gep Journal of Geoscience and Environment Protection 2327-4344 2327-4336 Scientific Research Publishing 10.4236/gep.2026.147018 gep-152784 Article Earth Environmental Sciences Spatiotemporal Dynamics of Forest Cover in the West Coast of Mount Cameroon National Park Taboko Akongenow Lucy 1 Ngomba Longonje Simon 1 Asongwe Godswill Azinwie 1 Bumtu Kamah Pascal 2 Simeon Tchakonte 1 Tataw Guilen-Noel Nghokapin 1 Atianjoh Echuaba Laris 3 Demianus Nkemndem Agendia 3 Mboza Jerry Clinton 3 Department of Environmental Science, Faculty of Science, University of Buea, Buea, Cameroon Deparment of Forestry and Wildlife, Faculty of Agriculture and Veterinary Medicine, University of Buea, Buea, Cameroon Department of Geography, Faculty of Social and Management Science, University of Buea, Buea, Cameroon

The authors declare that they have no competing interests.

 08 07 2026 07 2026 14 07 306 329 29 04 2026 24 07 2026 27 07 2026 © 2026 by the authors and Scientific Research Publishing Inc. 2026 This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ( https://creativecommons.org/licenses/by/4.0/ ). https://doi.org/10.4236/gep.2026.147018

Tropical forest ecosystems play a critical role in regulating global carbon cycles, supporting biodiversity, and sustaining ecosystem services. However, increasing anthropogenic and environmental pressures are driving significant forest cover changes, particularly within protected areas. This study assesses forest cover dynamics and their implications for ecosystem functioning in the West Coast Cluster of Mount Cameroon National Park, Cameroon. A geospatial approach integrating multi-temporal Landsat imagery (1985, 2005, and 2025) and Geographic Information Systems (GIS) was employed to quantify land use/land cover (LULC) changes. Supervised classification using Support Vector Machine (SVM) algorithms enabled the mapping of key land cover classes, while post-classification comparison techniques were used to detect temporal transitions. Results reveal a substantial decline in dense forest cover from 83.22% in 1985 to 53.02% in 2025, representing a loss of approximately 5560 ha. Concurrently, fragmented/secondary forest increased markedly from 1.81% to 29.42%, while bare soils/rocks expanded from 0.81% to 13.73%, indicating intensified degradation. These changes reflect a transition from intact forest systems to a more fragmented and heterogeneous landscape. The observed dynamics are driven by agricultural expansion, logging, settlement encroachment, and natural disturbances such as volcanic activity. The decline in dense forests suggests reduced carbon storage capacity and weakened ecosystem functions, despite partial compensation through secondary forest regrowth. Overall, the findings highlight a progressive loss of ecosystem integrity within the park. Strengthening conservation strategies, enhancing restoration efforts, and integrating forest landscapes into carbon market mechanisms are essential to sustain ecosystem functions and support climate change mitigation.

 Land Use/Land Cover Change Detection GIS Remote Sensing Agricultural Expansion Logging Encroachment Mount Cameroon National Park 1. Introduction

Forests are vital components of the Earth’s biosphere, providing essential ecological functions that sustain global environmental stability ([15]). They regulate the carbon cycle, maintain hydrological processes, preserve biodiversity and support numerous ecosystem services crucial for human well-being. Tropical forests alone harbor over half of the world’s terrestrial biodiversity and act as significant carbon sinks that help mitigate climate change ([33]; [47]). Forest cover dynamics involve not only forest loss but also recovery and the development of secondary forests ([18]). Secondary forests often arise after disturbances or land abandonment, playing a critical role in restoring ecological functions and biodiversity ([48]).

Factors influencing forest cover dynamics include local agricultural expansion, shifting cultivation and fuelwood extraction, which drive forest degradation ([35]). On a broader scale, global market demands for commodities like timber, palm oil and cocoa have accelerated land conversion in many tropical regions ([39]). Forests play a crucial role in carbon storage, biomass productivity, soil formation, water regulation and biodiversity maintenance ([12]). Research indicates that converting natural forests into agricultural landscapes or plantations diminishes aboveground biomass, tree diversity and structural complexity ([19]). Such changes adversely affect ecosystem processes like carbon cycling, soil fertility and hydrological regulation ([62]).

Changes in forest cover have significant implications for ecosystem functioning and global environmental sustainability ([20]). Fragmented forests store lower carbon stocks compared to areas dominated by dense vegetation ([2]). Tropical ecosystems, rich in carbon, account for over 40% of global biodiversity ([13]). In recent decades, forest cover dynamics—spatial and temporal changes in forest extent, structure and composition—have become a major environmental concern ([32]; [53]). Forest ecosystems are closely linked to biodiversity, which drives ecosystem functioning ([51]). Diverse forest communities tend to exhibit greater ecological resilience and productivity, with studies indicating that species diversity significantly contributes to ecosystem stability and functional performance ([57]). Fragmented forests often face edge effects, altered species composition and reduced habitat quality for wildlife ([3]). Across the tropics, deforestation and forest degradation have intensified due to agricultural expansion, logging, infrastructure development and population pressures ([66]; [4]; [55]).

Protected areas are crucial for mitigating forest loss and conserving biodiversity ([72]), yet they face pressures from surrounding communities and illegal resource extraction ([8]). These designated areas protect ecosystems from disturbances like deforestation and land conversion, enabling long-term carbon sequestration ([19]). Beyond species conservation, protected areas sustain essential ecosystem functions, including carbon storage, watershed protection and landscape resilience ([1]; [11]). Landscapes with high biomass density are vital assets in climate mitigation strategies ([9]; [73]).

Using high-resolution satellite observations, global studies have revealed extensive changes in forest landscapes ([45]; [75]). A landmark global analysis based on Landsat imagery estimated that approximately 2.3 million km2 of forest cover was lost worldwide between 2000 and 2012, while only 0.8 million km2 experienced regrowth, highlighting the increasing pressure on global forest ecosystems. Notably, the tropics recorded the most significant increases in forest loss during this period ([40]; [47]; [69]).

Remote sensing technologies have enhanced the monitoring of forest cover changes across various spatial and temporal scales. Satellite imagery from platforms like Landsat, Sentinel and MODIS provides crucial data for mapping land use changes ([40]). Researchers use classification algorithms and image processing techniques to identify forest types, detect deforestation and quantify landscape transformations ([50]). These methods are especially valuable in areas with limited field data. The integration of machine learning and advanced classification techniques allows for more accurate detection of deforestation and land cover changes in tropical forests, supporting environmental monitoring and informing policies aimed at reducing forest loss and promoting sustainable land management ([43]).

In Central Africa, forest cover change is increasingly monitored using geospatial technologies such as remote sensing and Geographic Information Systems (GIS) ([6]; [14]; [21]). These tools allow researchers to analyze multi-temporal satellite imagery to detect patterns of deforestation, forest fragmentation and regeneration over time ([45]). Such approaches have proven essential for understanding long-term landscape transformations and for guiding conservation planning ([30]). Studies suggest that distance-driven urban climate effects associated with nearby urban centers have an effect on protected areas ([36]). Within Africa, the Congo Basin does not just represent the second largest contiguous tropical rainforest system after the Amazon, but it also plays a crucial role in maintaining global carbon balance, regional climate stability and biodiversity conservation ([10]; [61]).

In many rural communities across Cameroon, forest resources remain essential for livelihoods and cultural identity ([31]; [56]). Surveys conducted in forest regions of Cameroon have shown that local populations strongly value provisioning services, particularly food and fuel resources derived from forests ([41]). In Cameroon, protected areas constitute an essential component of national strategies for biodiversity conservation and climate change mitigation ([38]). Despite its ecological importance, Mount Cameroon National Park in general and the west coast cluster has experienced increasing environmental disturbances driven by both climatic and human-induced factors. Climate change is influencing vegetation dynamics through alterations in rainfall patterns, increasing temperatures and the occurrence of extreme weather events ([49]).

This research aims to assess forest cover dynamics in the West Coast Cluster of Mount Cameroon National Park and their impact on key ecosystem functions, particularly carbon stock dynamics and ecosystem integrity, by integrating remote sensing analysis with ecological interpretation to inform conservation strategies and climate change mitigation efforts.

 2. Materials and Methods

This study adopts a geospatial approach that integrates remote sensing and Geographic Information Systems (GIS) to assess forest-cover dynamics and their implications for ecosystem functioning within the West Coast Cluster of Mount Cameroon National Park. The methodology focuses on detecting land-cover changes over time and linking these changes to ecosystem functions, particularly forest carbon stock.

Study Area

The West Coast Cluster of Mount Cameroon National Park (MCNP) is one of the compartments of Mount Cameroon national park, located in the South-West Region of Cameroon along the Gulf of Guinea in Central Africa. Mount Cameroon National Park constitutes one of the most ecologically significant protected landscapes within the Cameroon Volcanic Line, encompassing diverse altitudinal gradients, complex volcanic geomorphology, and highly productive tropical forest ecosystems. This study focuses specifically on the western coastal section of the park, an ecologically dynamic interface between montane forest systems and lowland coastal environments.

Geographically, the west Coast Cluster of Mount Cameroon National Park lies approximately between 4˚00'N and 4˚20'N latitude and 8˚55'E and 9˚20'E longitude, locating it on the southwestern flanks of Mount Cameroon, the highest peak in West and Central Africa (4095 m above sea level). The location map Figure 1 illustrates the spatial extent of the West Coast Cluster extending from near-coastal settlements toward inland forested slopes, forming an important ecological corridor connecting marine-influenced ecosystems with submontane and montane forest zones. This geographic positioning places the study area within a region strongly influenced by both Atlantic maritime climatic systems and orographic processes generated by Mount Cameroon’s massive volcanic structure ([28]).

Figure1. Location map of west coast cluster of Mount Cameroon national park (MCNP).

Topographically, the landscape is characterized by steep volcanic slopes, dissected valleys, and lava-derived landforms resulting from historical eruptions of Mount Cameroon. Elevation increases rapidly from near sea level along the coast to high-altitude forest zones inland, creating pronounced ecological gradients over relatively short horizontal distances. These gradients strongly influence vegetation distribution, soil formation, hydrological processes, and ecosystem productivity ([24]). The underlying geology is dominated by basaltic lava flows and pyroclastic deposits, producing fertile but erosion-prone Andosols that support dense forest growth while remaining sensitive to disturbance ([16]).

Climatically, the West Coast Cluster experiences a humid tropical climate controlled by moist southwesterly winds from the Atlantic Ocean. Orographic uplift along the mountain slopes generates extremely high rainfall, often exceeding 3000 - 5000 mm annually, making the Mount Cameroon region one of the wettest environments in Africa ([37]). Persistent cloud cover, high atmospheric humidity, and relatively stable temperatures create favourable conditions for evergreen tropical forests and high biomass accumulation, which are critical for carbon storage and ecosystem functioning.

Vegetation within the West Coast Cluster of Mount Cameroon National Park comprises a mosaic of lowland rainforest, submontane forest, secondary forest regrowth, and localised disturbed patches, reflecting both natural ecological processes and anthropogenic pressures, including smallholder agriculture, fuelwood collection, and settlement expansion along park boundaries ([42]). These forests provide essential ecosystem services, including biodiversity conservation, watershed protection, climate regulation, and soil stabilization. The West Coast Cluster is particularly important as a biodiversity refuge supporting endemic plant species, primates, birds, and other wildlife associated with the Mount Cameroon ecosystem, such as the African Forest Elephants.

Multi-temporal satellite imagery was used to evaluate land-cover dynamics within the study area (West Coast Cluster of Mount Cameroon National Park). Satellite image data from the Landsat Program were obtained for selected years representing long-term land-cover changes. Specifically, Landsat 5 TM, Landsat 7 ETM+, and Landsat 9 OLI images with a spatial resolution of 30 meters were acquired from the United States Geological Survey Earth Explorer platform. These datasets provide consistent historical coverage suitable for detecting forest cover change over several decades. To improve classification accuracy and reduce atmospheric disturbances, images with minimal cloud cover during the dry and transitional seasons were selected, as seen in Table 1.

Table 1. Image classification characteristics from Landsat TM 1985 (Landsat thematic mapper), enhanced thematic mapper plus 2005 (ETM+) and OLI 2 operational land imager 2, 2025.

Years Satellite Sensor Satellite Mission Date Acquired Path/Row Cloud Cover (%) Spatial Resolution Spectral Bands Used for Analysis
1985 TM Landsat Program (Landsat 5) 26-Oct-85 187/057 >30 30 m Bands 3, 2, 1 (R,G,B) - True Color Composite
2005 ETM+ Landsat Program (Landsat 7) 9-Nov-05 187/057 >50 30 m Bands 3, 2, 1 (R,G,B) - True Color Composite; Bands 4,3,2
2025 OLI-2 Landsat Program (Landsat 9) 15-Nov-25 187/057 >40 30 m Bands 4,3,2 (R,G,B) - True Color Composite; Bands 5,4,3

Prior to analysis, satellite images were subjected to standard pre-processing procedures to ensure data consistency and accuracy ([68]). These steps included radiometric correction, atmospheric correction, and geometric correction to align all images within the same coordinate reference system. Cloud masking and image clipping were performed to extract the spatial extent corresponding to the West Coast Cluster of Mount Cameroon National Park. Pre-processing procedures were carried out using ENVI software 4.5 ([34]; [46]). The Panchromatic band (Band 8) was utilised to improve the image resolution prior to classification, particularly for 2005 and 2025, which was available given the Landsat Mission.

The Land use and land cover (LULC) classification was conducted using a supervised classification approach ([64]). Training samples representing major land-cover classes such as dense forest, Fragmented Forest/secondary forest, Grassland, Baresoils/Bare Rocks were selected based on spectral signatures derived from the satellite imagery. This landuse and Landcover classes as were classified based on specific descriptions that align with their state (Table 2). A supervised classification approach was adopted to categorize the satellite imagery into distinct land use and land cover classes, such as Dense Forest, Secondary Forest/Agricultural Land and Settlements/Bare soils. The ENVI 4.5 software ([22]) was used to perform the image classification.

Table 2. Landuse/landcover classification.

Land Use Type Description
Bare Soils/Rocks Areas with little or no vegetation cover, exposing soil surfaces or rock outcrops. These zones may result from natural processes (erosion, landslides) or human activities such as deforestation and Urbanization.
Dense Forest Regions dominated by continuous, closed-canopy vegetation with high tree density and minimal disturbance.
Fragmented Forest/Secondary Forest Forest areas that have been disturbed, broken into patches, or are in stages of regrowth following human or natural disturbances (e.g., logging, agriculture, fire).
Grassland These landscapes may be natural (savannas) or anthropogenic (resulting from deforestation or grazing).

The classification process applied machine-learning algorithms, particularly the Support Vector Machine (SVM)classifier, due to its robustness and high accuracy in forest landscape mapping ([26]; [74]). Using the Envi software, the training samples were selected using the pixel identification reflection in relation to the Ground Control Points that were imported into the Envi Software. Image classification was performed in ENVI using the Radial Basis Function (RBF) kernel, which is widely recognized for its ability to model complex non-linear relationships in multispectral remote sensing data. The classifier parameters (penalty parameter C and kernel parameter γ) were optimised iteratively through repeated testing to maximise classification accuracy while minimising misclassification among spectrally similar land-cover classes. Training samples representing all land-use/land-cover (LULC) classes were used to train the classifier, and classification performance was evaluated using independent validation samples.

SVM was selected because it performs exceptionally well with high-dimensional remote sensing datasets, effectively handles limited training samples, and minimises overfitting through structural risk minimisation. The RBF kernel was particularly suitable for this heterogeneous mountainous landscape because it can capture complex spectral separability among vegetation.

Cloud-contaminated areas were corrected through image substitution using temporally comparable Landsat scenes with minimal cloud cover, ensuring complete spatial coverage of the study area. For Landsat 7 ETM+ imagery affected by the Scan Line Corrector (SLC)-off failure, data gaps were corrected using the GapFill algorithm implemented in QGIS prior to image import and classification in ENVI. All imagery was subsequently processed using a common study-area mask and standardized to the same spatial resolution to ensure temporal consistency and comparability across all assessment years. These procedures as shown in Figure 2 minimized potential biases associated with sensor-specific artifacts and missing data.

Figure 2. Image classification methodology scheme.

The classification procedure generated thematic land-cover maps for each selected year, allowing for the detection of spatial changes in forest distribution across the study area. Below demonstrates the Geospatial image classification work frame used for the analysis.

To evaluate the reliability of the classification results, an accuracy assessment was performed using ground reference data collected in the field using a Hand Garmin 65 GPS (Global Positioning system) and high-resolution imagery from Google Earth ([23]; [54]). Only classification outputs achieving acceptable accuracy thresholds (generally above 85%) were considered suitable for further analysis. Our Kappa accuracy was as follow 94.57% (0.9), 96.31% (0.94) and 94.57% (0.92) for 1985, 2005 and 2025 respectively. Post-classification comparison techniques were applied to detect changes in forest cover between different time periods ([70]). This approach involved comparing classified land-cover maps to quantify transitions between forest and non-forest classes ([71]).

To link land-cover changes with ecosystem functioning, forest cover classes were interpreted as indicators of ecosystem services, particularly carbon storage potential integrity ([7]). Areas classified as dense forest were considered to represent high carbon storage capacity, whereas converted landscapes such as Fragmented areas and Bare soils/rock land were associated with reduced ecosystem functionality. By integrating spatial change detection results with ecological interpretation, the study provides insights into how forest cover dynamics influence ecosystem functions and conservation outcomes in Mount Cameroon National Park.

 3. Results 3.1. Land Use Landcover

Table 3. LULC statistics of Landsat TM 1985 (Landsat thematic mapper), enhanced thematic mapper plus 2005 (ETM+), and OLI 2 operational land imager 2, 2025.

LULC Area 1985 % Cover 1985 Area 2005 % Cover 2005 Area 2025 % Cover 2025
Baresoils/Rocks 149.00 0.81 634.55 3.45 2526.32 13.73
Dense Forest 15320.66 83.22 12369.80 67.20 9759.74 53.02
Fragmented Forest/Secondary Forest 333.20 1.81 2612.01 14.19 5414.88 29.42
Grassland 2607.58 14.16 2792.14 15.17 705.23 3.83
TOTAL 18410.43 100 18408.49 100 18406.17 100

Understanding long-term land-use and land-cover (LULC) dynamics is essential for assessing ecosystem structure, function and resilience, particularly in protected tropical forest areas like Mount Cameroon National Park. The analysis of LULC from 1985 to 2025 reveals significant changes impacting biodiversity preservation, forest health, water regulation and overall ecosystem stability. The findings quantify spatial and temporal changes in key LULC classes with dense forest decreasing from 15320.66 ha (83.22%) in 1985 to 9759.74 ha (53.02%) in 2025, highlighting an alarming loss of 5560.91 ha, or 30.19%, over the 40-year period (Table 3).

 3.2. Land Use Landcover 1985

In 1985, the land use and land cover structure of the West Coast Cluster was predominantly characterized by dense forest, which occupied 15320.66 ha, representing 83.22% of the total area. This significant coverage illustrates the critical role of intact forests in maintaining biodiversity and carbon storage. Grassland constituted the second-largest class at 2607.58 ha (14.16%), while fragmented/secondary forest was minimal at just 333.20 ha (1.81%). The dominance of dense forest in this period indicates a landscape primarily reliant on continuous, closed-canopy forest, essential for conservation efforts aimed at preserving ecological integrity.

Landsat image classified revealed that in 1985, the land use and land cover structure of the West Coast Cluster of Mount Cameroon National Park was strongly dominated by dense forest, with relatively minor representation from other classes of the LULC. The statistical distribution seen in Table 2 reflects a highly forested landscape with limited exposure of non-forest surfaces. Dense forest constituted the overwhelmingly dominant land cover class in 1985, occupying 15320.66 ha, which represented 83.22% of the total study area. This proportion indicates that more than 4/5 of the landscape was under continuous, closed-canopy forest. The magnitude of this coverage far exceeded all other land cover categories, with a spatial extent nearly six times greater than the next largest class (grassland) (Figure 3).

Moreso, results revealed that grassland was the second most prominent land cover class, covering 2607.58 ha and accounting for 14.16% of the total area. Although significantly smaller than dense forest, grassland still represented a notable portion of the landscape, constituting over one-seventh of the total coverage. The difference between grassland and dense forest coverage was substantial, with grassland occupying approximately 12,700 ha less than dense forest. Satellite image results show that fragmented/secondary forest occupied 333.20 ha, corresponding to 1.81% of the total area in 1985. This class represented a very small fraction of the landscape compared to both dense forest and grassland. In relative terms, its coverage was less than 2%, indicating that disturbed forested areas were minimal at this time. The spatial extent of this class was nearly eight times smaller than that of grassland and more than 45 times smaller than dense forest.

Bare soils and rock surfaces formed the least represented land cover class, covering 149.00 ha, which accounted for only 0.81% of the total study area. This indicates that exposed, non-vegetated surfaces were extremely limited in spatial extent. Statistically, this class occupied less than 1% of the landscape, making it the smallest LULC category in 1985, with an area more than two times smaller than fragmented forest and over 100 times smaller than dense forest (Figure 3).

 3.3. Land Use Landcover 2005

By 2005, the land use and land cover composition reflected a shift towards greater diversification, with dense forest still dominant at 12369.80 ha (67.20%), although this marked a decline of 2950.86 ha (16.02%) since 1985. The increase in fragmented/secondary forest to 2612.01 ha (14.19%) represents a notable rise of 2278.81 ha (12.38%), suggesting a transition driven by anthropogenic pressures such as agricultural expansion and illegal logging. This change underscores the urgent need for conservation strategies that address fragmentation and enhance the recovery of disturbed areas to maintain ecosystem functions (Figure 3).

Classified satellite images of Landsat 7 ETM+ revealed that in 2005, the land use and land cover composition of the West Coast Cluster of Mount Cameroon National Park reflects a more diversified distribution compared to earlier observations, with noticeable shifts in the relative proportions of the different land cover classes. The statistical structure shows a reduction in dominance by a single class and a more pronounced presence of transitional land cover types.

Based on the results obtained from the satellite image classification, it revealed that dense Forest remained the dominant land cover class in 2005, covering 12369.80 ha and accounting for 67.20% of the total study area. Although it still occupied more than two-thirds of the landscape, its proportional dominance was notably lower compared to earlier conditions in 1985. In terms of spatial extent, dense forest exceeded all other classes individually, maintaining an area more than four times larger than the second-largest class, fragmented forest (Figure 3).

Also, Fragmented/secondary forest represented a significant portion of the landscape in 2005, covering 2612.01 ha, which corresponded to 14.19% of the total area. This positioned it as the second most extensive land cover class during this period. Statistically, its coverage was nearly equivalent to that of grassland, with only a marginal difference of about 0.98%. In absolute terms, the fragmented forest occupied slightly less area than the grassland but remained within a comparable range. Grassland covered 2792.14 ha in 2005, accounting for 15.17% of the study area. This made it the second-largest land cover class, slightly surpassing fragmented forest in both area and percentage share. The difference between grassland and fragmented forest was relatively small (approximately 180 ha), indicating a close distribution between these two classes. Despite this, grassland remained substantially smaller than dense forest, occupying less than one-quarter of its area. Bare soils and rock surfaces occupied 634.55 ha, representing 3.45% of the total study area in 2005. While still the least extensive class, its spatial coverage was notably larger than in other categories, such as fragmented forest in earlier periods. Statistically, this class remained below 5% of the total area, but its extent was more than double that of the fragmented forest in 1985 and approximately one-quarter of the grassland coverage in 2005 (Figure 3).

 3.4. Land Use Landcover 2025

In 2025, the LULC composition showed a further decline in dense forest to 9759.74 ha (53.02%), indicating a continued loss of forest cover with a decrease of 2610.06 ha (14.17%) from 2005. Fragmented/secondary forest expanded significantly to 5414.88 ha (29.42%), reflecting a gain of 2802.87 ha (15.23%). This increase emphasizes the importance of managing fragmentation to sustain ecosystem health and resilience. The notable rise in bare soils and rocks to 2526.32 ha (13.73%) also indicates a shift towards more degraded land cover, highlighting the need for targeted conservation actions to restore ecological integrity (Figure 3).

Figure 3. Land use/land cover map of west coast cluster of Mount Cameroon national park.

In 2025, the land use and land cover composition of the West Coast Cluster of Mount Cameroon National Park shows a markedly redistributed pattern among the major classes, with no single class exceeding three-quarters of the total area. The statistical structure reflects a more even spread across multiple categories, although clear differences in magnitude remain. Dense forest continued to be the largest land cover class in 2025, occupying 9759.74 hectares and accounting for 53.02% of the total study area. This indicates that just over half of the landscape was covered by dense forest. In relative terms, its share was only marginally greater than the combined proportion of all other land cover classes. Despite maintaining the highest individual coverage, its extent was less than twice that of fragmented forest, showing a reduced dominance compared to earlier distributions.

Fragmented or secondary forest covered 5414.88 hectares, representing 29.42% of the total area in 2025. This made it the second most dominant land cover class, occupying nearly one-third of the study area. Statistically, its extent was more than half that of dense forest, with a difference of approximately 4300 hectares. The proportion of this class was substantially higher than that of grassland and bare surfaces, positioning it clearly as a major component of the landscape.

Bare soils and rock surfaces accounted for 2526.32 hectares, which corresponds to 13.73% of the total area. This placed it as the third largest land cover class in 2025. In comparative terms, its coverage was roughly half that of fragmented forest and about one-quarter of dense forest. The proportion exceeded that of grassland by a considerable margin, with a difference of nearly 10 percentage points, indicating a strong representation within the overall distribution. Grassland occupied 705.23 hectares in 2025, accounting for 3.83% of the total study area. This made it the least represented land cover class during this period. Statistically, its extent was significantly lower than all other categories, being more than three times smaller than bare surfaces and nearly eight times smaller than fragmented forest. Its proportional share remained below 5%, indicating a minimal contribution to the overall land cover distribution in the year 2025 (Figure 3).

1985-2005: Comparative LULC Changes

Between 1985 and 2005, in the West Coast Cluster of Mount Cameroon National Park, clear shifts occurred across all land cover classes. Dense forest decreased from 15320.66 ha (83.22%) to 12369.80 ha (67.20%), reflecting a net loss of 2950.86 ha and a decline of 16.02%. Despite remaining the dominant class, its proportional share reduced substantially over these 20 years. Fragmented/secondary forest exhibited a pronounced increase, rising from 333.20 ha (1.81%) to 2612.01 ha (14.19%). This corresponds to a gain of 2278.81 ha and an increase of 12.38%, making it one of the most rapidly expanding classes during this interval. The magnitude of this increase brought it to near parity with grassland in percentage terms. Grassland showed a modest increase from 2607.58 ha (14.16%) to 2792.14 ha (15.17%), representing a gain of 184.56 ha and a slight rise of 1.01%. Its proportional share remained relatively stable compared to other classes, with only minimal variation over the period. Bare soils and rocks increased from 149.00 ha (0.81%) to 634.55 ha (3.45%), reflecting a gain of 485.55 ha and an increase of 2.64%. Although still the smallest class in 2005, it recorded a noticeable relative expansion compared to its 1985 extent (Table 4).

Table 4. LULC statistics of Landsat TM 1985 (Landsat thematic mapper), enhanced thematic mapper plus 2005.

LULC Area Δ 1985-2005 % Δ 1985-2005
Baresoils/Rocks 485.55 2.64
Dense Forest −2950.86 −16.02
Fragmented Forest/Secondary Forest 2278.81 12.38
Grassland 184.56 1.00
TOTAL − 1.94 0.00

2005-2025: Comparative LULC Changes

From 2005 to 2025, the LULC distribution continued to shift significantly. Dense forest declined further from 12369.80 ha (67.20%) to 9759.74 ha (53.02%), representing a loss of 2610.06 ha and a decrease of 14.18 percentage points. This indicates a continued reduction in its overall share, although the rate of percentage decline was slightly lower than in the previous period. Fragmented or secondary forest increased markedly from 2612.01 ha (14.19%) to 5414.88 ha (29.42%). This corresponds to a gain of 2802.87 ha and an increase of 15.23 percentage points, making it the class with the largest absolute and proportional growth during this period. Its share more than doubled, reinforcing its position as a major land cover category. Grassland experienced a sharp decline, decreasing from 2792.14 ha (15.17%) to 705.23 ha (3.83%). This represents a loss of 2086.91 ha and a reduction of 11.34 percentage points. The magnitude of this decrease contrasts strongly with its relative stability in the earlier period. Bare soils and rocks showed a substantial increase from 634.55 ha (3.45%) to 2526.32 ha (13.73%), reflecting a gain of 1891.77 ha and an increase of 10.28 percentage points. This significant rise elevated it from a minor to a more prominent land cover class by 2025 (Table 5).

Table 5. LULC statistics of enhanced thematic mapper plus 2005 (ETM+), and OLI 2 operational land imager 2, 2025.

LULC Area Cover Δ 2005-2025 % Cover Δ 2005-2025
Baresoils/Rocks 1891.77 10.28
Dense Forest −2610.05 −14.17
Fragmented Forest/Secondary Forest 2802.87 15.23
Grassland −2086.91 −11.34
TOTAL −2.33 0.00

1985-2025: Long-Term Comparative LULC Changes

Over the entire 40-year period from 1985 to 2025, the cumulative changes in LULC are pronounced. Dense forest declined from 15320.66 ha (83.22%) to 9759.74 ha (53.02%), resulting in a total loss of 5560.92 ha and a decrease of 30.20%. This represents the largest absolute reduction among all classes. Fragmented/secondary forest increased from 333.20 ha (1.81%) to 5414.88 ha (29.42%), corresponding to a gain of 5081.68 ha and an increase of 27.6%. This is the most substantial overall increase, both in area and percentage, across the study period. Grassland decreased from 2607.58 ha (14.16%) to 705.23 ha (3.83%), reflecting a loss of 1902.35 ha and a decline of 10.33%. This indicates a notable contraction in its spatial extent over the long term. Bare soils and rocks increased from 149.00 ha (0.81%) to 2526.32 ha (13.73%), resulting in a gain of 2377.32 ha and an increase of 12.92%. This long-term growth represents a substantial relative expansion, moving from a minimal to a more significant share of the total landscape (Table6).

The land use and land cover (LULC) dynamics in the West Coast Cluster of Mount Cameroon National Park are shaped by a complex interplay of anthropogenic pressures and natural environmental factors. Notably, agricultural expansion, particularly by smallholder farmers, has led to the clearing of dense forest areas, resulting in a loss of 2950.86 ha (16.02%) of dense forest between 1985 and 2005. Additionally, illegal logging has contributed to an increase in fragmented forest by 2278.81 ha (12.38%) during the same period, underscoring the need for enhanced conservation efforts. Natural processes, including volcanic activity, have also influenced LULC, with bare soils and rocks rising from 149.00 ha (0.81%) in 1985 to 2526.32 ha (13.73%) in 2025, highlighting the importance of integrating these factors into conservation strategies. Furthermore, limitations in enforcement and community engagement in conservation management have hindered effective restoration efforts, as evidenced by the total loss of 5560.91 ha of dense forest over the 40-year period, emphasizing the urgent need for comprehensive actions to restore ecological integrity and ensure long-term sustainability in the region.

Table 6. LULC statistics of Landsat TM 1985 (Landsat thematic mapper), enhanced thematic mapper plus 2005 (ETM+), and OLI 2 operational land imager, 2025.

LULC Area Cover Δ 1985-2025 % Cover Δ 1985-2025
Baresoils/Rocks 2377.32 12.92
Dense Forest −5560.91 −30.19
Fragmented Forest/Secondary Forest 5081.68 27.61
Grassland −1902.35 −10.33
TOTAL − 4.27 0.00

LULC Transition Matrix

Below is a table showing the transition matrices for the time periods 1985-2005, 2005-2025, and 1985-2025.

1985-2005 (%)
Dense Forest Fragmented Forest Grassland Baresoils Row Total Class Total
Dense Forest 76.568 41.931 2.671 10.85 100 100
Fragmented Forest 13.656 43.719 2.624 2.833 100 100
Grassland 8.025 12.852 10.59 14.105 100 100
Baresoils 1.752 1.498 84.115 72.212 100 100
Class Total 100 100 100 100 0 0
Class Changes 23.432 56.281 89.41 27.788 0 0
Image Difference −19.63 559.611 −80.648 8508.077 0 0
2005-2025 (%)
Dense Forest Fragmented Forest Grassland Baresoils Row Total Class Total
Unclassified 0 0 0 0 0 0
Dense Forest 56.435 30.548 20.232 1.449 100 100
Fragmented Forest 41.454 55.603 42.114 1.278 100 100
Grassland 2.004 13.744 37.547 95.57 100 100
Baresoils 0.107 0.105 0.107 1.704 100 100
Class Total 100 100 100 100 0 0
Class Changes 43.565 44.397 62.453 98.296 0 0
Image Difference −30.716 198.963 386.957 −98.149 0 0
1985-2025 (%)
Dense Forest Fragmented Forest Grassland Baresoils Row Total Class Total
Dense Forest 51.129 36.544 3.26 43.882 100 100
Fragmented Forest 45.253 57.845 5.024 31.826 100 100
Grassland 2.356 4.984 91.653 1.929 100 100
Baresoils 1.261 0.626 0.063 22.363 100 100
Class Total 100 100 100 100 0 0
Class Changes 48.871 42.155 8.347 77.637 0 0
Image Difference −44.316 1871.994 −5.764 59.373 0 0
4. Discussion

The land use and land cover (LULC) dynamics observed in the West Coast Cluster of Mount Cameroon National Park exhibit notable parallels with the findings of [65] in the Ebo Forest Reserve, particularly regarding significant transformations driven by anthropogenic pressures such as agricultural expansion and logging. Both studies document a marked decline in dense forest cover, with the West Coast Cluster experiencing a reduction from 83.22% to 53.02%, similar to the fragmentation and degradation reported by [65]. Similar to [65], the increase in bare soils and rock surfaces, rising from 0.81% to 13.73%, underscores intensified disturbances common to both regions. These multifaceted drivers of LULC changes in both studies emphasize the complex interplay between human activities and natural factors, necessitating comprehensive management approaches to mitigate degradation ([65]).

The land use and land cover (LULC) dynamics observed in the West Coast Cluster of Mount Cameroon National Park over the 40-year period from 1985 to 2025 reveal a pronounced transformation of the landscape, consistent with findings from other studies. The substantial decline in dense forest cover, which decreased from 83.22% to 53.02%, aligns with global trends reported by [27] and [69], indicating that agricultural expansion remains a primary driver of deforestation. This shift from a predominantly intact forest system to a more heterogeneous landscape reflects cumulative disturbances over time, corroborating the observations of [67] regarding the emergence of mosaic landscapes in response to both anthropogenic and natural pressures.

The notable rise in bare soils and rock surfaces, increasing from 0.81% to 13.73%, underscores intensifying disturbance processes within the study area. This trend aligns with findings from [52] and [60], who document similar patterns of forest-to-savanna transitions linked to land degradation and climate variability across sub-Saharan Africa. The simultaneous decline in grassland cover further emphasizes the extent of landscape transformation, suggesting that even transitional vegetation types are being altered or replaced over time, corroborating the broader ecological implications identified in other studies.

These LULC changes have significant repercussions for carbon stock distribution and ecosystem functioning. The reduction in dense forests, which serve as primary carbon reservoirs, leads to a redistribution of carbon from stable, high-density pools to more dynamic and lower-density systems, as noted by [25]. While the growth in fragmented forests suggests ongoing regeneration, the overall carbon storage capacity of the landscape is likely diminished due to reduced biomass. This observation is supported by [44], who link deforestation to increased atmospheric carbon concentrations, emphasizing the urgency of implementing conservation measures to mitigate these impacts.

The implications for ecosystem function are profound, as the reduction in dense forest cover disrupts key ecological processes such as nutrient cycling and hydrological regulation. This aligns with findings from [17], which highlight how fragmentation modifies environmental conditions and influences species composition. The increase in degraded surfaces disrupts soil processes and water dynamics, limiting the ecosystem’s capacity to sustain productivity and resilience, a concern echoed in studies conducted in Ethiopia ([59]).

The drivers of these LULC transitions are multifaceted, involving both anthropogenic and natural factors. Agricultural expansion, logging, and settlement growth have played dominant roles in the conversion and degradation of forested areas, confirming results from [58]. These pressures are often concentrated in accessible regions, leading to spatially uneven patterns of change. Meanwhile, natural processes such as volcanic activity associated with Mount Cameroon contribute to localized disturbances, echoing findings from [29] regarding the interaction between human activities and natural vulnerabilities that exacerbate landscape transformation.

An important aspect emerging from these findings is the apparent inadequacy of restoration efforts within the park, as highlighted by [5], who discuss successful restoration projects in Mozambique. Despite the observed degradation, limited evidence of effective restoration interventions suggests that current management strategies may not adequately address the scale or intensity of landscape change. This gap not only prolongs the persistence of degraded land cover but also hinders the recovery of ecosystem functions and carbon storage capacity, a concern reiterated by [63].

 5. Conclusion

The analysis of land use and land cover (LULC) dynamics in the West Coast Cluster of Mount Cameroon National Park over the period 1985-2025 reveals a clear and sustained transformation of the landscape, characterized by a substantial decline in dense forest cover and a concurrent expansion of fragmented forests and degraded land surfaces. The progressive reduction of dense forest from a dominant 83.22% to 53.02%, alongside the marked increase in secondary vegetation and bare surfaces, underscores a transition from a largely intact forest ecosystem to a more heterogeneous and structurally altered landscape. Dense forest cover is considered representative of relatively high ecosystem integrity because it maintains greater structural complexity, biodiversity support, habitat connectivity, and ecological functionality. Similarly, dense forests are associated with the highest aboveground biomass and carbon storage potential, whereas degraded vegetation, agricultural land, bare land, and built-up areas generally exhibit progressively lower carbon stocks. Consequently, multiple related studies corroborate key findings indicating that interpretations of ecosystem integrity and carbon dynamics are based on established ecological relationships among land-cover type, vegetation structure, and biomass accumulation rather than on direct field-based carbon measurements. Drivers of these transformations are multifaceted, with local anthropogenic pressures such as agricultural expansion, logging and settlement encroachment playing a central role, compounded by natural factors, including volcanic activity and geomorphological processes associated with the Mount Cameroon landscape, as confirmed by other literature from different parts of the world. The limited evidence of successful restoration efforts emphasizes a critical gap in current conservation strategies, especially in areas requiring active ecological recovery.

Despite the application of advanced geospatial techniques and rigorous analytical procedures, this study was subject to several limitations. Persistent cloud cover, inherent differences among satellite sensors, and the complex topography of the mountainous landscape introduced uncertainties into the detection and interpretation of forest cover dynamics. Furthermore, the absence of direct field-based carbon measurements constrained the validation of remotely sensed estimates and limited the assessment of carbon stock changes associated with observed land-cover transitions. Nevertheless, integrating multiple geospatial datasets and methodological approaches provides a robust basis for understanding the spatio-temporal dynamics of forest cover within the West Coast Cluster of Mount Cameroon.

 Authors Contributions

Taboko Akongenow Lucy, Longonje Simon Ngomba, Asongwe Godwill, Athanasius Fuashi Nkwatoh and Mokake Seraphine were responsible for conceptual contributions and research design. Nkemdem Agendia D., Taboko Akongenow Lucy and Guilen-Noel N. Tataw contributed to the GIS and Remote sensing and Data analyses for this study. Levenje Kinge Romoe, Echuaba Laris, Mboza Jerry Clinton, Guilen-Noel N. Tataw and Kamah Pascal Bumtu were responsible for experimental work. Taboko Akongenow Lucy is the principal researcher and wrote the manuscript. All authors read and approved the final manuscript.

 Acknowledgements

The authors would like to express their sincere gratitude to several individuals and organizations who have contributed to the successful completion of this study. First and foremost, we would like to thank the University of Buea and Mount Cameroon National Park service for support that made this research possible.

 References Adi, B. T. S., Putra, S., Sari, R. K., Dewantara, B., & Suryaningtyas, P. (2025). Managing High Conservation Value and High Carbon Stock Areas in Coffee Landscapes of Mandailing Natal. Jurnal Locus Penelitian dan Pengabdian, 4, 9308-9316. https://doi.org/10.58344/locus.v4i10.4848 10.58344/locus.v4i10.4848https://doi.org/10.58344/locus.v4i10.4848 Adi, B. Putra, S. Sari, R. Dewantara, B. Suryaningtyas, P. 2025 10.58344/locus.v4i10.4848 Ahmad, M. J., Ismail, R., & Ghani, F. A. (2023). Review on Socioeconomic and Sustainability of Oil Palm Plantations among Rural Communities in Malaysia. IOP Conference Series: Earth and Environmental Science, 1208, Article ID: 012054. https://doi.org/10.1088/1755-1315/1208/1/012054 10.1088/1755-1315/1208/1/012054https://doi.org/10.1088/1755-1315/1208/1/012054 Ahmad, M. Ismail, R. Ghani, F. 2023 012054 ID 10.1088/1755-1315/1208/1/012054 Ajonina, S. A., Gerhard, W., Nkwatoh, A. F., & Hofer, H. (2014). Endangering the Endangered: Are Protected Areas Save Havens for Threatened Species in Cameroon? Case of Banyang-Mbo Wildlife Sanctuary, South Western Cameroon. Journal of Ecology and The Natural Environment, 6, 42-55. https://doi.org/10.5897/jene2013.0430 10.5897/jene2013.0430https://doi.org/10.5897/jene2013.0430 Ajonina, S. Gerhard, W. Nkwatoh, A. Hofer, H. Sanctuary, S 2014 10.5897/jene2013.0430 Akoh, N. R., Kelvin, N., & Tiafack, O. (2021). The Environmental and Socio Economic Signatures of Landcover and Landuse Dynamics in the Ndu Tea Estate, North West Cameroon. Journal of Geography, Environment and Earth Science International,25, 27-43. https://doi.org/10.9734/jgeesi/2021/v25i1030319 10.9734/jgeesi/2021/v25i1030319https://doi.org/10.9734/jgeesi/2021/v25i1030319 Akoh, N. Kelvin, N. Tiafack, O. Estate, N Geography, E 2021 10.9734/jgeesi/2021/v25i1030319 Ameja, L. G., Ribeiro, N., Sitoe, A. A., & Guillot, B. (2022). Regeneration and Restoration Status of Miombo Woodland Following Land Use Land Cover Changes at the Buffer Zone of Gile National Park, Central Mozambique. Trees, Forests and People, 9, Article ID: 100290. https://doi.org/10.1016/j.tfp.2022.100290 10.1016/j.tfp.2022.100290https://doi.org/10.1016/j.tfp.2022.100290 Ameja, L. Ribeiro, N. Sitoe, A. Guillot, B. Park, C Trees, F 2022 100290 ID 10.1016/j.tfp.2022.100290 Anh, D. T. L., Hung, D. V., & Hung, N. T. (2021). Monitoring of Forest Land Cover Change in Nui Ba Ra Protected Landscape in Vietnam Using Remote Sensing Methods and GIS Techniques. IOP Conference Series: Earth and Environmental Science, 666, Article ID: 052009. https://doi.org/10.1088/1755-1315/666/5/052009 10.1088/1755-1315/666/5/052009https://doi.org/10.1088/1755-1315/666/5/052009 Anh, D. Hung, D. Hung, N. 2021 052009 ID 10.1088/1755-1315/666/5/052009 Bajracharya, R. M., Sitaula, B., Lal, S. H., Bajracharya, R. M., & Sitaula, B. K. (2012). Forest and Soil Carbon Stocks, Pools and Dynamics and Potential Climate Change Mitigation in Nepal. Formerly Part of Journal of Environmental Science and Engineering, 1,800-811. https://www.researchgate.net/publication/269976435 Bajracharya, R. Sitaula, B. Lal, S. Bajracharya, R. Sitaula, B. Stocks, P 2012 Belote, R. T., & Wilson, M. B. (2020). Delineating Greater Ecosystems around Protected Areas to Guide Conservation. Conservation Science and Practice, 2, e196. https://doi.org/10.1111/csp2.196 10.1111/csp2.196https://doi.org/10.1111/csp2.196 Belote, R. Wilson, M. 2020 10.1111/csp2.196 Berisha, N., Ćušterevska, R., Lluga-Rizani, K., Millaku, F., & Matevski, V. (2020). Relation between Boundaries of Protected Areas and the Distribution of Vulnerable Natural Habitats—A Case Study from Sharri National Park, SE Europe. Ecological Questions, 32, 1-18. https://doi.org/10.12775/eq.2021.008 10.12775/eq.2021.008https://doi.org/10.12775/eq.2021.008 Berisha, N. Lluga-Rizani, K. Millaku, F. Matevski, V. Park, S 2020 10.12775/eq.2021.008 Blake, S., Dewachter, P., Hart, J., Hugel, B., Huijbregts, B., Julien, P., Kaboza, Y., Knight, K., & Lefeuvre, J. (2010). World Heritage in the Congo Basin. https://whc.unesco.org/uploads/activities/documents/activity-628-1.pdf Blake, S. Dewachter, P. Hart, J. Hugel, B. Huijbregts, B. Julien, P. Kaboza, Y. Knight, K. Lefeuvre, J. 2010 Condro, A. A., Prasetyo, L. B., Rushayati, S. B., Santikayasa, I. P., & Iskandar, E. (2022). Protected Areas Slow down Tropical Rainforest Disturbance in the Leuser Ecosystem, Indonesia. Journal of Land Use Science, 17, 454-470. https://doi.org/10.1080/1747423x.2022.2115571 10.1080/1747423x.2022.2115571https://doi.org/10.1080/1747423x.2022.2115571 Condro, A. Prasetyo, L. Rushayati, S. Santikayasa, I. Iskandar, E. Ecosystem, I 2022 10.1080/1747423x.2022.2115571 Correia, A. M., & Lopes, L. F. (2023). Revisiting Biodiversity and Ecosystem Functioning through the Lens of Complex Adaptive Systems. Diversity, 15, Article 895. https://doi.org/10.3390/d15080895 10.3390/d15080895https://doi.org/10.3390/d15080895 Correia, A. Lopes, L. 2023 895 10.3390/d15080895 de Groot, R., Brander, L., van der Ploeg, S., Costanza, R., Bernard, F., Braat, L. et al. (2012). Global Estimates of the Value of Ecosystems and Their Services in Monetary Units. Ecosystem Services, 1, 50-61. https://doi.org/10.1016/j.ecoser.2012.07.005 10.1016/j.ecoser.2012.07.005https://doi.org/10.1016/j.ecoser.2012.07.005 Groot, R. Brander, L. Ploeg, S. Costanza, R. Bernard, F. Braat, L. 2012 10.1016/j.ecoser.2012.07.005 Demichelis, C., Oszwald, J., Gasquet‐Blanchard, C., Narat, V., Bokika, J., Pennec, F. et al. (2020). Multidimensional Analysis of Landscape Dynamics in a Central African Forest‐Savannah Mosaic. African Journal of Ecology, 58, 692-708. https://doi.org/10.1111/aje.12750 10.1111/aje.12750https://doi.org/10.1111/aje.12750 Demichelis, C. Oszwald, J. Blanchard, C. Narat, V. Bokika, J. Pennec, F. 2020 10.1111/aje.12750 Duarte, A. C., Guerra, C. A., Cano-Díaz, C., Zeiss, R., Carvalho-Santos, C., Carvalho, R. P. et al. (2023). Effects of Protected Areas on Soil Nematode Communities in Forests of the North of Portugal. Biodiversity and Conservation, 33, 73-89. https://doi.org/10.1007/s10531-023-02732-6 10.1007/s10531-023-02732-6https://doi.org/10.1007/s10531-023-02732-6 Duarte, A. Guerra, C. Zeiss, R. Carvalho-Santos, C. Carvalho, R. 2023 10.1007/s10531-023-02732-6 Ewane, E. B., Deh-Nji, A., Mfonkwet, N. Y., & Nkembi, L. (2021). Agricultural Expansion and Land Use Land Cover Changes in the Mount Bamboutos Landscape, Western Cameroon: Implications for Local Land Use Planning and Sustainable Development. International Journal of Environmental Studies, 80, 186-206. https://doi.org/10.1080/00207233.2021.2006911 10.1080/00207233.2021.2006911https://doi.org/10.1080/00207233.2021.2006911 Ewane, E. Deh-Nji, A. Mfonkwet, N. Nkembi, L. Landscape, W 2021 10.1080/00207233.2021.2006911 Faria, D., Morante-Filho, J. C., Baumgarten, J., Bovendorp, R. S., Cazetta, E., Gaiotto, F. A. et al. (2023). The Breakdown of Ecosystem Functionality Driven by Deforestation in a Global Biodiversity Hotspot. Biological Conservation, 283, Article ID: 110126. https://doi.org/10.1016/j.biocon.2023.110126 10.1016/j.biocon.2023.110126https://doi.org/10.1016/j.biocon.2023.110126 Faria, D. Morante-Filho, J. Baumgarten, J. Bovendorp, R. Cazetta, E. Gaiotto, F. 2023 110126 ID 10.1016/j.biocon.2023.110126 Faye, D. (2015). Assessment of the Recovering Effects of Protected Areas Using a Multiple Functional Approaches: The Bamboung Model (Senegal, West Africa). International Journal of Biology, 7, 37-58. https://doi.org/10.5539/ijb.v7n3p37 10.5539/ijb.v7n3p37https://doi.org/10.5539/ijb.v7n3p37 Faye, D. Senegal, W 2015 10.5539/ijb.v7n3p37 Gakaev, R. (2023). Functional Classification of Forests: Study of Carbon Sequestration. BIO Web of Conferences, 76, Article ID: 06004. https://doi.org/10.1051/bioconf/20237606004 10.1051/bioconf/20237606004https://doi.org/10.1051/bioconf/20237606004 Gakaev, R. 2023 06004 ID 10.1051/bioconf/20237606004 Hallaj, Z., Bijani, M., Karamidehkordi, E., Yousefpour, R., & Yousefzadeh, H. (2024). Forest Land Use Change Effects on Biodiversity Ecosystem Services and Human Well-Being: A Systematic Analysis. Environmental and Sustainability Indicators, 23, Article ID: 100445. https://doi.org/10.1016/j.indic.2024.100445 10.1016/j.indic.2024.100445https://doi.org/10.1016/j.indic.2024.100445 Hallaj, Z. Bijani, M. Karamidehkordi, E. Yousefpour, R. Yousefzadeh, H. 2024 100445 ID 10.1016/j.indic.2024.100445 Hansen, M. C., Roy, D. P., Lindquist, E., Adusei, B., Justice, C. O., & Altstatt, A. (2008). A Method for Integrating MODIS and Landsat Data for Systematic Monitoring of Forest Cover and Change in the Congo Basin. Remote Sensing of Environment, 112, 2495-2513. https://doi.org/10.1016/j.rse.2007.11.012 10.1016/j.rse.2007.11.012https://doi.org/10.1016/j.rse.2007.11.012 Hansen, M. Roy, D. Lindquist, E. Adusei, B. Justice, C. Altstatt, A. 2008 10.1016/j.rse.2007.11.012 Harris Geospatial (2023). Harris Geospatial Solutions n.d.b. ENVI (Version 4.5) [Computer Software], “K-Means”. 2023 Hsiao, L. H., & Cheng, K. S. (2013). Assessing Uncertainties in Accuracy of Landuse Classification Using Remote Sensing Images. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2, 19-23. https://doi.org/10.5194/isprsarchives-xl-2-w1-19-2013 10.5194/isprsarchives-xl-2-w1-19-2013https://doi.org/10.5194/isprsarchives-xl-2-w1-19-2013 Hsiao, L. Cheng, K. Photogrammetry, R 2013 10.5194/isprsarchives-xl-2-w1-19-2013 Innocent, N. M., & Dieudonne, B. (2012). From Land Cover to Landscape Structure: Change and Fragmentation Analysis in Korup National Park, Cameroon. Journal of Geography and Geology, 4, 103-117. https://doi.org/10.5539/jgg.v4n1p103 10.5539/jgg.v4n1p103https://doi.org/10.5539/jgg.v4n1p103 Innocent, N. Dieudonne, B. Park, C 2012 10.5539/jgg.v4n1p103 Islam, M., Deb, G. P., & Rahman, M. (2017). Forest Fragmentation Reduced Carbon Storage in a Moist Tropical Forest in Bangladesh: Implications for Policy Development. Land Use Policy, 65, 15-25. https://doi.org/10.1016/j.landusepol.2017.03.025 10.1016/j.landusepol.2017.03.025https://doi.org/10.1016/j.landusepol.2017.03.025 Islam, M. Deb, G. Rahman, M. 2017 10.1016/j.landusepol.2017.03.025 Jangid, A., Gupta, M. K., & Shrivastava, V. (2023). Techniques and Challenges of the Machine Learning Method for Land Use/Land Cover (LU/LC) Classification in Remote Sensing Using the Google Earth Engine. International Journal on Recent and Innovation Trends in Computing and Communication, 11, 85-92. https://doi.org/10.17762/ijritcc.v11i7.7833 10.17762/ijritcc.v11i7.7833https://doi.org/10.17762/ijritcc.v11i7.7833 Jangid, A. Gupta, M. Shrivastava, V. 2023 10.17762/ijritcc.v11i7.7833 Jayathilake, H. M., Prescott, G. W., Carrasco, L. R., Rao, M., & Symes, W. S. (2021). Drivers of Deforestation and Degradation for 28 Tropical Conservation Landscapes. Ambio, 50, 215-228. https://doi.org/10.1007/s13280-020-01325-9 10.1007/s13280-020-01325-932152906https://doi.org/10.1007/s13280-020-01325-9 Jayathilake, H. Prescott, G. Carrasco, L. Rao, M. Symes, W. 2021 10.1007/s13280-020-01325-9 32152906 Joseph, O., Marie, O., Beri, N. G., Mana, D., Biyon, N., Bruno, J., Godlov, N., Neculina, A., & Martin, K. (2023). Diversity and Host Range of Loranthaceae along the Windward Slope of Mount Cameroon, South-West Cameroon. World Journal of Advanced Research and Reviews, 20,157-172. https://api.semanticscholar.org/CorpusId:266047773 Joseph, O. Marie, O. Beri, N. Mana, D. Biyon, N. Bruno, J. Godlov, N. Neculina, A. Martin, K. Cameroon, S 2023 266047 Karátson, D., & Thouret, J. (2025). Chapter 9 the Role of Geomorphology in Assessing Volcanic Hazards. In D. Karátson, & J. Thouret (Eds.), Volcano Geomorphology (pp. 509-581). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-77693-9_10 10.1007/978-3-031-77693-9_10https://doi.org/10.1007/978-3-031-77693-9_10 Thouret, J. 2025 10.1007/978-3-031-77693-9_10 Kazadi Mukaku, G., Mukenza, M. M., Tchowa, J. K., Malaisse, F., Kabeya, C. K., Hulu, J. P. P. M. T., Bogaert, J., & Sikuzani, Y. U. (2024). Assessment of Spatial Dynamics of Forest Cover in Lomami National Park (DR Congo), 2008-2024: Implications for Conservation and Sustainable Ecosystem Management. Ecologies, 6,Article 2. https://api.semanticscholar.org/CorpusId:275199171 Mukaku, G. Mukenza, M. Tchowa, J. Malaisse, F. Kabeya, C. Hulu, J. Bogaert, J. Sikuzani, Y. 2024 275199 2 Kettunen, M., & Brink, P. T. (2013). Social and Economic Benefits of Protected Areas. Routledge. https://doi.org/10.4324/9780203095348 10.4324/9780203095348https://doi.org/10.4324/9780203095348 Kettunen, M. Brink, P. 2013 10.4324/9780203095348 Küçüker, D. M., & Sari, B. (2021). Determining the Spatio-Temporal Dynamics of Land Cover along with Some Socio-Economic Factors in the Olur Planning Unit (Turkey). Bartın Orman Fakültesi Dergisi. Journal of Bartin Faculty of Forestry, 23, 926-940. https://api.semanticscholar.org/CorpusId:239572301 Sari, B. 2021 239572 Li, C., Li, B., Zhao, W., Jiang, J., & Tang, J. (2025). Forest Tree Breeding under the Global Environmental Change: Challenges and Opportunities. Trees, Forests and People, 20, Article ID: 100867. https://doi.org/10.1016/j.tfp.2025.100867 10.1016/j.tfp.2025.100867https://doi.org/10.1016/j.tfp.2025.100867 Li, C. Li, B. Zhao, W. Jiang, J. Tang, J. Trees, F 2025 100867 ID 10.1016/j.tfp.2025.100867 Li, E., Samat, A., Liu, W., Lin, C., & Bai, X. (2019). High-Resolution Imagery Classification Based on Different Levels of Information. Remote Sensing, 11, Article 2916. https://doi.org/10.3390/rs11242916 10.3390/rs11242916https://doi.org/10.3390/rs11242916 Li, E. Samat, A. Liu, W. Lin, C. Bai, X. 2019 2916 10.3390/rs11242916 Maghah, J. A., & Fokeng, R. M. (2021). The Degradation of the Bafut-Ngemba Forest Reserve Revisited: A Spatio-Temporal Analysis of Forest Cover Change Dynamics. American Journal of Remote Sensing, 9,33-41. https://api.semanticscholar.org/CorpusId:233624829 Maghah, J. Fokeng, R. 2021 233624 Mcdonald, R. I., Forman, R. T. T., Kareiva, P., Neugarten, R., Salzer, D., & Fisher, J. (2009). Urban Effects, Distance, and Protected Areas in an Urbanizing World. Landscape and Urban Planning, 93, 63-75. https://doi.org/10.1016/j.landurbplan.2009.06.002 10.1016/j.landurbplan.2009.06.002https://doi.org/10.1016/j.landurbplan.2009.06.002 Mcdonald, R. Forman, R. Kareiva, P. Neugarten, R. Salzer, D. Fisher, J. Effects, D 2009 10.1016/j.landurbplan.2009.06.002 Mekuate Kamga, S., Tamungang, S., Awa, T., Chmel, K., Ewome, F. L., Molua, L. L. et al. (2023). Changes in Bird Community Structure on Mount Cameroon Driven by Elevational and Vertical Gradients. Diversity, 15,Article 727. https://api.semanticscholar.org/CorpusId:259040206 Kamga, S. Tamungang, S. Awa, T. Chmel, K. Ewome, F. Molua, L. 2023 259040 727 MINEPDED (2022). NationalBiodiversityandEcosystemServicesAssessmentinCameroon. BES-Net. https://www.besnet.world/national-ecosystem-assessment/ 2022 Mohamed, M. A. (2021). An Assessment of Forest Cover Change and Its Driving Forces in the Syrian Coastal Region during a Period of Conflict, 2010 to 2020. Land, 10, Article 191. https://doi.org/10.3390/land10020191 10.3390/land10020191https://doi.org/10.3390/land10020191 Mohamed, M. 2021 191 10.3390/land10020191 Myroniuk, V., Kutia, M., J. Sarkissian, A., Bilous, A., & Liu, S. (2020). Regional-Scale Forest Mapping over Fragmented Landscapes Using Global Forest Products and Landsat Time Series Classification. Remote Sensing, 12, Article 187. https://doi.org/10.3390/rs12010187 10.3390/rs12010187https://doi.org/10.3390/rs12010187 Myroniuk, V. Kutia, M. Sarkissian, A. Bilous, A. Liu, S. 2020 187 10.3390/rs12010187 Ndikebeng, K. R., Forba, C. F., Paul Tume, S. J., Yenlajai, B. J., & Kimengsi, J. N. (2023). Land Cover Dynamics and Implications on Water Resources in Bamenda III Sub-Division, North West Region, Cameroon. International Journal of Global Sustainability, 7, 37-52. https://doi.org/10.5296/ijgs.v7i1.21031 10.5296/ijgs.v7i1.21031https://doi.org/10.5296/ijgs.v7i1.21031 Ndikebeng, K. Forba, C. Tume, S. Yenlajai, B. Kimengsi, J. Sub-Division, N Region, C 2023 10.5296/ijgs.v7i1.21031 Nebasifu, A. A., & Atong, N. M. (2019). Rethinking Institutional Knowledge for Community Participation in Co-Management. Sustainability, 11,Article 5788. https://api.semanticscholar.org/CorpusId:211415289 Nebasifu, A. Atong, N. 2019 211415 5788 Nesha, K., Herold, M., De Sy, V., Duchelle, A. E., Martius, C., Branthomme, A. et al. (2021). An Assessment of Data Sources, Data Quality and Changes in National Forest Monitoring Capacities in the Global Forest Resources Assessment 2005-2020. Environmental Research Letters, 16, Article ID: 054029. https://doi.org/10.1088/1748-9326/abd81b 10.1088/1748-9326/abd81bhttps://doi.org/10.1088/1748-9326/abd81b Nesha, K. Herold, M. Sy, V. Duchelle, A. Martius, C. Branthomme, A. Sources, D 2021 054029 ID 10.1088/1748-9326/abd81b Nzabarinda, V., Bao, A., Tie, L., Uwamahoro, S., Kayiranga, A., Ochege, F. U. et al. (2025). Expanding Forest Carbon Sinks to Mitigate Climate Change in Africa. Renewable and Sustainable Energy Reviews, 207, Article ID: 114849. https://doi.org/10.1016/j.rser.2024.114849 10.1016/j.rser.2024.114849https://doi.org/10.1016/j.rser.2024.114849 Nzabarinda, V. Bao, A. Tie, L. Uwamahoro, S. Kayiranga, A. Ochege, F. 2025 114849 ID 10.1016/j.rser.2024.114849 Pettorelli, N., Schulte to Bühne, H., Tulloch, A., Dubois, G., Macinnis‐Ng, C., Queirós, A. M. et al. (2018). Satellite Remote Sensing of Ecosystem Functions: Opportunities, Challenges and Way Forward. Remote Sensing in Ecology and Conservation, 4, 71-93. https://doi.org/10.1002/rse2.59 10.1002/rse2.59https://doi.org/10.1002/rse2.59 Pettorelli, N. Tulloch, A. Dubois, G. Ng, C. Opportunities, C 2018 10.1002/rse2.59 Politi, E., Rowan, J. S., & Cutler, M. E. J. (2016). Assessing the Utility of Geospatial Technologies to Investigate Environmental Change within Lake Systems. Science of the Total Environment, 543, 791-806. https://doi.org/10.1016/j.scitotenv.2015.09.136 10.1016/j.scitotenv.2015.09.13626521989https://doi.org/10.1016/j.scitotenv.2015.09.136 Politi, E. Rowan, J. Cutler, M. 2016 10.1016/j.scitotenv.2015.09.136 26521989 Psistaki, K., Tsantopoulos, G., & Paschalidou, A. K. (2024). An Overview of the Role of Forests in Climate Change Mitigation. Sustainability, 16, Article 6089. https://doi.org/10.3390/su16146089 10.3390/su16146089https://doi.org/10.3390/su16146089 Psistaki, K. Tsantopoulos, G. Paschalidou, A. 2024 6089 10.3390/su16146089 Qasha, V., Manyevere, A., Flynn, T., & Mashamaite, C. V. (2024). Assessing the Impact of Ecological Forest Restoration on Soil Carbon Stocks in Sub-Saharan Africa: A Systematic Review. Carbon Management, 15, Article ID: 2404409. https://doi.org/10.1080/17583004.2024.2404409 10.1080/17583004.2024.2404409https://doi.org/10.1080/17583004.2024.2404409 Qasha, V. Manyevere, A. Flynn, T. Mashamaite, C. 2024 240440 ID 10.1080/17583004.2024.2404409 Qi, G., Cong, N., Qiu, T., Rong, L., Ren, P., & Xiao, J. (2025). Evaluation of Impact of Climate Extremes on Vegetation Change in Southwest China Considering Time-Lag Effect. Global Ecology and Conservation, 58, e03497. https://doi.org/10.1016/j.gecco.2025.e03497 10.1016/j.gecco.2025.e03497https://doi.org/10.1016/j.gecco.2025.e03497 Qi, G. Cong, N. Qiu, T. Rong, L. Ren, P. Xiao, J. 2025 10.1016/j.gecco.2025.e03497 Ramachandra, T. V., Mondal, T., & Setturu, B. (2023). Relative Performance Evaluation of Machine Learning Algorithms for Land Use Classification Using Multispectral Moderate Resolution Data. SN Applied Sciences, 5, Article No. 274. https://doi.org/10.1007/s42452-023-05496-4 10.1007/s42452-023-05496-4https://doi.org/10.1007/s42452-023-05496-4 Ramachandra, T. Mondal, T. Setturu, B. 2023 No 10.1007/s42452-023-05496-4 Rees, S. E., Austen, M. C., Attrill, M. J., & Rodwell, L. D. (2012). Incorporating Indirect Ecosystem Services into Marine Protected Area Planning and Management. International Journal of Biodiversity Science, Ecosystem Services & Management, 8, 273-285. https://doi.org/10.1080/21513732.2012.680500 10.1080/21513732.2012.680500https://doi.org/10.1080/21513732.2012.680500 Rees, S. Austen, M. Attrill, M. Rodwell, L. Science, E 2012 10.1080/21513732.2012.680500 Roba, Z. R., Moisa, M. B., Purohit, S., Deribew, K. T., & Gemeda, D. O. (2025). Assessing the Impacts of Forest Cover Change on Carbon Stock and Soil Moisture Dynamics Using Geospatial Techniques: A Case Study of Nensebo Forest, Southern Ethiopia. Environmental Monitoring and Assessment, 197, Article No. 169. https://doi.org/10.1007/s10661-024-13611-0 10.1007/s10661-024-13611-039808343https://doi.org/10.1007/s10661-024-13611-0 Roba, Z. Moisa, M. Purohit, S. Deribew, K. Gemeda, D. Forest, S 2025 No 10.1007/s10661-024-13611-0 39808343 Rodrigues, A. R., Botequim, B., Tavares, C., Pécurto, P., & Borges, J. G. (2020). Addressing Soil Protection Concerns in Forest Ecosystem Management under Climate Change. Forest Ecosystems, 7, Article No. 34. https://doi.org/10.1186/s40663-020-00247-y 10.1186/s40663-020-00247-yhttps://doi.org/10.1186/s40663-020-00247-y Rodrigues, A. Botequim, B. Tavares, C. Borges, J. 2020 No 10.1186/s40663-020-00247-y Rwanga, S. S., & Ndambuki, J. M. (2017). Accuracy Assessment of Land Use/Land Cover Classification Using Remote Sensing and Gis. International Journal of Geosciences, 8, 611-622. https://doi.org/10.4236/ijg.2017.84033 10.4236/ijg.2017.84033https://doi.org/10.4236/ijg.2017.84033 Rwanga, S. Ndambuki, J. 2017 10.4236/ijg.2017.84033 Ryan, S. J., Palace, M. W., Hartter, J., Diem, J. E., Chapman, C. A., & Southworth, J. (2014). Population Pressure and Global Markets Drive a Decade of Forest Cover Change in Africa’s Albertine Rift. Applied Geography, 81, 52-59. https://doi.org/10.1016/j.apgeog.2017.02.009 10.1016/j.apgeog.2017.02.009https://doi.org/10.1016/j.apgeog.2017.02.009 Ryan, S. Palace, M. Hartter, J. Diem, J. Chapman, C. Southworth, J. 2014 10.1016/j.apgeog.2017.02.009 Sebastian, I. M., Nkwatoh, A. F., Melle, E. M., & Kamah, P. B. (2023). Contributions of Mount Cameroon National Park (MCNP) Activities to Livelihood of the Local People, South West Region Cameroon. International Journal of Biodiversity and Conservation, 15, 48-69. https://doi.org/10.5897/ijbc2023.1587 10.5897/ijbc2023.1587https://doi.org/10.5897/ijbc2023.1587 Sebastian, I. Nkwatoh, A. Melle, E. Kamah, P. People, S 2023 10.5897/ijbc2023.1587 Seifert, T., Teucher, M., Ulrich, W., Mwania, F., Gona, F., & Habel, J. C. (2022). Biodiversity and Ecosystem Functions across an Afro-Tropical Forest Biodiversity Hotspot. Frontiers in Ecology and Evolution, 10, Article 816163. https://doi.org/10.3389/fevo.2022.816163 10.3389/fevo.2022.816163https://doi.org/10.3389/fevo.2022.816163 Seifert, T. Teucher, M. Ulrich, W. Mwania, F. Gona, F. Habel, J. 2022 816163 10.3389/fevo.2022.816163 Sisay, G., Gessesse, B., Kassie, M., Kebede, B., & de Aza, C. H. (2024). Exploring Drivers of Land Use/Land Cover Transformations in Goang Watershed Ethiopia: Integrating Local Community Perceptions with Remote Sensing Data. Environmental Challenges, 17, Article ID: 101043. https://doi.org/10.1016/j.envc.2024.101043 10.1016/j.envc.2024.101043https://doi.org/10.1016/j.envc.2024.101043 Sisay, G. Gessesse, B. Kassie, M. Kebede, B. Aza, C. 2024 101043 ID 10.1016/j.envc.2024.101043 Sitotaw, T. M., Willemen, L., Meshesha, D. T., Weldemichael, M., & Nelson, A. (2025). Modelling the Impact of Ecosystem Fragmentation on Ecosystem Services in the Degraded Ethiopian Highlands. Ecological Informatics, 87, Article ID: 103100. https://doi.org/10.1016/j.ecoinf.2025.103100 10.1016/j.ecoinf.2025.103100https://doi.org/10.1016/j.ecoinf.2025.103100 Sitotaw, T. Willemen, L. Meshesha, D. Weldemichael, M. Nelson, A. 2025 103100 ID 10.1016/j.ecoinf.2025.103100 Solomon, N., Segnon, A. C., & Birhane, E. (2019). Ecosystem Service Values Changes in Response to Land-Use/Land-Cover Dynamics in Dry Afromontane Forest in Northern Ethiopia. International Journal of Environmental Research and Public Health, 16, Article 4653. https://doi.org/10.3390/ijerph16234653 10.3390/ijerph1623465331766672https://doi.org/10.3390/ijerph16234653 Solomon, N. Segnon, A. Birhane, E. 2019 4653 10.3390/ijerph16234653 31766672 Soudan-Nonault, A. (2021). Conservation of the Congo Basin.8822334. https://www.brazzavillefoundation.org/app/uploads/2022/12/Page-BassinCongo-EN.pdf Soudan-Nonault, A. 2021 Sovann, C., Tagesson, T., Vestin, P., Sakhoeun, S., Kim, S., Kok, S. et al. (2025). Land-Cover Change Alters Stand Structure, Species Diversity, Leaf Functional Traits, and Soil Conditions in Cambodian Tropical Forests. Biogeosciences, 22, 4649-4677. https://doi.org/10.5194/bg-22-4649-2025 10.5194/bg-22-4649-2025https://doi.org/10.5194/bg-22-4649-2025 Sovann, C. Tagesson, T. Vestin, P. Sakhoeun, S. Kim, S. Kok, S. Structure, S Diversity, L 2025 10.5194/bg-22-4649-2025 Stanturf, J. A., & Mansourian, S. (2020). Forest Landscape Restoration: State of Play. Royal Society Open Science, 7, Article ID: 201218. https://doi.org/10.1098/rsos.201218 10.1098/rsos.20121833489272https://doi.org/10.1098/rsos.201218 Stanturf, J. Mansourian, S. 2020 201218 ID 10.1098/rsos.201218 33489272 Sun, B., Zhang, Y., Zhou, Q., & Zhang, X. (2022). Effectiveness of Semi-Supervised Learning and Multi-Source Data in Detailed Urban Landuse Mapping with a Few Labeled Samples. Remote Sensing, 14, Article 648. https://doi.org/10.3390/rs14030648 10.3390/rs14030648https://doi.org/10.3390/rs14030648 Sun, B. Zhang, Y. Zhou, Q. Zhang, X. 2022 648 10.3390/rs14030648 Tataw, G. N., Maurice, M. E., Nkwatoh, A. F., Bumtu, K. P., Atianjoh, E. L., & Demianus, N. A. (2024). Spatiotemporal Dynamics of Land Use and Land Cover Change in the Ebo Forest Reserve, Littoral Region of Cameroon. American Journal of Natural Sciences, 5, 80-102. https://doi.org/10.47672/ajns.2272 10.47672/ajns.2272https://doi.org/10.47672/ajns.2272 Tataw, G. Maurice, M. Nkwatoh, A. Bumtu, K. Atianjoh, E. Demianus, N. Reserve, L 2024 10.47672/ajns.2272 Ustainable, O. N. S., Anagement, F. O. M., Sia, I. N. A., & Int, G. C. P. (2007). Manual on Deforestation, Degradation, and Fragmentation Using Remote Sensing and Gis. Food and Agriculture Organization of the United Nations. Ustainable, O. Anagement, F. Sia, I. Int, G. Deforestation, D 2007 van der Valk, A. G., & Warner, B. G. (2009). The Development of Patterned Mosaic Landscapes: An Overview. Plant Ecology, 200, 1-7. https://doi.org/10.1007/s11258-008-9469-z 10.1007/s11258-008-9469-zhttps://doi.org/10.1007/s11258-008-9469-z Valk, A. Warner, B. 2009 10.1007/s11258-008-9469-z Vittek, M., Brink, A., Donnay, F., Simonetti, D., & Desclée, B. (2014). Land Cover Change Monitoring Using Landsat MSS/TM Satellite Image Data over West Africa between 1975 and 1990. Remote Sensing, 6, 658-676. https://doi.org/10.3390/rs6010658 10.3390/rs6010658https://doi.org/10.3390/rs6010658 Vittek, M. Brink, A. Donnay, F. Simonetti, D. 2014 10.3390/rs6010658 Wade, C. M., Austin, K. G., Cajka, J., Lapidus, D., Everett, K. H., Galperin, D. et al. (2020). What Is Threatening Forests in Protected Areas? A Global Assessment of Deforestation in Protected Areas, 2001-2018. Forests, 11, Article 539. https://doi.org/10.3390/f11050539 10.3390/f1105053933123319https://doi.org/10.3390/f11050539 Wade, C. Austin, K. Cajka, J. Lapidus, D. Everett, K. Galperin, D. 2020 539 10.3390/f11050539 33123319 Wang, T., Yue, W., Wu, T., Xiong, J., Xia, H., & Huang, B. (2025). The Carbon Sink Conservation Areas (CSCAs) as a Land Use Strategy for Climate Change Mitigation. Sustainable Horizons, 15, Article ID: 100141. https://doi.org/10.1016/j.horiz.2025.100141 10.1016/j.horiz.2025.100141https://doi.org/10.1016/j.horiz.2025.100141 Wang, T. Yue, W. Wu, T. Xiong, J. Xia, H. Huang, B. 2025 100141 ID 10.1016/j.horiz.2025.100141 Xie, X., Ye, L., Kang, X., Yan, L., & Zeng, L. (2022). Land Use Classification Using Improved U-Net in Remote Sensing Images of Urban and Rural Planning Monitoring. Scientific Programming, 2022, Article ID: 3125414. https://doi.org/10.1155/2022/3125414 10.1155/2022/3125414https://doi.org/10.1155/2022/3125414 Xie, X. Ye, L. Kang, X. Yan, L. Zeng, L. 2022 312541 ID 10.1155/2022/3125414 Zeng, J., Chen, T., Yao, X., & Chen, W. (2020). Do Protected Areas Improve Ecosystem Services? A Case Study of Hoh Xil Nature Reserve in Qinghai-Tibet Plateau. Remote Sensing, 12, Article 471. https://doi.org/10.3390/rs12030471 10.3390/rs12030471https://doi.org/10.3390/rs12030471 Zeng, J. Chen, T. Yao, X. Chen, W. 2020 471 10.3390/rs12030471 Zeng, Y., Koh, L. P., & Wilcove, D. S. (2022). Gains in Biodiversity Conservation and Ecosystem Services from the Expansion of the Planet’s Protected Areas. Science Advances, 8, eabl9885. https://doi.org/10.1126/sciadv.abl9885 10.1126/sciadv.abl988535648855https://doi.org/10.1126/sciadv.abl9885 Zeng, Y. Koh, L. Wilcove, D. 2022 10.1126/sciadv.abl9885 35648855 Zhang, T., Hou, M., Zhou, T., Liu, Z., Cheng, W., & Cheng, Y. (2020a). Land-Use Classification via Ensemble Dropout Information Discriminative Extreme Learning Machine Based on Deep Convolution Feature. Computer Science and Information Systems, 17, 427-443. https://doi.org/10.2298/csis191222010z 10.2298/csis191222010zhttps://doi.org/10.2298/csis191222010z Zhang, T. Hou, M. Zhou, T. Liu, Z. Cheng, W. Cheng, Y. 10.2298/csis191222010z Zhang, Y., Shen, W., Li, M., & Lv, Y. (2020b). Integrating Landsat Time Series Observations and Corona Images to Characterize Forest Change Patterns in a Mining Region of Nanjing, Eastern China from 1967 to 2019. Remote Sensing, 12, Article 3191. https://doi.org/10.3390/rs12193191 10.3390/rs12193191https://doi.org/10.3390/rs12193191 Zhang, Y. Shen, W. Li, M. Lv, Y. Nanjing, E 1967 3191 10.3390/rs12193191