)	701 (+25.1)	286 (−17.9)	13,157 (−1.0)
Aridisols	4319.9	6.5	335 (+2.7)	1 (+3.0)	170 (+28.6)	164 (−15.2)	2994 (−1.4)
Vertisols	540.1	0.8	27 (+2.7)	0 (+34.0)	14 (+38.3)	13 (−19.3)	409 (−2.5)
All Soils
Totals	66,930.4	100.0	2269 (+6.6)	265 (+1.9)	1385 (+23.3)	619 (−17.0)	28,506 (+5.2)

Note: Entisols, Inceptisols, Alfisols, Mollisols, Vertisols, and Aridisols are mineral soils. Histosols are organic soils. Land that is considered anthropogenically degraded was calculated as the sum of degraded land from developed, agricultural, and barren land. Developed land includes categories: developed, open space; developed, low intensity; developed, medium intensity; developed, high intensity. Agriculture includes categories: cultivated crops; hay/pasture. Potential land for nature-based solutions (NBS) is assumed to include only the land cover classes of shrub/scrub, barren land, and herbaceous, to identify potential areas for NBS without altering current land uses. The change in land area was determined by the following equation: ((2024 Area − 2001 Area)/2001 Area) × 100%.

). Between 2001 and 2024, all types of development increased (1385 km²) across various soil types in the CO mountains, resulting in a midpoint C loss of 5.5 × 10⁹ kg of C and an associated midpoint SC-CO₂ of $1B (Table S7). Similarly, pasture/hay and cultivated crops have declined for all soil types (Table 12).

As of 2024, CO mountainous regions have experienced 3.4% anthropogenic LD, mainly in developed areas (61%) (Table 10). The second most significant factor is agriculture (27%). The greatest impact occurred on Mollisols (1006 km²), where agriculture and development were the main causes. Entisols (379 km²) and Inceptisols (243 km²) were moderately affected, while Histosols and Vertisols showed only minor disturbance. Areas with high potential for nature-based restoration cover approximately 28,506 km², primarily found on Mollisols, Entisols, and Alfisols, indicating where recovery and conservation efforts could be most effective in the mountain landscape.

4. Discussion

4.1. Limitations of MGCI and Enhancing the MGCI Definition for the United Nations (UN) MGCI Analysis

The MGCI aims to monitor land cover changes in mountain regions by quantifying the land area covered by green vegetation. This metric is a critical sub-indicator for evaluating the mountain ecosystem, as the presence and distribution of green cover reflect the overall ecological stability of these areas [10]. Therefore, monitoring these green areas over time is crucial for understanding the condition of mountain ecosystems and supports conservation strategies. The current method used by UN-FAO for monitoring MGCI uses the European Space Agency Climate Change Initiative Land Cover (ESA-CCI-LC) product at a resolution of 300 m for identifying and aggregating areas with green coverage (forest, shrub, cropland, etc.) on the country scale and then computing the ratio of green mountain coverage to the total mountain area [14]. However, this method fails to detect the finer-scale variability of MGCI within the country. Similarly, the current sub-indicator 15.4.2a focuses solely on MGCI by land cover type, without disaggregating land cover areas by soil type, thereby overestimating the health of degraded areas with low soil quality. Assessing the health of the mountain ecosystem solely on the basis of the LC assumes that all green vegetation provides the same ecosystem services (ESs) (e.g., carbon storage). Based on the MGCI sub-indicator, all artificial and natural vegetation is considered green coverage. For example, cropland does not provide the same ESs as grassland does, and ESs can also vary depending on soil type [50,51]. Therefore, future studies should focus on distinguishing between natural and artificial green cover and on exploring the varying ESs provided by different green cover types across diverse soil types to more accurately monitor the condition of mountain ecosystems.

This study proposes incorporating soil types into the MGCI analysis by disaggregating LCs by soil type (

Table 13. Mountain Green Cover Index (MGCI) [14] by soil order within the state of Colorado (CO), USA, in 2024 for mountains with available soil spatial data.

Land Type	Degree of Weathering and Soil Development
	Slightly Weathered			Moderately Weathered
	Entisols	Inceptisols	Histosols	Aridisols	Vertisols	Alfisols	Mollisols
	Mountain Green Cover Index (MGCI) (km2, %)
Mountains (with soil data)	9599.8	9342.1	489.6	4148.2	526.3	14,024.2	27,150.2
	14.7%	14.3%	0.8%	6.4%	0.8%	21.5%	41.5%

Note: Entisols, Inceptisols, Aridisols, Vertisols, Alfisols, and Mollisols are mineral soils. Histosols are organic soils.

), which showed highest MGCI for the Mollisols and Alfisols which are the most productive soils. The evaluation of CO data showed that green cover in different soil types was degraded, despite overall green cover increasing. For example, woody wetlands decreased across the state and in the mountainous region for all soil types. This was higher in Histosols, which are considered wetland soils and have greater potential as carbon sinks; this shift could lead to reduced long-term carbon storage capacity and increased carbon emissions, as drained or degraded Histosols release stored carbon into the atmosphere. Therefore, not accounting for soil types led to an overestimation of the potential NBS area, as green cover in low SQ areas is likely unsuitable for NBS.

4.2. Limitations and Enhancing the Sub-Indicator “Proportion of Degraded Mountain Land” in the United Nations (UN) Mountain Land Degradation Analysis

The proportion of degraded mountain land is an important metric for evaluating the extent of LD resulting from changes in land cover within a specified country and reporting period. This sub-indicator aligns with the sub-indicator “trends in land cover” under SDG Indicator 15.3.1, providing a systematic approach to monitoring landscape transformations and their implications for ecosystem stability [52]. While LD is being studied globally, a gap remains in the research, particularly for mountain ecosystems, which are unique and heterogeneous landscapes varying in altitude; therefore, using remote sensing techniques is crucial for effective monitoring [53], and monitoring LD over time supports conservation and land management strategies. This sub-indicator only includes LC for the mountain LD analysis. However, adopting a conceptual framework based on land degradation neutrality (LDN) incorporates LC, land productivity, and soil organic carbon stocks. Adding this sub-indicator provides an accurate estimation of the LD [54]. Similarly, another limitation is that the current mountain LD sub-indicator, UN SDG 15.4.2b, focuses on the total degraded mountain land without disaggregation by soil type, which can mask the actual mountain LD, because it is difficult to identify which areas are degrading and easy to miss areas that need urgent conservation or restoration.

This study proposes incorporating inherent soil quality (SQ) into LD assessment by stratifying the analysis by soil type. Application of this approach to CO data revealed that the estimated LD nearly doubled when inherent soil quality was incorporated. Evaluating the status and change in mountain LD can be enhanced by incorporating soil information (Figure 10). Separating inherent from anthropogenic LD in these systems is important because inherent degradation can help mask human impact. Soil data includes soil type, which can identify weathered or poorly developed soil areas with low soil quality (SQ) (Figure 10) that are inherently degraded and have a lower potential to support net primary productivity (NPP). Low-quality soil often exhibits higher soil salinity, lower soil moisture, and lower vegetation density, which are the most significant natural factors contributing to the decline in soil quality and reduction in NPP [55,56]. This contrasts with soils such as Alfisols and Mollisols, which have high potential to support vegetation. Mountain areas where spatial soil data are not available are often dominated by bedrock outcrops and steep slopes, with pockets of shallow soils [57]. Steep slopes and a lack of soil make it difficult to survey these areas effectively. These areas, which are often not surveyed, can be considered inherently degraded and analogous to low-SQ soil areas (Figure 10). Identifying inherently degraded mountain areas is critical because it enables the evaluation of past and future anthropogenic changes through the lens of soil resources, allowing for the identification of where critical, highly productive soil resources are lost or reduced due to development or new agricultural production. Additionally, using this method to calculate total mountain LD can provide a more realistic view of mountain LD status and identify important areas that should be protected from future anthropogenic change. For example, in the CO mountains, the historical LD (prior to and until 2024) accounts for approximately 4.7% of the total mountain area, based on land cover in accordance with the current UN SDG indicator 15.4.2 concepts. In contrast, inherent soil degradation accounts for 35.5% of the mountain area, while total LD (inherent + anthropogenic) accounts for 38.9% of the mountain area. Including inherently degraded soils provides a better understanding of the overall LD, which is important because it can help understand the true impact of future anthropogenic LD on mountain resources. Standard LD analyses, including SEPAL, do not account for inherent soil degradation and therefore underestimate the overall mountain LD.

Due to widespread anthropogenic LD and the complexity of changing these trends, restoring the degraded land to its near-natural condition is essential [58] and adopting NBS may be the most feasible and cost-effective solution to combat LD [59]. There are several applications of NBS for land restoration, such as natural regeneration, afforestation, or reforestation, that enhance soil quality, increase carbon sequestration, provide habitat for wildlife, support biodiversity conservation, enhance ecosystem services, mitigate climate change and improve the livelihood of local communities [59,60]. However, failing to account for inherent SQ could lead to overestimating the potential NBS area, as low-SQ areas are likely unsuitable for NBS. For example, a large proportion of Entisols and Inceptisols are often inherently low-SQ soils. In the CO mountains, the total potential area for NBS is higher, i.e., 42.6%, without considering the inherent soil quality. However, given the soil quality, 83.4% of the land is of low quality. Therefore, the actual land suitable for NBS is approximately 7% of the total mountain area. Furthermore, many mountain regions are facing rapid climate change, accelerating land-use expansion and population growth, and there will be an even larger increase in climate change and land use change in the near future, as compared to the lowlands [61] which results in the shrinking and degradation of natural ecosystems and a reduction in available land for NBS.

This study highlights the need to incorporate soil inorganic carbon (SIC) and total soil C (TSC) to accurately estimate the potential damage from LD (e.g., caused by land development). However, indicator 15.3.1 only suggests the SOC for LD analysis, which can underestimate actual losses and damages (e.g., GHG emissions) from LD. Entisols, Inceptisols, and Aridisols are mineral soils and can contain considerable amounts of SIC. The conversion of this land could lead to increased greenhouse gas emissions. Therefore, this study determines the actual mountain land that is potentially degraded and the potential land for NBS. Mountains are complex and integrated systems, and a single indicator will not be sufficient for the sustainability of mountainous regions. Therefore, future studies should incorporate additional indicators to improve the monitoring of mountain LD, such as vegetation structure and composition, invasive species, and biomass.

4.3. Significance of Results in a Broader Context

This research focuses on SDG target 15.4, which emphasizes the conservation of mountain ecosystems and their biodiversity. The results of this study were linked with other relevant UN initiatives, such as the SDGs, adopted in 2015 [8], the UN Convention on Biological Diversity [62], the UN Convention to Combat Desertification [63], and the UN Kunming Montreal Global Biodiversity Framework [64], to see the importance of SDG target 15.4 in a broader context. As advised by the UN, which recommends disaggregating indicators when feasible, this study was conducted using CO as an example, which integrates soil and land use relationships for better land management to fulfill the UN SDGs [8].

• Within the period of 2001 to 2024, there was a decline in hay/pasture and cultivated crops overall and in the mountain areas of CO. In the mountain areas, the results showed reductions in hay/pasture (−12.9%) and cultivated crops (−29.5%) across all soil orders (Table 11). This indicates a decrease in the land available for agricultural use, resulting in an overall decline in food production (relevant to UN SDG 2: Zero Hunger).

• Regarding the historical mountain LD until 2024, development has occurred throughout the soil orders, with the highest area in Mollisols, which are the most agriculturally important soils for food production (Table 10). Similarly, between 2001 and 2024, development occurred in other agriculturally important soils, mainly in Alfisols and Vertisols (Table 11). Developments have occurred at the cost of deciduous forests (−2.0%), evergreen forests (−4.5%), and mixed forests (−18.1%) (Table 11) (relevant for UN SDG 12: Responsible Consumption and Production).

• The state of CO completed and released comprehensive climate change action plans in 2014 to improve the state’s ability to adapt to future climate change (https://www.georgetownclimate.org/adaptation/plans.html (accessed on 7 March 2024) [65]. Still, emissions continue to rise due to LD and increased development activities. The results from this study can help policymakers revise the climate change action plan. Continuous LD in CO degrades the dynamic quality of soil (soil health) and increases carbon dioxide (CO₂) emissions from soil, contributing to climate change. Developments in the mountainous area of CO prior and through 2024 were the major sources of LD damages, with 1385.1 km² developed, causing midpoint losses of 2.9 × 10¹⁰ kg of total soil carbon (TSC) with a midpoint social cost of carbon dioxide emissions (SC-CO₂) of $5.3B (where B = billion = 10⁹, USD) (Table S6). In the period from 2001 to 2024, the recently developed land area increased (+23.3%), resulting in a midpoint loss of 5.5 × 10⁹ kg of TSC and a resultant midpoint value of $1.0B in SC-CO₂ (Table S7). The total potential land for C sequestration using nature-based solutions is 42.6%. Entisols, Inceptisols and Aridisols occupy 83.4% of the state, and these are inherently low-quality soils. This study uses the monetary damage estimates derived from fixed (non-market) and theoretical SC-CO₂ values, which are not directly assessed as fines or damages from responsible parties. (Addressing UN SDG 13: Climate Action);

• As of 2024, overall, the state of CO has faced 19% anthropogenic LD (mostly associated with agriculture, 80%). In 2024, the mountain area had 3.4% anthropogenic LD, due to development in the mountainous region (61%). All seven soil orders had various levels of inherent and anthropogenic LD. Increases in soil degradation and LD between 2001 and 2024 (+6.6%) indicated a degrading land status in the mountainous region (Table 12). The recent developments from 2001 to 2024 resulted in a +6.6% increase in anthropogenic LD and a substantial +23.3% increase in LD in developed land cover types. Although the CO mountain area has 42.6% of its land potential for NBS, 35.5% of this land is most likely unsuitable due to poor soil quality (e.g., Entisols, Inceptisols and Aridisols). Mountain land cover changes driven by development between 2001 and 2024 have decreased the availability of valuable soil resources throughout the state. Additionally, there have been significant declines in key land covers essential for atmospheric pollution reduction and carbon sequestration, including woody wetlands (−1.3), deciduous forests (−2.0%), evergreen forests (−4.5%), mixed forests (−18.1), hay/pasture (−12.9%) and cultivated crops (−29.5%) (Table 11). (Addressing UN SDG 15: Life on Land; UN Convention on Biological Diversity; UN Convention to Combat Desertification; UN Kunming–Montreal Global Biodiversity Framework);

• Global efforts are increasingly directed towards preserving ecosystem resilience and integrity, which are at risk from LD. Considering this, the UN’s 15th meeting of the Conference of the Parties (COP15) adopted the UN Kunming–Montreal Global Biodiversity Framework [64]. The principal objective of this framework (Goal A) was to maintain, enhance, and restore ecosystem resilience, integrity, and connectivity, and included a target (Target 11) to maintain, restore, and enhance ecosystem functions and services (e.g., soil health, air, water and climate regulation). This study demonstrated that CO did not achieve land degradation neutrality (LDN) between 2001 and 2024, as development occurred across all soil orders, including agriculturally important Alfisols and Mollisols, as well as C-rich Histosols. New development construction is likely to cause biodiversity loss by reducing pedodiversity (soil diversity). The analytical approaches and data interpretation techniques applied in this study can be instrumental in advancing Target 21 of the framework, which aims to enhance biodiversity management. By generating spatial data that provides a deeper understanding of how changes in land cover impact biodiversity, these methods can contribute to improved resource management and more effective conservation strategies. (Relevant to the UN Kunming–Montreal Global Biodiversity Framework).

• The Revised World Soil Charter 2015 [66], which was endorsed by Food and Agriculture Organization (FAO) member states, highlights the pressing need to limit soil degradation to safeguard critical soil ecosystem services and support efforts to achieve LDN. This charter provides clear guidelines to ensure that “soils are managed sustainably and that degraded soils are rehabilitated or restored [67]”. Our research demonstrates that mountainous areas of CO have experienced a significant increase in soil degradation and LD, with a notable +6.6% rise in LD between 2001 and 2024, as indicated in Table 12. This degradation has affected all soil types across the state, primarily associated with the growth of development-related land use during the study period. The findings reveal that CO has not achieved LDN, as evidenced by the data presented. (Relevant to the Revised World Soil Charter).

5. Conclusions

The SDG sub-indicators 15.4.2a (Mountain green cover index) and 15.4.2b (Proportion of degraded mountain areas) do not account for soil diversity when monitoring mountain ecosystems. This study proposed to enhance these sub-indicators with soil- and land-related data and measures to monitor mountain ecosystems. The novelty of this study lies in its use of soil and satellite data to track mountain greenness and degradation, a method that can be replicated worldwide. This study used the state of Colorado (CO), USA, as an example, demonstrating that it has experienced significant LD. The integration of soil data enriched LD analysis in mountain areas. Without disaggregating by soil type, the mountain areas showed a very small proportion of land degraded; however, the state has experienced considerable LD and has a very low potential area for NBS. This study reveals that, as mountains are diverse and fragile ecosystems, LULC alone is insufficient for tracking and monitoring large-scale mountain ecosystems; incorporating soil data and using medium-resolution remote sensing data is necessary. Land cover and soil data are global, and there are global datasets—such as the CORINE Land Cover (CLC) [68], Global 2000–2020 Land Cover and Land Use Change Dataset (GLAD) [69], ESA-CCI-LC [35], the Harmonized World Soil Database (HWSD) [70], SoilGrids [71], and local national-level datasets such as NLCD [15] and SSURGO [26]—that can be used to apply the methodology from this study to other countries and other US states. Conversely, the applicability of this methodology is restricted to mountain ecosystems, as land cover and soil–vegetation interactions are strongly influenced by terrain gradients (elevation, slope, etc.), making green cover a reliable sub-indicator of ecosystem condition. However, in low-relief or heavily human-altered ecosystems, these natural controls are lower, and land cover patterns are driven mainly by land use rather than mountain ecosystem processes, reducing the interpretability of the sub-indicator. Furthermore, although human-modified croplands can maintain green cover, they do not provide the same ecological functions, such as supporting native biodiversity, as natural mountain forests or shrublands, which can lead to an underestimation of ecosystem degradation.

The findings of this study provide important information for local, regional, and federal governments to protect degraded soil by integrating soil science into land-use planning in highly sensitive mountain environments that are prone to disturbance, erosion, and land-use change. Authorities can use conservation or ecological restoration zoning to isolate and protect areas with significant carbon sequestration potential (e.g., Histosols) and fertile agricultural soils (e.g., Mollisols and Alfisols) from further degradation due to development. Overall, this approach reflects USDA NRCS soil health principles [72] and supports LDN [54] by helping prevent further soil degradation and by targeting where restoration and more sustainable land-use decisions can be most effective.