Integrated eco-environmental risk mapping in the mountainous tropical landscapes of the Western Ghats, Kerala

Rapid tourism expansion and infrastructure growth in the Western Ghats are reshaping ecological stability in steep, topographically complex landscapes that still retain substantial forest cover. These changes are intensifying habitat fragmentation, degrading ecosystems, and placing increasing pressure on biodiversity-rich environments. Focusing on Idukki district in southern India, this study integrates the Urban Nature Access (UNA) and Habitat Risk Assessment (HRA) modules within the InVEST framework to quantify how accessibility-driven development influences habitat vulnerability between 2011 and 2025. Results reveal a widening spatial imbalance between population demand and accessible natural areas, with strongly negative urban nature balance values expanding across valley settlements and plantation corridors by 2025. Built-up land exhibits the highest mean habitat risk (R̄ = 0.42), followed by plantations (R̄ = 0.38) and croplands (R̄ = 0.34), while deciduous forests (R̄ = 0.22) and water bodies (R̄ = 0.05) remain comparatively less vulnerable. More than one-third of built-up and plantation landscapes fall within medium to high-risk categories. Spatial overlap between high-risk zones and documented landslide-affected areas during the 2018–2020 extreme monsoon events highlight the cumulative impact of tourism-driven development and associated forest conversion on steep slopes. The findings demonstrate that ecological vulnerability in Idukki is closely linked to localized forest loss and land-use transformation driven by concentrated human activities in hazard-prone terrain. These patterns highlight how steep and environmentally sensitive landscapes are increasingly affected by development pressures. This underscores the need to integrate ecosystem accessibility metrics with slope-sensitive land-use planning in tropical mountain systems.

Western Ghats of India are one of the most significant biodiversity hotspots in the world, recognized for their exceptional levels of species richness, endemism, and ecosystem services (Thakur et al., 2021; Balasubramanian and Sangha, 2023). Stretching parallel to the western coast of India, this mountain chain plays a critical role in regulating regional climate, hydrology, and biogeochemical cycles, while supporting livelihoods through agriculture, forestry, hydropower, and tourism (Raghavan et al., 2016; Balasubramanian and Sangha, 2023). Despite its ecological importance, the Western Ghats are increasingly subjected to anthropogenic pressures, including rapid urban expansion, land use and land cover (LULC) transformation, infrastructure development, plantation expansion, and resource extraction, leading to widespread habitat modification and fragmentation (Ramachandra et al., 2023). Idukki District, located in the central Western Ghats region of Kerala (Figure 1), represents a unique and ecologically sensitive landscape characterized by steep topography, dense forest cover, river basins, and protected areas interspersed with agricultural land, plantations, and expanding urban settlements (Easa and Rajesh, 2025). The district hosts diverse forest types ranging from evergreen and semievergreen forests to shola–grassland ecosystems, which provide critical habitat for endemic and threatened species (Suresh and Arr, 1998; Easa and Rajesh, 2025). At the same time, Idukki has experienced increasing anthropogenic pressure driven by population growth, tourism, hydropower development, road expansion, and urbanization, particularly around urban and peri-urban centres (Pradeep et al., 2023; Easa and Rajesh, 2025; Renu et al., 2025). These changes have altered traditional land use patterns and intensified interactions between human activities and natural ecosystems.

Land use and land cover change are one of the most significant drivers of habitat degradation and biodiversity loss globally, particularly in mountainous and tropical regions (Gupta et al., 2020; Thakur et al., 2020). In the Western Ghats, conversion of natural forests to plantations, agricultural land, built-up areas, and infrastructure corridors has resulted in habitat loss, reduced connectivity, and altered ecosystem functioning. In Idukki District, the expansion of urban areas and associated activities such as road construction, quarrying, tourism facilities, and hydropower infrastructure have contributed to substantial LULC transitions, reshaping landscape structure and increasing pressure on ecologically sensitive habitats (Kumar and Jayarajan, 2025). These transformations not only threaten biodiversity but also compromise the ecosystem services upon which local communities depend, including water regulation, soil conservation, climate regulation, and recreation (Easa and Rajesh, 2025; Kumar and Jayarajan, 2025).

Similar environmental and developmental challenges have been reported across several hill regions of India, where rapid urbanization, infrastructure expansion, and population growth increasingly interact with environmentally sensitive landscapes. Studies from Himalayan hill towns and mountain ecosystems have highlighted the growing vulnerability of these regions to hazards such as landslides, slope instability, environmental degradation, and ecosystem fragmentation (Saini et al., 2024). Hazard appraisal studies conducted in Almora emphasized the implications of unplanned urban expansion on environmental stability and future development planning (Prapanchan et al., 2024). The Landslide Susceptibility Map of Kerala, developed by the National Centre for Earth Science Studies (NCESS), further demonstrated the high sensitivity of Kerala’s mountainous terrain to both natural and anthropogenic disturbances (Pradeep et al., 2023). Similarly, multi-hazard assessment frameworks developed for communities in the Indian Himalayan Region have underscored the importance of integrating hazard exposure, vulnerability, and community resilience in regional planning (Sekhri et al., 2020). Hazard, vulnerability, capacity, and risk assessments conducted for hill areas such as Shimla have also revealed how urban growth can amplify environmental risks and place additional pressure on fragile mountain ecosystems (Lal and Sharma, 2026). Collectively, these studies demonstrate that hill regions across India face common sustainability challenges; however, most investigations primarily focus on disaster risk, hazard susceptibility and urban vulnerability, with comparatively limited attention given to cumulative ecological impacts and human–nature interactions within rapidly changing landscapes.

Urban expansion in hilly regions such as Idukki poses particular challenges due to fragile geomorphology, steep slopes, and high rainfall intensity (Kuriakose, 2010; Gupta et al., 2020; Kumar and Jayarajan, 2025). Unplanned or poorly regulated development can exacerbate environmental risks, including landslides, soil erosion, flooding, and habitat degradation. Moreover, the encroachment of urban and peri-urban land uses into forested and semi-natural areas intensifies human–wildlife interactions and disrupts ecological processes (Gupta et al., 2020; Pradeep et al., 2023). Understanding how urban growth and other anthropogenic activities spatially intersect with natural habitats is therefore essential for sustainable regional planning and conservation in the Western Ghats context (Chhachhiya et al., 2025). While numerous studies have documented this, there remains a need for integrated, spatially explicit assessments that move beyond simple land cover transitions to evaluate the cumulative risk posed to habitats by multiple human activities (Kale et al., 2016; Ramachandran et al., 2018; Balasubramanian and Sangha, 2023; Pradeep et al., 2023; Jagadesh et al., 2024). Traditional LULC analyses often quantify the extent and rate of change but provide limited insight into how different stressors interact to influence habitat condition, resilience, and long-term sustainability. In landscapes like Idukki, where diverse human activities coexist with high ecological sensitivity, cumulative impact assessments are particularly important (Abraham and Shaji, 2013; Kumar and Jayarajan, 2025).

Urban Nature Access (UNA) model within the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) Workbench provides a way to quantify the balance between the supply of accessible urban nature and the demand generated by urban populations (Silva, 2024; The Natural Capital Project, 2024). By considering the type, size, proximity, and quality of natural spaces, as well as population distribution, the model identifies areas with deficits or surpluses of nature access. In districts like Idukki, where urbanization is occurring within an ecologically sensitive landscape, understanding spatial inequalities in urban nature access is essential for equitable planning and for reducing pressure on vulnerable habitats (Aparna et al., 2024). Understanding urban nature access analysis offers a comprehensive perspective on the coupled human environment system. This analysis reveals where growing populations may experience limited access to natural spaces, potentially increasing pressure on remaining habitats. Together, these approaches enable the evaluation of trade-offs between conservation objectives and human wellbeing, supporting ecosystem-based planning strategies (The Natural Capital Project, 2024).

The Habitat Risk Assessment (HRA) model, developed under the InVEST framework, offers a structured and quantitative approach to assessing the cumulative risk to habitats from multiple anthropogenic stressors (Bastos et al., 2023; The Natural Capital Project, 2023). The HRA model evaluates risk as a function of two complementary dimensions: exposure, which reflects the degree to which habitats overlap with human activities and the effectiveness of management, and consequence, which represents habitat-specific sensitivity and resilience to those stressors (Bastos et al., 2023). By integrating spatial information on human activities, habitat distribution, and vulnerability attributes, the model enables the identification of high-risk areas, dominant drivers of risk, and relative contributions of exposure and consequence. Applying habitat risk assessment in a Western Ghats setting is particularly relevant, as ecosystems in this region often exhibit high sensitivity to disturbance but vary in resilience depending on habitat type, management regime, and landscape context. In Idukki District, protected forests, plantations, agricultural mosaics, and urban areas form a complex landscape where cumulative impacts may not be apparent through single stressor analyses (Abraham and Shaji, 2013; George et al., 2025). A spatially explicit habitat risk framework can thus provide valuable insights into how urban expansion and associated LULC changes influence habitat vulnerability across the district (Bastos et al., 2023; Fang and Xu, 2023).

Despite the extensive research on land use and land cover change, biodiversity loss, and urbanization impacts in the Western Ghats, these processes are often addressed in isolation. Most analyses focus either on ecological degradation or on urban growth dynamics, without explicitly linking human demand for nature with cumulative habitat vulnerability (Chandu et al., 2022; Prakasam and Aravinth, 2024). Spatially explicit evaluations of how multiple human activities jointly impact habitat risk remain limited. At the same time, understanding how urban expansion alters access to natural spaces is particularly important in ecologically sensitive mountain regions like Idukki. LULC analyses often focus on changes in land cover but do not account for the combined effects of urban expansion, road networks, hydropower development, and plantation expansion on habitat condition. They also overlook how restricted access to urban nature can increase pressure on the remaining natural areas. Hazards, landslide susceptibility, urban vulnerability, and environmental risks have been widely studied in Indian hill regions. However, very few studies have simultaneously examined cumulative habitat vulnerability and urban nature accessibility within a single spatially explicit framework. Consequently, the interactions between ecological integrity and anthropogenic pressures have not been comprehensively examined, particularly in biodiversity-rich mountain landscapes such as the Western Ghats.

To address this knowledge gap, the present study integrates the InVEST Urban Nature Access (UNA) and Habitat Risk Assessment (HRA) models into a spatially explicit framework to evaluate both habitat vulnerability and human access to natural spaces in Idukki District. By integrating LULC patterns, multiple anthropogenic stressors, habitat sensitivity, and population distribution, this study goes beyond traditional land cover change analysis to identify high-risk habitats, dominant drivers of risk, and spatial inequalities in urban nature access. Unlike conventional hazard and vulnerability assessments that primarily focus on human settlements, infrastructure, or disaster susceptibility, the present study explicitly evaluates cumulative ecological risks to habitats while simultaneously assessing the accessibility of natural spaces to urban populations. The combination of the InVEST HRA and UNA models within a Western Ghats landscape enables the integrated assessment of ecosystem conservation, habitat vulnerability, and urban nature accessibility within a single spatially explicit framework. This integrated approach helps identify which habitats are most vulnerable to multiple human activities and where urban populations have limited access to natural spaces. The results can guide targeted conservation actions, risk mitigation, and urban planning in Idukki District while providing insights for other mountainous regions facing similar ecological and developmental pressures.

Study area and datasets

Idukki district is located in the central part of the Western Ghats mountain range in Kerala, southern India, extending approximately between 9°15′–10°21′N latitude and 76°30′–77°25′E longitude and covering an area of about 4,356 km 2, shown in Figure 1. The district is predominantly mountainous, with elevations ranging from less than 100 m above mean sea level in the lower valleys to over 2,600 m at Anamudi Peak, the highest point in peninsular India (Karuppusamy, 2024). This pronounced altitudinal gradient results in substantial climatic, ecological, and land use heterogeneity. Idukki experiences a humid tropical montane climate with strong orographic control on rainfall, receiving high annual precipitation mainly during the southwest monsoon (June–September), often exceeding 3,000 mm (Aditya et al., 2024).

The district forms a core part of the Western Ghats biodiversity hotspot, encompassing diverse forest types such as evergreen, semi-evergreen, moist deciduous forests, and high-altitude shola–grassland ecosystems (Aparna et al., 2024). A substantial portion of the landscape is protected under conservation areas, including Eravikulam National Park (Eravikulam NP), Idukki Wildlife Sanctuary (Idukki WS), and the Periyar Tiger Reserve (PTR) (Figure 1), along with extensive reserve forests (Sivakami et al., 2023). Land use and land cover patterns are highly heterogeneous, comprising natural forests, grasslands, plantations (tea, coffee, cardamom, and rubber), agricultural lands, reservoirs, and expanding built up areas associated with urbanization, tourism, transportation, and hydropower development (Raju and Kumar, 2006; Kumar and Jayarajan, 2025). Although Idukki remains less urbanized than the coastal districts of Kerala, recent urban expansion around centers such as Thodupuzha, Munnar, and Kattappana (Figure 1) has intensified anthropogenic pressure, leading to habitat fragmentation and altered ecosystem dynamics (George et al., 2025). The combination of ecological sensitivity, steep terrain, and increasing land-use change makes Idukki a particularly suitable region for examining habitat risk and land-use dynamics in the Western Ghats.

Idukki was selected as the study area because it exhibits many of the environmental and developmental characteristics that are increasingly observed across the Western Ghats and other mountainous regions of India, including high biodiversity, protected natural ecosystems, plantation-based land use, tourism-driven development, expanding urban settlements, and growing infrastructure networks. The district, therefore, provides a representative landscape for examining the interactions between anthropogenic pressures and ecologically sensitive habitats.

The datasets provide the main environmental and socio-economic information needed for the study. The spatial data used in this study were collected for the years 2011 and 2025. A summary of the datasets and the variables used is given in Table 1.

Methodology

Habitat suitability modelling, ecological vulnerability assessment, landscape fragmentation analysis, and ecological risk assessment have been used to evaluate ecological vulnerability and human–environment interactions, including habitat suitability models, ecological vulnerability indices, landscape fragmentation metrics, and cumulative impact assessments (Kerr and Remmel, 2024; Vaissi and Mohammadi, 2025). However, many of these approaches focus either on habitat condition or human accessibility independently and often do not explicitly account for the cumulative influence of multiple anthropogenic stressors. The InVEST UNA model was selected for its precise quantification of the equilibrium between urban nature supply and population demand while accounting for accessibility constraints and competition among users. Similarly, the InVEST HRA model was chosen for its spatially explicit framework for evaluating cumulative ecological risk by integrating habitat sensitivity, exposure to multiple stressors, and habitat resilience. The combined application of UNA and HRA therefore enables a comprehensive assessment of both ecological vulnerability and human access to natural spaces within a unified framework. Alternative approaches such as habitat suitability modelling, ecological vulnerability assessments, and landscape fragmentation analyses have been widely used to evaluate ecosystem condition. However, these methods often focus on individual stressors or habitat characteristics and may not explicitly account for the cumulative impacts of urban expansion, road networks, and other land-use patterns.

The step-by-step procedure for the UNA and HRA using the InVEST Workbench is shown in Figure 2. The analysis began with the collection and pre-processing of secondary datasets, including satellite imagery, habitat and stressor layers, and a digital elevation model (DEM). Land Use–Land Cover (LULC), representing the physical and human-modified characteristics of the Earth’s surface (Skariah and Suriyakala, 2022), was derived from Landsat 7 ETM+ and Landsat 8 OLI satellite imagery with a spatial resolution of 30 m. The imagery was visually interpreted and classified into eight land-cover categories, such as evergreen forest, deciduous forest, plantation, cropland, scrubland, built-up land, barren land, and water body. LULC classification was carried out using the supervised Maximum Likelihood Classification (MLC) algorithm using Landsat datasets.

InVEST urban nature access

Urban nature accessibility in Idukki district was quantified using the Urban Nature Access (UNA) model implemented within the InVEST framework (The Natural Capital Project, 2024). The model evaluates the spatial relationship between urban nature supply and population demand, while accounting for distance-based accessibility and competition among users. All analyses were performed at the raster pixel level, and the resulting outputs were subsequently aggregated to administrative units to facilitate interpretation at the planning scale. LULC dataset derived from Landsat 7 ETM+ and Landsat 8 OLI imagery were used to identify urban nature classes, including forests, plantations, grasslands, and other green or open spaces. LULC is represented by the variable L. For each urban nature pixel j, the area of urban nature was calculated as A j (m 2), which represents the primary indicator of urban nature supply. Population data were obtained from the 2011 Census of India and projected to 2025 using the exponential population growth method. Population projections were derived using the exponential growth method, a widely used approach for estimating population between census years. Although it assumes a constant growth rate, it provides a reasonable approximation for regional-scale analyses (Goswami, 2024). The projected population was spatially allocated to raster pixels to represent the spatial distribution of population demand. Population at pixel i is denoted as P i (persons). Urban nature demand was estimated using a predefined per capita requirement D (m 2 person −1). In this study, D was set to 9 m 2 per person, following the minimum urban green space recommendation of the World Health Organization (WHO) for ensuring basic access to green spaces and supporting human wellbeing (Hassan and Jombach, 2025). This internationally recognized threshold is widely applied in urban accessibility assessments and provides a consistent benchmark for evaluating urban nature demand across different regions. However, it represents a generalized estimate and does not account for local variations in socio-economic characteristics or ecosystem service preferences. The total urban nature demand at population pixel i was calculated using Equation 1. where D i (m 2) represents the total urban nature area required to meet the demand at the population pixel i.

Spatial separation between population pixels (i) and urban nature pixels (j) was measured using Euclidean distance, denoted as d i j (m). A maximum search radius d 0 defines the catchment distance within which urban nature is considered accessible to the population. In this study, d 0 was set to 1,200 m, representing an approximate 10–15 min walking distance commonly used in urban planning and accessibility studies of green spaces (Fan et al., 2022; The Natural Capital Project, 2024). Accessibility was assessed using Euclidean distance and a fixed catchment threshold, which may differ from actual travel behaviour and transportation networks. Therefore, the findings are to be interpreted as relative measures of accessibility rather than exact representations of individual movement patterns. Accessibility to urban nature areas decreases with increasing distance and is represented using an exponential distance decay function, as shown in Equation 2. where β is the distance decay coefficient, which controls the rate at which accessibility declines with increasing distance from urban nature locations. The exponential decay function assumes that the probability of visiting green spaces decreases progressively with increasing travel distance, which reflects commonly observed behaviour in recreational accessibility studies (Handy and Niemeier, 1997). Pixels with d i j > d 0 were excluded from the accessibility analysis. This exclusion is based on the assumption that urban nature areas located beyond the defined catchment distance are less likely to contribute to routine access for recreation and wellbeing. The selected threshold emphasizes local accessibility that can reasonably be achieved on foot and is consistent with the conceptual framework of the InVEST UNA model. However, this assumption may underestimate the contribution of large recreational destinations located beyond the catchment but occasionally visited by residents.

Urban nature supply was estimated using the Two-Step Floating Catchment Area (2SFCA) approach, which accounts for both distance-based accessibility and competition among users. In the first step, the urban nature–population ratio for each urban nature pixel j was calculated using Equation 3. where R j (m 2 person −1) represents the effective urban nature supply capacity of pixel j, A j is the urban nature area derived from the LULC layer, and P i represents the population demand within the catchment. In the second step, the per capita urban nature supply available to each population pixel was computed using Equation 4. where S i (m 2 person −1) represents the accessible urban nature supply per person at population pixel i. The per capita balance between urban nature supply and demand at population pixel i was calculated using Equation 5. where B i (m 2 person −1) indicates surplus (B i > 0) or deficit (B i < 0) in urban nature accessibility. To express this balance in terms of total area, the population-weighted balance was computed using Equation 6. where B i t o t a l (m 2) represents the total surplus or deficit of urban nature associated with population pixel i.

The total urban nature area effectively accessible to the population pixel i, accounting for distance decay effects, was calculated using Equation 7. where A i a c c (m 2) represents the distance-weighted urban nature area available within the defined catchment. This metric provides an integrated measure of accessibility and recreational potential consistent with the InVEST Urban Nature Access framework (The Natural Capital Project, 2024). All continuous UNA model outputs were classified into discrete categories using the Jenks natural breaks classification method, which identifies natural groupings in the data and reflects the inherent spatial clustering of accessibility patterns.

Habitat risk assessment (HRA)

Habitat Risk Assessment (HRA) was performed using the InVEST Habitat Risk Assessment model to quantify the cumulative impacts of anthropogenic stressors on natural and semi-natural habitats. The model conceptualizes ecological risk as a function of exposure to stressors and the ecological consequence of that exposure, consistent with established cumulative impact assessment frameworks (Bastos et al., 2023). The resulting risk estimates are relative rather than absolute and allow spatial comparison of vulnerability across habitats and locations. The analysis was conducted at the raster pixel level using spatial datasets representing habitats and anthropogenic stressors. Habitat layers were derived from the LULC map and included water bodies, built-up areas, barren land, scrub land, cropland, plantations, and deciduous forest. Stressor layers included hydropower dams, urban expansion, and road networks, which represent major anthropogenic pressures influencing ecosystem structure and functioning in the study region. To represent the spatial influence of stressors, buffer distances were applied to each stressor layer based on ecological impact zones reported in previous studies. Hydropower dams and urban expansion were assigned a 1,000 m buffer, representing the spatial extent of hydrological alteration and urban ecological footprint, while road networks were assigned a 500 m buffer to represent road-induced disturbance and habitat fragmentation effects. The selected buffer distances were derived from published studies and represent generalized zones of influence. Actual ecological responses may vary depending on local environmental conditions and habitat characteristics. For each habitat–stressor interaction, two components of risk were quantified, exposure (E) and consequence (C). Exposure represents the degree to which a habitat is subjected to a stressor and incorporates criteria such as disturbance frequency, temporal overlap, and management effectiveness. Consequence represents the sensitivity of a habitat to disturbance and includes ecological attributes such as habitat structural change, recovery time, recruitment rate, and landscape connectivity. Exposure for habitat j under stressor k at pixel l was calculated as a weighted average of exposure criteria, as expressed in Equation 8 (The Natural Capital Project, 2023). where e i j k l represents the exposure rating assigned to criterion i, w e i denotes the weight assigned to the exposure criterion, q e i represents the data-quality score for the criterion, and N E is the total number of exposure criteria.

The ecological consequence for habitat j under stressor k at pixel l was calculated using a weighted average of consequence criteria, as shown in Equation 9. where c i j k l represents the consequence rating assigned to criterion i, w c i denotes the weight of the consequence criterion, q c i represents the data-quality score, and N C is the number of consequence criteria. Exposure and consequence ratings were assigned using literature-based and expert-informed criteria, introducing a degree of uncertainty into the assessment (Li et al., 2022). However, this approach remains suitable for identifying relative spatial patterns of habitat vulnerability (Arkema et al., 2014).

The risk score for each habitat–stressor–pixel combination was then calculated using a Euclidean risk formulation that integrates exposure and consequence while accounting for distance-based stressor influence. The risk score was calculated using Equation 10. where R j k l represents the risk score for habitat j from stressor k at pixel l, and D k l is a distance-decay function representing the decline in stressor influence with increasing distance from the stressor source. The HRA model assumes additive effects among multiple stressors and may not fully capture synergistic or non-linear ecological responses. Consequently, the resulting risk values should be interpreted as relative estimates for spatial comparison rather than absolute measures of ecological condition. To facilitate interpretation of the results, continuous risk scores were classified into four categories based on fractions of the maximum possible risk score (R max). The classification scheme used to categorize risk levels is defined in Equation 11.

Pixels where no spatial interaction occurred between a habitat and a stressor (R j k l = 0) were classified as No Risk, indicating the absence of stressor influence. Habitat and stressor layers used in the analysis are summarized in Supplementary Table S1, and the scoring matrix for exposure and consequence criteria is provided in Supplementary Table S2.

Land use pattern change

LULC pattern of Idukki District (Figure 3) reveals significant transformations between 2011 and 2025, reflecting broader landscape-level changes within the ecologically sensitive Western Ghats. Evergreen forest remained the dominant land-cover class; however, its extent declined from 2,350.18 km 2 (53.95%) in 2011–2,200.20 km 2 (50.49%) in 2025, representing a net loss of approximately 150 km 2. This decline is consistent with the studies of Reddy et al., 2016; Rai et al., 2025 that attribute forest loss to sustained anthropogenic pressures. However, the magnitude of reduction observed here highlights the continued vulnerability of primary forests even within ecologically sensitive highland landscapes. Deciduous forest also decreased from 740.66 km 2 (17%) to 697.09 km 2 (16%), aligning with reported patterns of secondary forest degradation, although the relatively smaller decline compared to evergreen forests suggests differential responses among forest types.

Plantation area increased from 520.82 km 2 (11.95%) to 600.24 km 2 (13.77%), supporting earlier findings from the Western Ghats by Rege and Lee, 2022, which document the expansion of monoculture plantations. The increase of nearly 80 km 2 over the study period indicates that plantation-driven land transformation remains a dominant process, and the magnitude observed at the district level suggests a more pronounced shift than typically generalized in broader regional assessments. Cropland declined from 304.98 km 2 (7%) to 261.41 km 2 (6%), which is consistent with earlier observations of agricultural contraction in the region by Fox et al., 2017. However, the extent of decline identified in this study points to a relatively rapid transition, likely driven by reduced agricultural viability and conversion to plantation systems, reflecting ongoing socio-economic restructuring in highland areas. Anthropogenic land-use expansion is particularly evident in the increase of built-up land, which nearly doubled from 87.14 km 2 (2%) in 2011 to 174.27 km 2 (4%) in 2025. While similar trends of urban growth and infrastructure expansion have been reported by Chhachhiya et al., 2025, the scale of increase observed here underscores the intensification of infrastructure expansion in Idukki, contributing to landscape fragmentation.

Barren land increased from 87.14 km 2 (2%) to 121.99 km 2 (2.8%), indicating the land degradation due to quarrying and construction activities. Scrubland also expanded from 152.49 km 2 (3.5%) to 170.92 km 2 (3.92%), reflecting the emergence of transitional or degraded landscapes following forest disturbance. While such increases have been noted in degraded ecosystems, the present results provide clearer district-level evidence of these intermediate land-cover states. Water bodies showed a slight increase from 112.41 km 2 (2.58%) to 130.70 km 2 (3%), reflecting hydrological modifications in the region. The observed change likely reflects reservoir expansion and regulated water systems, and this study further highlights how such changes manifest spatially within a mountainous terrain. The observed change in the area of each land use class, shown in Figure 4, also sheds light on a gradual shift from natural forest ecosystems toward plantations and built-up areas, raising concerns regarding biodiversity conservation, hydrological stability, and sustainable land management in the Western Ghats region.

Urban nature access

Urban nature demand

Urban nature demand patterns in Idukki district (Figure 5) both align with and extend existing evidence on spatial heterogeneity in human–environment interactions in the Western Ghats. In 2011, demand values ranged from 139.44 to 174.31 across much of the district, indicating a relatively dispersed dependence on nearby natural spaces. This pattern is consistent with the study of Gupta et al., 2020, which highlights the integration of ecosystem use within rural livelihood systems. Higher demand values (212.8–312.6) were concentrated in the central and southern regions, where plantation-dominated landscapes and clustered settlements are prevalent, supporting findings that link intensified ecosystem dependence to economically active and accessible areas (Balasubramanian and Sangha, 2023). Lower demand values (5.95–105.76) in northern and remote areas reflect reduced accessibility and weaker integration of natural resources into daily activities.

By 2025, the spatial pattern shows a clear shift toward greater concentration of demand. High-demand zones (212.9–312.6) expanded across the southern and south-eastern parts and extended into central regions. While demand in 2011 was primarily associated with livelihood support, the 2025 distribution suggests increasing pressure on natural spaces associated with population concentration, settlement expansion, and changing land-use patterns. The current findings explicitly illustrate the transition of ecosystem demand at the district level, emphasizing the influence of improved road connectivity, expanding plantation systems, and increasing settlement density on demand distribution, as previously observed in regional studies (CITATION). These results indicate that urban nature demand in Idukki is transitioning from a dispersed, livelihood-oriented pattern to a more spatially clustered and functionally diversified structure. The concentration of demand in specific parts of the district suggests increasing localized pressure on natural systems, emphasizing the need for targeted management strategies that address emerging demand hotspots and their potential ecological impacts.

Urban nature supply

The spatial patterns of urban nature supply in Idukki district (Figure 6) reflect both the distribution and temporal changes in natural areas that support ecosystem service delivery. The relative stability in overall supply between 2011 and 2025 observed in this study aligns with previous findings that forest-dominated mountainous regions can retain substantial vegetation cover despite increasing anthropogenic pressures. However, the spatial reconfiguration and emerging fragmentation identified here reinforce the view that ecosystem service provision may decline even when total green cover remains relatively unchanged. The distribution of moderate supply values (3.27–4.57) across much of the district in 2011, along with the concentration of high supply zones (8.4–13.17) in the northern, northeastern, and western high-elevation forested areas, is in agreement with Anitha et al., 2024, who highlight the critical role of intact forest patches in sustaining regulating and supporting ecosystem services. The present analysis further substantiates this by demonstrating the continued functional importance of these ecologically intact landscapes. The lower supply values observed in the central and southern regions, characterized by plantation-dominated land use. The study by Pradheep et al., 2021 highlights that plantation systems, although contributing to overall vegetation cover, differ structurally from native forests and consequently provide reduced habitat heterogeneity and ecosystem service capacity, particularly for biodiversity support. The results of this study reaffirm this distinction within the context of Idukki.

The temporal changes observed by 2025, specifically the fragmentation and contraction of high supply classes alongside the expansion of moderate supply areas, are also in line with previously reported trends in the Western Ghats. Studies by Kuriakose 2010; Badhe et al., 2023 document similar processes driven by land-use transitions, including plantation expansion, infrastructure development, and increasing accessibility, which contribute to forest fragmentation, intensified edge effects, and disruption of ecological processes. The present findings provide spatially explicit evidence of these dynamics at the district scale. The shift from high to moderate supply values suggests a decline in ecosystem service quality rather than a complete loss of vegetation, supporting earlier observations of functional degradation in human-modified tropical landscapes. This indicates that while the structural presence of green cover may persist, its ecological effectiveness is being progressively reduced.

Per capita urban nature balance

The spatial distribution of per capita urban nature balance in Idukki district (Figure 7) reflects the combined influence of population-driven demand and the availability and accessibility of natural spaces between 2011 and 2025. These patterns highlight the extent to which ecosystem service supply is able to meet localized human demand, a relationship that has been widely emphasized in studies of human–environment interactions in the Western Ghats. The concentration of strongly negative balance values (−8.73 to −4.02) in the central and southern regions in 2011, particularly within settlement clusters situated in plantation-dominated valley landscapes. Prior studies have observed that extensive forest cover does not necessarily translate into equitable access to ecosystem services due to legal, topographic, and socio-economic constraints (Raghavan et al., 2016; Gupta et al., 2020). The present results reinforce this perspective by showing that, despite the forest-rich setting of Idukki, these areas experience pronounced deficits in per capita urban nature availability. The less negative to near-neutral balance values observed in peripheral forest regions follow a similar pattern that is primarily driven by low population density rather than improved accessibility, indicating persistent spatial inequities in nature benefits.

The role of increasing population density, settlement expansion, and infrastructure development identified in this study is associated with declines in per capita ecosystem service availability under conditions of stable or minimally changing forest cover. Accordingly, the temporal changes observed by 2025, particularly the expansion of strongly negative balance values across much of the district, reflect similar processes operating in Idukki. The influence of built-up expansion and infrastructure development, particularly in rapidly developing settlements, reflects how localized growth intensifies demand and limits access to natural spaces. Similarly, the effects of plantation expansion and intensification observed in this study indicate that the replacement of heterogeneous landscapes with monoculture systems reduces ecosystem service diversity and accessibility. The findings indicate that such land-use transitions contribute to measurable declines in per capita urban nature balance. Similar patterns have been noted by Kumar and Jayarajan, (2025), showing that fragmentation, infrastructure expansion, and concentrated demand reduce per capita ecosystem service availability in mountainous regions. The observed deterioration in balance values suggests that land-use configuration and accessibility constraints, rather than large-scale forest loss, are the dominant drivers of emerging imbalances in Idukki.

Total population-based urban nature balance

The spatial patterns of urban nature balance based on total population in Idukki district (Figure 8) are consistent with existing ecosystem service assessments in the Western Ghats, while also providing detailed spatial insights at the district level. The observed disparities between cumulative population demand and the availability of accessible natural areas support earlier studies emphasizing that ecosystem service pressure is strongly shaped by population concentration and landscape configuration. The concentration of strongly negative balance values (−231.32 to −110.75) in the central, eastern, and southern regions in 2011, particularly in densely populated valley settlements and plantation-dominated landscapes. While settlement clustering along valleys and transport corridors can intensify ecosystem service demand (Balasubramanian and Sangha, 2023), the present study highlights that in Idukki, cumulative population pressure is especially concentrated within spatially constrained and fragmented green spaces, identifying localized hotspots of urban nature demand. In contrast, diminished negative balance values in the western and interior forest regions reflect lower population density and reduced aggregate demand, despite extensive natural vegetation. By 2025, strongly negative balance values had intensified and spread across a larger area. Increasing population, settlement expansion, and infrastructure development, combined with the uneven distribution of natural areas, decreases the ability of ecosystems to meet growing demand. The lower imbalance in interior forests suggests that ecosystem service balance is driven primarily by population distribution, not by forest extent.

Accessible urban nature

The spatial pattern of accessible urban nature in Idukki district (Figure 9) reveals a clear spatio-temporal shift in the physical and functional accessibility of natural areas between 2011 and 2025. These changes are strongly influenced by terrain constraints and development pathways characteristic of the Western Ghats, where steep topography, valley-based settlements, and linear infrastructure shape human–nature interactions. The widespread distribution of high accessibility values (4.68–5.31) in 2011, particularly across the central and southern regions, where settlements were embedded within forest–plantation mosaics. Mixed land-use systems and relatively lower levels of infrastructure development allow closer integration between human settlements and natural areas, enhancing functional access to ecosystem services. Similarly, the lower accessibility observed in northern and north-eastern regions corresponds with areas where earlier development had already begun to modify landscape connectivity. The temporal changes observed by 2025, particularly the contraction and fragmentation of high accessibility zones and the expansion of lower accessibility classes, are consistent with patterns reported in previous studies. Bharath et al., 2014; Ramachandra et al., 2023 document that road expansion, settlement growth, and land-use transitions in the Western Ghats contribute to habitat fragmentation and reduced connectivity. In line with this, the present study shows that development concentrated along valley floors and transport corridors has increased physical separation between settlements and natural areas, thereby reducing accessibility. The influence of infrastructure expansion, particularly road development and settlement growth, observed in this study is evident. Linear infrastructure can create barriers, increase edge effects, and reduce functional accessibility to ecosystem services in mountainous landscapes (Thomas et al., 2021). This study also shows how these processes are spatially expressed in Idukki, particularly within rapidly developing nodes. Despite increasing isolation, interior forest regions maintain relatively high accessibility values. This highlights the distinction between ecological presence and functional accessibility. However, these areas contribute less effectively to meeting human demand due to spatial disconnection.

The spatial distribution of habitat risk in Idukki district (Figure 10) reflects the combined effects of urban nature accessibility and imbalance, which together determine how human pressure is redistributed across the landscape. Figure 10 illustrates that regions of high habitat risk are spatially concentrated along valley floors, plantation belts, and major development corridors, where accessibility, transportation networks, and settlement expansion have intensified human–environment interactions. Large tracts of ecologically significant forest remain functionally inaccessible due to terrain constraints and legal protection, resulting in a spatial mismatch between ecosystem availability and human use. The research conducted by Raghavan et al., 2016 indicates that, improved connectivity in mountain ecosystems results in localized ecological stress, a pattern that is consistent with the accessibility-driven concentration of pressure observed in this study. The HRA results, summarized in Table 2 and spatially represented in Figure 10, quantitatively support this pattern. Built-up areas exhibit the highest mean risk (R̄ = 0.42), with 12.5% of their extent classified as high risk, reflecting concentrated anthropogenic pressure in highly accessible yet environmentally sensitive zones. This pattern corresponds with the findings of Anitha et al., 2024; Aparna et al., 2024 where infrastructure-driven built-up expansion has been shown to increase fragmentation, edge effects, and disturbance intensity, resulting in elevated habitat risk.

Plantation landscapes, which dominate mid- and high-elevation regions, also show elevated risk levels (R̄ = 0.38), with nearly 35% of the area falling under medium to high-risk classes, as evident in Figure 10ii. These zones also correspond to areas experiencing declining per capita urban nature balance, indicating spatial overlap between increasing human demand and ecologically modified landscapes. Study by Kanda et al., 2025 in the Western Ghats demonstrate that plantation systems reduce habitat connectivity and ecological resilience, thereby contributing to increased vulnerability despite maintaining vegetative cover. Contiguous forest cover has been reported to act as a buffer against ecological degradation in the Western Ghats (Gupta et al., 2020). Similarly, deciduous forest regions in Idukki, characterized by limited accessibility and lower human disturbance, exhibit substantially lower habitat risk (R̄ = 0.22), with nearly 48% of their area classified as no risk, as reflected in the low-risk zones (Figure 10).

Land-use disturbances, particularly those associated with construction activity and agricultural expansion, contribute to ecological degradation and habitat fragmentation in the Western Ghats. The relatively high habitat risk observed in barren land (R̄ = 0.36) and cropland (R̄ = 0.34) (Figure 10) indicates that these land-cover types represent disturbed or transitional states associated with elevated ecological risk. These patterns suggest that habitat risk in Idukki is strongly governed by the spatial reorganization of human activities driven by accessibility and land-use change (Raghavan et al., 2016; Anitha et al., 2024). The clustering of high-risk zones in accessible and economically active regions, as illustrated in Figure 10, highlights the need for regulating infrastructure expansion, maintaining vegetation cover, and preserving landscape connectivity to reduce ecological degradation and hazard vulnerability in this ecologically fragile region.

The results further indicate that habitat risk is influenced not only by the extent of urban expansion but also by the spatial pattern of development and land-use management. High-risk areas are concentrated where built-up expansion, transportation infrastructure, and ecologically sensitive habitats intersect. This suggests that ecological vulnerability may be reduced through improved planning approaches that incorporate habitat conservation into development decisions. Measures such as regulating development in environmentally sensitive zones, maintaining ecological corridors, conserving remaining natural habitats, and integrating green infrastructure into local planning frameworks could help reduce habitat risk while supporting future development needs. The spatial identification of high-risk areas provided by this study can therefore support evidence-based planning and sustainable landscape management in mountain regions.

Implications and future directions

The results indicate that ecological vulnerability in Idukki is shaped by the interaction between accessibility, population pressure, and land-use configuration. The observed spatial concentration of habitat risk in accessible and rapidly developing areas suggests that future planning interventions should focus on balancing development needs with ecological conservation. The integration of habitat risk and urban nature accessibility assessments can support the identification of priority areas for conservation, ecological restoration, and sustainable land management.

Future studies could extend the present framework by incorporating additional environmental and socio-economic variables including habitat connectivity, vegetation quality and ecosystem service preferences that influence habitat condition and accessibility patterns. Integrating detailed information on demographic characteristics, ecosystem service preferences, and alternative development scenarios would improve the representation of human–environment interactions. Scenario-based assessments evaluating different land-use planning and development pathways could also provide valuable insights into the long-term implications of urban expansion and landscape transformation.

The current representation of nature demand is generalized and does not account for socio-economic heterogeneity or variations in ecosystem service requirements (e.g., recreational versus provisioning uses). Future studies could refine demand estimation by incorporating such distinctions to better represent localized human–environment interactions. From a management perspective, the identification of areas where high human pressure overlaps with ecologically sensitive habitats can support targeted regulation of infrastructure expansion and land-use change. Maintaining vegetation cover, improving landscape connectivity, and protecting ecologically important areas can further enhance ecosystem resilience. Further validation through field-based assessment of habitat condition and ecosystem service provision would strengthen the robustness and applicability of the HRA results.

This study demonstrates that ecological vulnerability in Idukki is shaped more by the spatial configuration of accessibility, population concentration, and land-use intensity than by the mere extent of forest cover. The results show that, although overall forest cover remains relatively stable between 2011 and 2025, high urban nature supply areas have become increasingly fragmented, with a transition toward moderate supply classes, indicating a decline in ecosystem service quality rather than outright loss. At the same time, both per capita and total population-based urban nature balance reveal a clear intensification of negative values, particularly in valley settlements and plantation-dominated regions, reflecting growing pressure from concentrated population and infrastructure expansion. The accessibility analysis further indicates a contraction and fragmentation of highly accessible natural areas, while habitat risk assessment highlights the clustering of moderate to high-risk zones in regions influenced by road expansion, plantation agriculture, and settlement growth. Together, these findings reveal a pronounced spatial mismatch: ecologically valuable forest areas remain physically intact but are often functionally inaccessible, whereas accessible landscapes experience concentrated and unsustainable levels of human pressure. This imbalance is most evident along valley corridors, plantation belts, and rapidly developing settlement clusters, where human–environment interactions are intensified and ecological degradation is more likely to occur.

From a planning perspective, these results emphasize the need for targeted and spatially explicit interventions. Regulating infrastructure expansion, particularly roads and built-up growth in ecologically sensitive and high-risk zones, is critical to minimizing further fragmentation. Enhancing landscape connectivity through the conservation of ecological corridors and restoration of degraded patches can improve both ecosystem resilience and functional accessibility. In plantation-dominated landscapes, management strategies should focus on increasing structural diversity and ecological functionality, for example, through the integration of native vegetation buffers and the reduction of intensive land-use practices. Future development activities should be guided by ecological considerations to avoid excessive pressure on environmentally sensitive areas.

The findings further suggest that ecological vulnerability is influenced not only by the extent of urban expansion but also by the spatial pattern of development and land-use management. High-risk areas were concentrated where built-up expansion and transportation infrastructure intersected with ecologically sensitive habitats, indicating the importance of integrating ecological information into planning and decision-making processes. The spatial outputs generated in this study can support evidence-based planning by identifying priority areas for conservation, ecological corridor protection, and development regulation. Incorporating habitat risk and urban nature accessibility considerations into local land-use planning frameworks can help balance development needs with long-term ecosystem sustainability in mountain regions.