From Planetary Boundaries to Regional Action: Remote Sensing Within Absolute Environmental Sustainability Assessments

Abstract

Accelerating environmental degradation and the continued overshoot of planetary boundaries highlight the urgent need for scientifically grounded sustainability assessments that operate across scales. While the planetary boundaries framework provides a global reference for safe environmental limits, its translation to regional and local contexts remains a methodological and practical challenge. In response, this study presents a novel scalable framework for conducting regionally explicit assessments of absolute environmental sustainability, grounded in the planetary boundaries framework. The central objective is to enable scientifically robust and globally comparable evaluations that remain sensitive to local environmental and socioeconomic conditions. The method integrates historical environmental datasets, and satellite-based Earth observation, to assess environmental impacts at the regional scale. A structured three-step process is introduced: (1) regional thresholds are derived from historical reference conditions; (2) thresholds are validated using Earth observation; and (3) environmental impacts are quantified against the validated thresholds to detect transgressions. The framework was tested in the urban core of Kiruna, northern Sweden, across five planetary boundary indicators. The results reveal substantial boundary transgressions, most notably for genetic diversity, which reaches 269 extinctions per million species-years, and for land system change, where the regional threshold is fully exceeded. These findings illustrate both the analytical value and the methodological challenges of applying planetary boundaries at fine spatial scales. Kiruna, northern Sweden, was selected as a case study due to its role as a European mining center, its location within Sámi territories, and the overlap between resource extraction and settlement. The case study illustrates the difficulty of applying planetary boundaries at fine spatial scales. This highlights the need for careful interpretation and improved calibration when downscaling global thresholds to local conditions. Ultimately, the framework reveals the potential and limitations of regionalizing planetary boundaries, highlighting the importance of methodological transparency and contextual nuance in sustainability assessment.

1. Introduction

The world is currently facing an escalating polycrisis, driven by the entanglement and nonlinear amplification of various global challenges, such the lingering socioeconomic effects of the COVID-19 pandemic, climate change, and armed conflicts [1]. While sustainability remains a widely discussed concept, its practical implementation often results in ambiguity, where almost anything or nothing can be deemed sustainable [2,3]. To overcome this challenge and enable quantifiable and verifiable sustainability assessments, the concept of Absolute Environmental Sustainability Assessment [AESA] has gained prominence. AESA is particularly relevant in the context of intensifying environmental pressures, aiming to ensure that planetary boundaries are respected and remain within observable and manageable limits [4,5,6,7]. Achieving regionally meaningful AESA requires analytical tools and models that can be applied both regionally and globally. Such tools must enable the monitoring, validation, and calculation of environmental impacts, and the identification of boundary transgressions [8,9,10]. Effective solutions require coordinated global efforts, while simultaneously acknowledging regional heterogeneity. This dual requirement underscores the need for hybrid methodologies that combine globally harmonized datasets with historical reference conditions and local or regional observations, including satellite-based Earth observation [11,12]. Ecosystem states, resource use, emissions, and socioeconomic structures vary widely in space and time, calling for both macro-level perspectives and fine-grained regional modeling [13,14]. If the cause–effect relationships in such a systemic approach are not accurately represented, the quality of results and the validity of conclusions can be severely compromised [15,16,17].

At the same time, existing studies on AESA and planetary boundaries have focused predominantly either on the global scale or on highly aggregated assessments, while comparatively little attention has been given to how such approaches can be translated into regionally explicit and empirically validated applications. This represents an important research gap, because regional sustainability assessments require not only the downscaling of global environmental limits but also methods to evaluate whether these limits are meaningful under specific local biophysical and socioeconomic conditions. In particular, the role of Earth observation in supporting such validation remains insufficiently explored.

Based on these challenges, this study addresses the following research questions: (1) How can global environmental limits derived from the planetary boundaries framework be meaningfully downscaled and validated at the regional level? (2) What added value does such regional validation provide beyond conventional global assessments? The contribution of this study to the development of regionalized AESA is three-fold. First, it presents a theoretical framework for applying planetary boundaries at the regional level by combining satellite-based Earth observation (EO) data, historical reference models, and equity-based allocation mechanisms. Second, the framework is tested in a case study of Kiruna, applying it to five planetary boundary indicators derived from globally harmonized datasets. Third, the study offers a critical reflection on the limits of satellite data and global models in capturing regional variation, identifying key methodological and data-related gaps. In this way, the study responds to the need for methodological advances that connect globally defined sustainability limits with regionally specific environmental conditions and observations. Rather than providing definitive answers, the study aims to make these challenges visible and to advance the methodological basis for translating planetary boundaries into regionally actionable strategies.

Conceptual Framework: From Planetary Boundaries to Regional AESA

Bjørn et al. (2020) [8] emphasized that effective implementation of Absolute Environmental Sustainability Assessment [AESA] must incorporate either comprehensive environmental indicators that are directly linked to ecosystem quality and resource use, allowing their integration into Life Cycle Impact Assessment [LCIA] categories, or indicators that represent planetary boundary Earth system processes. In this regard, the AESA literature has increasingly drawn on concepts and methods from Life Cycle Assessment [LCA], particularly where environmental impacts need to be translated into quantifiable indicators that can support absolute sustainability evaluations. However, the existing connections between ecosystem quality indicators and LCIA categories are limited and often indirect [18,19]. This introduces significant uncertainties and undermines the reliability of policy recommendations derived from such assessments [15,16,17]. In response, we argue that direct and explicit assessments of planetary boundaries are essential to ensure methodological robustness. This is particularly relevant where LCA-based approaches are used for absolute sustainability assessment, as the linkage between impact assessment results and Earth-system limits remains methodologically challenging. This also requires the ability to translate global biophysical thresholds into actionable, regionally relevant targets. To enhance integration into policy and governance frameworks, planetary boundary indicators must be expressed as individual or regionally defined usage budgets [13,20,21]. Effective governance of such budgets depends on converting globally defined ecological limits into spatially explicit and socially equitable allocations. This includes accounting for ecological constraints, historical responsibility, and socioeconomic disparities [22,23,24]. To operationalize equity in this context, it is crucial to define what constitutes a minimum level of human well-being. The concept of Decent Living Standards [DLSs] offers a normative benchmark for the material conditions required to satisfy basic human needs and enable social participation. These standards encompass essential dimensions such as housing, nutrition, sanitation, mobility, education, and healthcare [24,25]. DLSs provide a valuable reference point for identifying populations whose current living conditions fall below universally acceptable thresholds. Regions where populations fall below Decent Living Standards must be granted additional resources to ensure that basic human needs can be met without transgressing planetary boundaries [26,27]. Accordingly, fair allocation should incorporate both a guaranteed baseline and a flexible development budget to account for differentiated needs and capacities [3,28]. In this study, AESA is understood as an approach to assessing environmental pressures against absolute ecological limits, rather than as a claim to provide a complete or context-independent measure of sustainability at the regional scale.

2. Materials and Methods

In this study, we adopt the model assumptions and historical reference values compiled in the most recent synthesis by Richardson et al. (2023) [29] to ensure consistency with the current quantification of planetary boundaries and to establish a transparent historical baseline for the regional assessment. Specifically, historical reference values were derived from pre-industrial or low-impact conditions and translated into indicator-specific baseline values against which cur-rent Earth observation-based conditions were assessed. For green water and functional-integrity-related variables, we used outputs from the LPJmL model forced by four general circulation models [GFDL, HadGEM2, IPSL, and MIROC5]. For these indicators, the historical reference period 1661–1861 was used, and the baseline value was calculated as the mean across all years in this period. Green water was operationalized as soil moisture minus frozen soil moisture, based on LPJmL outputs. For HANPP-related ecosystem pressure, we likewise used LPJmL outputs forced by the same four circulation models. For land system change, HYDE 3.4 was used as the historical reference dataset to identify the dominant land cover at the beginning of the Holocene in the study region.

2.1. Earth Observation Data

Global satellite datasets play a central role in this study because they provide spatially consistent and regularly updated ecological information at both global and regional scales. Our analysis relies primarily on freely accessible data products from the National Aeronautics and Space Administration [NASA] (see Supplementary Materials for a full list). The identification of relevant satellite systems followed a structured procedure. First, we used the NASA Worldview platform to screen which satellite missions and products are capable of observing variables related to the selected planetary-boundary indicators [30]. Second, we verified the suitability of these missions and products through Earthdata Search, which, although less intuitive for initial screening, is required for accessing, filtering, and downloading the corresponding datasets [31]. The technical specifications, spatial resolution, and temporal coverage of the selected products are summarized in the Supplementary Materials. In the present framework, Earth observation serves two functions: first, to derive spatially explicit indicator values where satellite-based products are directly applicable, and second, to validate whether regionally downscaled thresholds are consistent with observed environmental conditions. The selected products differ in spatial and temporal resolution depending on the indicator considered. For biodiversity-related indicators (especially genetic diversity), satellite data alone are insufficient to characterize species-level responses and extinction risk. We therefore operationalize the biodiversity-related planetary boundary primarily on the basis of established biodiversity datasets rather than on Earth observation products. Specifically, we use the IUCN Red List and associated range maps to identify threatened species and their global conservation status [32,33,34]. Because IUCN range maps represent broad species distributions rather than confirmed local occurrence, their application at the municipal scale requires careful interpretation, particularly in highly modified environments such as the urban core of Kiruna. In this study, the range maps are therefore used as a screening tool to identify potentially affected threatened species within the study area, rather than as direct evidence of local presence. This approach allows the assessment to capture potential biodiversity pressure while acknowledging that actual species occurrence at fine spatial scales may be lower than suggested by the mapped range extent. In addition, global Human Appropriation of Net Primary Production (HANPP) datasets as well as MODIS satellite data are employed as a proxy for anthropogenic pressure on ecosystems (functional integrity) [35,36]. These datasets are intersected with the municipal boundary of Kiruna to derive a regional species list and to quantify the number and proportion of threatened species in line with the planetary-boundary control variable for biosphere integrity.

Of the nine planetary boundaries, this study focuses on five indicators: atmospheric aerosol loading, green water, functional integrity, genetic diversity, and land system change. These were selected because (1) they have clearly defined control variables in the most recent planetary-boundary synthesis, and (2) they can be approximated using globally consistent datasets that are either directly derived from, or can be reliably combined with, satellite-based Earth observation. In addition, these five indicators were considered particularly relevant for the regional context of Kiruna, where land-use change, ecosystem integrity, freshwater availability, and airborne emissions are closely linked to mining activities, urban development, and surrounding ecological conditions. The selection therefore reflects both conceptual compatibility with the planetary boundaries framework and practical applicability to the environmental pressures observable in the case study region.

The framework itself is designed to be transferable to other regions and economic sectors, provided that comparable historical baselines, spatial datasets, and Earth observation products are available. However, the specific choice of indicators may need to be adapted to regional ecological, social and economic conditions, as not all planetary-boundary indicators can be robustly operationalized at the same spatial. Accordingly, the present indicator set should be understood as a context-specific application of a broader methodological framework rather than a universally fixed template for all regions.

Indicators related to climate change, stratospheric ozone depletion, biogeochemical flows, ocean acidification, and novel entities are not included in our analysis. While satellite instruments increasingly provide high-quality observations of relevant atmospheric constituents, such as column concentrations of CO₂, CH₄, NO₂, and ozone, the corresponding planetary-boundary control variables are defined at the global or hemispheric scale and require integration of multiple data streams (in situ measurements, inventories, and process-based models) over long time horizons. As a result, satellite observations alone are insufficient to derive a consistent, regionally allocated share of the global climate or ozone budgets for a single municipality. Similarly, the control variables for biogeochemical flows (e.g., nitrogen and phosphorus inputs), ocean acidification (e.g., surface ocean pH and aragonite saturation state), and novel entities (e.g., synthetic chemicals and plastics) rely primarily on process-based models and monitoring data that are not yet available with adequate spatial resolution or methodological consistency for application in our regional AESA framework.

To evaluate and contextualize satellite data, this study used historical reference values from ISIMIP2b and HYDE 3.4 (Inter-sectoral Impact Model Intercomparison Project; History Database of the Global Environment) [37,38]. Figure 1 visualizes the status of the nine planetary boundaries and its indicators across three dimensions. The innermost circle reflects the current transgression status of each boundary (based on global assessments) [29]. The middle circle indicates whether the respective indicator is conceptually scalable to the regional level, e.g., through per capita allocation or equity-based schemes such as Decent Living Standards (DLSs). The outermost circle shows whether the indicator is empirically observable through Earth observation (EO) or globally harmonized datasets. While all indicators except ocean acidification are conceptually scalable, only a subset is empirically verifiable through spatially explicit data.

2.2. Case Study

2.2.1. Location

This study includes an applicational case study and the region chosen is the municipality of Kiruna, located in the county of Norrbotten in northern Sweden. Geographically, Kiruna lies above the Arctic Circle at approximately 67.85 degrees north and 20.22 degrees east. The area includes boreal forests, tundra ecosystems, and mountainous terrain, and is characterized by a cold subarctic climate (Figure 2).

For this proof-of-concept application, we restrict the study area to the urban center of Kiruna rather than the entire municipality. This urban core represents the most densely populated part of the municipality and is currently undergoing extensive transformation due to land subsidence caused by iron ore mining. Focusing on this area serves two purposes. First, it allows the framework to be tested in the part of Kiruna where anthropogenic pressures, land-use change, and socioenvironmental interactions are most concentrated and directly relevant to ongoing spatial planning and relocation processes. Second, it provides a clearly defined and policy-relevant assessment unit for evaluating whether a regionalized planetary-boundary approach can be operationalized in a local decision-making context. We acknowledge that this spatial delimitation does not necessarily correspond to the most appropriate unit for all Earth-system processes. In particular, processes such as green-water dynamics are often more meaningfully analyzed at the catchment or landscape scale. The urban core was nevertheless selected deliberately for this initial application because the aim of this study is not to define the optimal ecological unit for each indicator, but to test whether the proposed framework can be applied in a spatially explicit urban transformation context. This limitation is taken into account in the interpretation of the results. Consequently, the results presented here should be interpreted as an urban-scale illustration of the framework’s applicability rather than as a comprehensive planetary-boundary assessment for the entire region. According to the most recent demographic data, the town of Kiruna has a population of approximately 18,000 residents [40,41].

2.2.2. Climate and Environmental Setting

Mean January temperatures are well below 0 °C, while July temperatures typically remain below 15 °C, and the area experiences pronounced seasonal contrasts in daylight, from polar night in winter to midnight sun in summer [42,43]. The municipality spans a transition from boreal coniferous and birch forests to alpine tundra and mountain landscapes, with extensive wetlands and lakes that support cold-adapted species assemblages and carbon-rich soils [44].

2.2.3. Natural Resources and Land Use

Kiruna’s development is closely tied to mineral resource extraction. The town hosts one of the world’s largest underground iron ore mines, and exploration of nearby deposits has identified additional iron, phosphorus, and rare earth element resources that are of strategic importance [45]. Mining activities, together with associated infrastructure and the ongoing relocation of the urban center due to land subsidence, significantly shape local land use and environmental pressures. Beyond mining, the region supports reindeer husbandry by Sámi communities, forestry, outdoor recreation, and tourism related to Arctic landscapes and the aurora borealis [46].

2.2.4. Urban Transformation and Governance Context

Ongoing ground deformation caused by the expansion of the iron ore mine has made it necessary to gradually move large parts of the town, including the historic center and key public buildings, several kilometers to a newly planned urban area. This multi-decade transformation requires the demolition, reconstruction, or physical relocation of housing, infrastructure, and cultural heritage sites, and directly affects a substantial share of the approximately 18,000 residents. The relocation process is negotiated between the mining company, the municipal authorities, regional and national governments, and affected residents and businesses [47]. It has attracted national and international attention as an emblematic example of how resource extraction, urban planning, and long-term sustainability objectives intersect in Arctic regions. At the same time, the project has raised concerns among Sámi communities, for whom reindeer herding and land-based cultural practices are increasingly constrained by mining infrastructure and land-use change [45].

2.2.5. Raw Material Security and European Policy Context

Beyond its local significance, Kiruna also plays an important role in the broader European raw material landscape. The iron ore deposits in the region have been a cornerstone of European steel production for more than a century, and recent exploration of the Per Geijer deposit has highlighted additional resources, including phosphorus and rare earth elements, that are of strategic importance for low-carbon and digital technologies [48]. As the European Union seeks to reduce its dependence on external suppliers and to secure a stable and diversified supply of critical raw materials, mining regions such as Kiruna are increasingly framed as potential contributors to European raw material security [49].

The EU Critical Raw Materials Act and associated strategies aim to strengthen domestic extraction, processing, and recycling capacities for selected critical and strategic raw materials, while at the same time improving environmental performance and social standards along the value chain [50]. On the one hand, additional mining activity is presented as a contribution to European energy transitions and industrial resilience; on the other hand, it may intensify local environmental pressures and exacerbate conflicts over land use. These overlapping scales, from local impacts to European supply security, underscore the need for assessment approaches that can relate regional environmental pressures to globally defined planetary boundaries.

3. Results

This section presents the empirical outcomes of applying the regional Absolute Environmental Sustainability Assessment (AESA) framework to the urban core of Kiruna. We first report the regional budgets derived from the allocation of global planetary-boundary thresholds to the study area. We then present results for the five selected indicators, atmospheric aerosol loading, green water, functional integrity, genetic diversity, and land-system change, by comparing the regional pressure indicators to their corresponding safe operating budgets. For each indicator, we summarize the level of utilization or transgression relative to the allocated budget and highlight key sources of uncertainty arising from data limitations and dataset inconsistencies. A final subsection aggregates these findings across indicators to provide an overview of where the urban system remains within, or exceeds, the allocated share of the global safe operating space.

3.1. Sharing Principles of Planetary Boundaries

In order to translate planetary boundaries into actionable and equitable regional targets, globally defined limits must be downscaled using allocation approaches that reflect both environmental constraints and social needs [3]. As outlined in Section Conceptual Framework: From Planetary Boundaries to Regional AESA, this requires the definition of usage budgets that combine a guaranteed minimum with a flexible development component. In order to explore how equity considerations can be incorporated into the allocation of regional safe operating space, we introduce a simple but novel weighting scheme based on the concept of Decent Living Standards [DLSs]. The aim is not to propose a definitive allocation rule, but to test how a needs-sensitive distribution of the development part of the budget would affect the results. For each region, we conceptually distinguish between a baseline budget, which guarantees a minimum level of environmental use associated with achieving DLSs, and a development budget, which can be allocated flexibly. The novel regional allocation is then given by

$$Allocation=B_{baseline}+\left({\displaystyle \frac{ {\left(1-DL{S}_i\right)}^2}{{\sum }_j{\left(1-DL{S}_j\right)}^2}}\right)*B_{development}$$

where DLS_i ∈ [0,1] denotes the normalised DLS attainment of region i [with 1 indicating that DLSs are fully met], B_baseline is the baseline budget for region i, B_development is the total development budget to be distributed across all regions, and the sum Σ_j [1 − DLS_j]² is taken over all regions j included in the allocation scheme. DLS_j therefore represents the normalized DLS attainment of each region j in the same way as DLS_i does for region i. The squared term [1 − DLS_i]² represents an inequality-averse weighting: regions that are further below DLSs receive disproportionately larger shares of the development budget compared to a linear scheme based on [1 − DLS_i]. In other words, the marginal priority assigned to additional environmental space increases more than proportionally with the shortfall from DLSs. This reflects a normative choice to emphasize improvements for regions with the largest basic-needs gaps, while still keeping the weighting function simple and transparent. We acknowledge that other functional forms [e.g., linear weights or alternative exponents] could be used to treat the chosen exponent as one illustrative parameter, rather than a uniquely justified solution. The weighting mechanism ensures fairness by emphasizing relative need, while the sum of all allocations must remain within the respective planetary boundary. Adjustments to both the baseline and development budgets must account for local ecological conditions, technological infrastructure, and socioeconomic context [24,26,51]. Transparent and just resource allocation is therefore central to achieving long-term sustainability and planetary stability [52]. To summarize this allocation logic,

Table 1. Conceptual allocation logic to achieve the Decent Living Standards within planetary limits. [Note: This table presents the normative structure of the proposed allocation logic. Empirical application requires threshold definitions and robust indicator data.].

Allocation Step	Description	Allocation Principle
1. Baseline Budget	Each individual receives an equal minimum share to fulfill basic DLSs within planetary limits.	Egalitarian → Universal right to minimum needs
2. Development Budget	Remaining budget [if any] is allocated to those furthest below DLS thresholds.	Prioritarian → Most to the least advantaged
Constraint	Total allocation must not exceed the safe global budget [planetary boundaries].	Ecological sustainability as hard boundary
Outcome	DLS convergence without ecological overshoot.	Social justice within environmental limits

provides a conceptual overview of the two-stage distribution mechanism. It illustrates how planetary boundaries can be equitably distributed by combining universal entitlements with need-based prioritization, while remaining within ecological constraints.

In the present case study, however, high-resolution DLS information is not available at the local level for Kiruna. As a result, we do not apply the DLS-based allocation scheme quantitatively in the empirical analysis. Instead, we use a simplified allocation based on per capita and per area budgets as a pragmatic proxy that can be parameterized consistently with the available data. The resulting budgets should therefore be interpreted as equity-neutral reference values that indicate the order of magnitude of Kiruna’s allocated share under a simple proportional scheme. The DLS-based formulation and the conceptual two-stage mechanism summarized in Table 1 should be understood as a normative extension of the framework, highlighting how equity-sensitive regional budgeting could be implemented once suitable socioeconomic data become available.

Case Study of Kiruna

For spatially variable boundaries such as freshwater availability, we adjust the regional budgets on the basis of both hydrological conditions and assessment area. Hydrological capacity is derived from gridded ISIMIP2b outputs for the relevant hydrological variables, which provide values for each model grid cell. For all grid cells overlapping the study area, we multiply the ISIMIP2b value by the corresponding cell area and sum across cells to obtain an area-integrated measure of renewable water availability for the region [36,37,53,54].

Table 2. Derivations of regional planetary boundaries assessable via remote sensing and global datasets.

Planetary Boundary	Global Threshold	Upper Limit	Regional Boundary	Threshold Based on
Freshwater Change	11.1% change in green water	50% change in green water	0.7–1 m3/m3/yr	[29]; values derived from ISIMIP2b (Supplementary Materials)
Land System Change	85% of original forest cover	60% of original forest cover	1103–1443 ha	[29]; values derived from Utrecht University [55]
Atmospheric Aerosol Loading	0.1 mean annual interhemispheric AOD	0.25 mean annual interhemispheric AOD	0.1–0.25 mean annual regional AOD	[29]
Biosphere Integrity	10 Ext./MSY	100 Ext./MSY	10–100 Ext./MSY	[29]
	90% remaining NPP	80% remaining NPP	0.3–0.4 kgC/m2/yr	[29]; values derived from ISIMIP2b (Supplementary Materials)

summarizes the regionally downscaled boundary values for Kiruna.

The analysis covers a continuous observation period from 31 May 2023 to 31 May 2024. To ensure consistency in spatial coverage and data handling, a predefined analysis grid encompassing the entire urban area of Kiruna was applied (Figure 3). This spatial unit served as the basis for all satellite-based measurements and comparisons with environmental model outputs.

3.2. Freshwater Change—Green Water

The assessment of green water focused on evaluating soil moisture using data from the SMAP [Soil Moisture Active Passive] satellite mission. Given that SMAP provides data at three-hour intervals, one specific value was selected for each day to reduce the data volume (see Supplementary Materials) [31,55,56]. The observation area in Kiruna was covered by four SMAP pixels, each representing an individual measurement point. The lowest value was recorded on 16 June 2023 with 0.24 m³/m³, while the highest value was observed on 15 May 2024 with 0.38 m³/m³. A clear seasonal trend emerged, with maximum moisture levels in May and minimum levels during the summer months. Despite the seasonal consistency, the validity of some measurements must be questioned. SMAP provides soil moisture estimates even during periods that, according to its operational protocols, should not produce reliable readings. These include times of snow and ice cover, which are common in Kiruna’s climate (Supplementary Materials). When SMAP data were compared with outputs from the ISIMIP2b model, substantial discrepancies appeared, especially during December until July (Figure 4) [37] (Supplementary Materials).

On an annual average, SMAP results deviate from the ISIMIP2b model data by approximately 73–78% [values compared against the 5th and 95th percentile]. When individual measurement points are evaluated against the 5th and 95th percentiles for each respective month, it becomes evident that only during the month of November do all observations fall within the defined variability range. Across the full annual dataset, merely 21% of values lie within the corresponding monthly thresholds, resulting in 79% of the measurement points exhibiting “local deviations” beyond historical variability. Given that the exceedance range for green water is defined between 11.1% and 50% of global ice-free land area, the conditions observed in Kiruna clearly indicate a transgression of the possible regional green water planetary boundary [29].

3.3. Land System Change

We identified a total of 511 land cover pixels within the defined observation area for Kiruna [31,57]. According to the MODIS dataset, the dominant vegetation types in the assessment area are savanna, shrublands, and barren land. However, MODIS satellite-derived classifications must be treated with caution. A comparison with high-resolution imagery from Google Maps (version 26.18.01) reveals clear mismatches between observed land features, such as forest patches and water bodies, and the MODIS land cover categories [58]. This discrepancy raises concerns about the reliability of automated classification algorithms, particularly in boreal or topographically complex landscapes. Historical land cover reconstructions from the HYDE 3.4 database indicate that around the year 10,000 BCE, the area was originally dominated by “wild, remote woodlands” and “semi-natural woodlands, remote” [38,55].

Based on MODIS data, none of the original forest cover persists within Kiruna today. Based on this comparison, the indicator for remaining original forest cover in the study area is set to 0%. This outcome constitutes a full transgression of the control variable for land system change. This result should, however, be interpreted in relation to the spatial focus of the assessment. Because the study area is limited to the urban core of Kiruna, a complete loss of original forest cover primarily reflects the intensity of local land transformation within a highly modified settlement area. It does not imply that all surrounding landscapes in the wider Kiruna region have undergone the same degree of change.

3.4. Atmospheric Aerosol Loading

After filtering the MODIS data, the remaining 277 days contained at least one valid observation point within the defined analysis grid. Among the 13,052 measurement points, 48% exceeded the threshold value of 0.1 Aerosol Optical Depth [AOD], while 3% surpassed the upper boundary limit of 0.25 [31,53]. Several days exhibited persistently elevated AOD values across the grid, exceeding the critical limit of 0.25. A more nuanced picture emerges when examining spatial distribution patterns. No single measurement point recorded a mean annual value above 0.11 AOD. Most points were slightly above the threshold of 0.1, but remained below the upper limit of 0.25. The highest AOD values were spatially dispersed throughout the urban area, with a possible concentration near the main traffic artery of Kiruna. These results are broadly consistent with the MERRA-2 reanalysis dataset, which reports an annual mean AOD of 0.08 for the same period [31,59,60].

3.5. Biosphere Integrity—Human Appropriation of Net Primary Production [HANPP]

When comparing satellite-derived NPP values with historical baselines, significant discrepancies become apparent. MODIS reports an average NPP of 0.22 kgC/m²/yr for the period 2001–2023. In contrast, model-based estimates from ISIMIP2b for the historical reference period 1661 to 1861 indicate a substantially higher average of 0.38 kgC/m²/yr [37] (Supplementary Materials). This represents a reduction in potential productivity of approximately 58%. To contextualize these findings, additional comparisons were made using HANPP datasets from Matej et al. (2024) [35], which suggest that potential NPP has increased relative to historical baselines [ISIMIP2b].

Table 3. Percentage of change in HANPP [Kiruna].

NPP[act.] & HANPP[LUC,harv.]	NPP[pot.]	% of Change	Unit
[35]	ISIMIP2b	+25, +44%	HANPP
[35]	[35]	−44%	HANPP
MODISaverage	[35]	−74%	HANPPLUC
MODISaverage	ISIMIP2b	−35, −42%	HANPPLUC
MODIS 2023	[35]	−72%	HANPPLUC
MODIS 2023	ISIMIP2b	−30, −38%	HANPPLUC

summarizes changes in HANPP and its components by comparing MODIS, ISIMIP2b, and Matej et al. (2024) [35].

3.6. Biosphere Integrity—Extinctions per Million Species Years

A total of 199 terrestrial and avian species [excluding spiders, fish, and insects] have been documented in the vicinity of the town of Kiruna [32,60,61]. This regional species pool comprises 49 plant species, one amphibian, one reptile, 27 mammals, and 118 bird species. According to current occurrence records, eight of these species are no longer present in the region and are thus considered “regionally extinct” (see Supplementary Materials). This includes seven plant species and one mammal. The wild reindeer is categorized as extinct in this context because no free-ranging individuals persist in the area. To estimate the regional extinction rate, the established formula for extinctions per million species years [Ext./MSY] is applied:

$${\displaystyle \frac{Ext.}{MSY}}={\displaystyle \frac{Extinct Species}{[Observation Period*Species in a given area]}}*{10}^6$$

Assuming an observation period of 150 years [start of strong human intervention within the ecosystem], this yields a regional extinction rate of approximately 269 Ext./MSY. This value exceeds the upper uncertainty threshold proposed for the planetary boundary on genetic diversity by more than a factor of 27 [29]. Such a magnitude indicates a significant transgression of the safe operating space for genetic diversity in the Kiruna region. It also underscores the critical importance of spatially resolved assessments to identify and quantify localized losses of species that may be obscured in global averages. At the same time, this estimate is sensitive to the assumed observation period, as the Ext./MSY metric decreases when species losses are distributed across a longer time horizon. However, current datasets lack the spatial and temporal granularity needed to capture such transitions effectively [62,63,64].

3.7. Synthesis of Case Study Findings and Methodological Implications

The regional application of planetary boundaries in the Kiruna case study yielded clear exceedances for several indicators, but also revealed important nuances regarding the spatial interpretation of results. The fact that urban centers like Kiruna show full transgressions for indicators such as land system change and biosphere integrity is hardly surprising (Figure 5). Cities, by their very nature, are zones of concentrated human impact, characterized by sealed surfaces, high energy demands, and disrupted natural habitats [65,66]. Thus, findings of near-zero remaining forest cover or high extinction rates within city boundaries may reflect more the characteristics of urbanization than a failure of sustainability per se.

This raises important methodological and normative questions: Should planetary boundaries be applied uniformly across all regions, including cities? Or do they require context-sensitive interpretation that accounts for the functional role of different land types [21]? Concentrated urban development in a pristine landscape may, paradoxically, preserve more biodiversity overall than a scenario in which the same number of people is dispersed evenly across the landscape. Kiruna, situated in a sparsely populated Arctic biome, may in fact exemplify this logic. Urban land conversion in one location can functionally protect surrounding ecosystems from fragmentation and degradation. In this sense, the finding of a “complete” transgression in Kiruna is both accurate and misleading: it captures the local signal but misses the broader systemic logic [67]. A similar caveat applies to the indicator on genetic diversity. The extinction estimates in this study are based on species that are no longer observed within their documented historical distribution ranges, which include the Kiruna region. The underlying biodiversity data, primarily from the IUCN Red List and BirdLife International, are based on species distribution maps rather than point-based observations. Therefore, while the spatial unit of analysis is Kiruna city, the extinction signal stems from a broader absence of species across their mapped range that encompasses this area. The result is locally situated but not limited to local processes. Furthermore, this case study illustrates the difficulty of interpreting transgressions without accounting for functional land use and spatial heterogeneity. The framework by Richardson et al. 2023 [29] focuses on the spatial distribution of deviations beyond historic variability, yet this assumes that each spatial unit must remain within its historical envelope. The results from Kiruna challenge that assumption. They show that while the urban footprint is ecologically intense, its concentration may help maintain surrounding areas within safe limits. In light of these findings, the question becomes not whether Kiruna transgresses its planetary boundaries, which is evident from the data, but whether and how such localized exceedances can be offset or justified within a globally consistent allocation logic. A key concern emerging from this observation is the sensitivity of results to the definition of spatial boundaries. Adjusting the assessment area, even marginally, can drastically alter the outcome. Without a standardized approach to delineating regional units for AESA, the framework remains vulnerable to manipulation. This opens the door to selective boundary-setting practices, which could be used to mask unsustainable conditions or to artificially align outcomes with political goals, thus risking methodological greenwashing and undermining the credibility of the entire approach. This reinforces the need to rethink per capita or per area downscaling not as ends in themselves but as entry points for deliberative trade-offs. It also highlights the potential value of complementing production-side indicators with consumption-based assessments, including those derived from Life Cycle Assessment (LCA) [18,67,68,69]. The impacts measured in Kiruna may partially reflect mining activities, but the environmental burden of these processes is ultimately tied to global steel demand, not local consumption. For regional planners and policymakers, this implies that planetary-boundary-based assessments should not be used as rigid pass–fail instruments, but rather as decision-support tools that help identify environmental pressure hotspots, data gaps, and trade-offs between land use, infrastructure development, and ecosystem protection. These case study results, while derived from a limited spatial unit, illustrate key methodological tensions and serve as a basis for broader reflection. Finally, this case study returns us to the guiding research questions: To what extent, and by what means, can global environmental models be meaningfully downscaled and validated at the regional level? The results from Kiruna suggest that such downscaling is technically feasible, but interpretive caution is warranted. Regionalized assessments are not just smaller versions of global models; they must navigate trade-offs between spatial justice, ecological functionality, and data granularity. Their added value lies in enabling place-based decision making grounded in globally derived thresholds, while acknowledging that not all exceedances are equal, and not all boundaries can be universally enforced at fine scales.

Finally, a comparative overview of the observed values, regional thresholds, and transgression status across all five indicators is provided in

Table 4. Summary of observed values, regional thresholds, and transgression status for the five indicators in Kiruna.

Indicator	Observed Value	Regional Threshold	Upper Limit	Status	Main Uncertainty
Green Water	0.24–0.38 m3/m3	0.7–1 m3/m3/yr	50% change	Transgressed	SMAP–ISIMIP mismatch; frozen-ground effects
Land System Change	0% remaining original forest cover	1103–1443 ha	60% remaining forest cover	Fully transgressed	MODIS misclassification; urban-core effect
Atmospheric Aerosol Loading	48% > 0.1 AOD; 3% > 0.25 AOD	0.1 mean annual AOD	0.25 mean annual AOD	Exceeded	MODIS/MERRA-2 dependency
Functional Integrity	0.22 kgC/m2/yr	0.3–0.4 kgC/m2/yr	80% remaining NPP	Transgressed	Dataset divergence; proxy-based indicator
Genetic Diversity	269 Ext./MSY	10 Ext./MSY	100 Ext./MSY	Strongly transgressed	Sensitive to 150-year observation period

.

4. Discussion

The discussion of the potential for regionalizing planetary boundaries forms the conceptual starting point of this study [8,22]. Planetary boundaries were developed to enable global assessments of environmental pressures and to quantify how far humanity has moved beyond Holocene conditions [29,65,66]. AESA represents an important step toward operationalizing these boundaries by linking them to policy and governance through spatial downscaling and regional contextualization. One of the major challenges is to integrate control variables such as land system change and biosphere integrity in ways that preserve global comparability while ensuring regional relevance [4,8]. Data availability remains a key constraint, affecting both the spatial resolution and the reliability of regional assessments. Critical improvements include the harmonization of global with local datasets and the development of systematic and transparent downscaling methods [33,34,60,70] (Supplementary Materials).

Satellite data offer a foundational basis for regional environmental monitoring within a globally consistent framework. They enable large-scale observation of key ecological parameters but are also limited by geographic and climatic conditions [30,71,72,73,74,75,76]. High latitudes, prolonged darkness, snow cover, and persistent cloud layers can affect data reliability and continuity. Moreover, most satellite systems only capture surface-level parameters, which may not reflect the complexity of underlying Earth system dynamics. These limitations underline the need for hybrid monitoring systems that combine satellite imagery with on-the-ground observations to ensure accuracy and contextual relevance (Supplementary Materials) [37,53,56,77,78].

The limitations of satellite-based observations become evident in the case of atmospheric aerosol loading in the Kiruna region. The surrounding landscape, characterized by snow cover, rugged topography, and seasonal darkness, poses persistent challenges for accurate remote sensing results [57,79,80,81,82]. While ground-based validation could theoretically improve confidence in such measurements, the sparse network of in situ monitoring stations in northern Sweden and logistical constraints for placing sensors in urban traffic corridors hamper this potential (Supplementary Materials) [67]. This raises broader questions about the policy relevance of localized exceedances: if specific hotspots, such as traffic intersections or industrial zones, consistently exceed defined AOD thresholds, what forms of intervention and environmental risk are both scientifically justified and socially acceptable? Scientific assessments may identify clear boundary transgressions, yet public perception of risk often diverges from expert analysis. Risk acceptance is shaped by local experience, perceived fairness, and the trustworthiness of institutions. Consequently, regional AESA must go beyond technical diagnostics and be embedded in participatory processes that recognize community values and lived realities. In this sense, environmental thresholds are not only biophysical markers but also socio-political constructs that require deliberation, negotiation, and transparent communication.

Historical land use datasets such as HYDE 3.3 are critical for establishing baselines in regional AESA. Yet, they face known limitations, particularly for earlier time periods where data resolution is coarse and estimates rely heavily on generalized assumptions. This is especially problematic for regions with decentralized or traditional land-use systems [55,83,84,85,86,87,88]. Similar limitations affect the ISIMIP2b data, which, although designed for cross-sectoral climate modeling, relies on uniform parameters that may oversimplify environmental variability in topographically complex landscapes (Supplementary Materials) [37]. These structural simplifications are additionally mirrored in Earth observation data. These inconsistencies raise concerns about the reliability of automated classification algorithms in boreal environments, where seasonal dynamics, snow cover, and mixed vegetation types present unique challenges. While manual image interpretation could offer more accurate results, it is resource-intensive and not feasible at global scales.

A closer look at green water availability in Kiruna highlights the importance of spatial and seasonal calibration. The ISIMIP2b dataset offers only two spatial grid cells for the entire urban area, reducing the representativeness of modeled soil moisture values [37,89,90,91,92,93,94,95]. Since factors such as soil type and vegetation cover vary locally, the modeled baseline may overstate deviations observed in SMAP data [56]. A sensitivity analysis excluding implausible winter data significantly shifts the interpretation. The share of satellite observations falling within historical bounds increases when winter months are excluded, suggesting that transgression status depends strongly on seasonal data quality. This underscores the need for refined spatial resolution and better local calibration in green water assessments [37,56,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112].

Similarly, the evaluation of functional ecosystem integrity using HANPP reveals substantial uncertainties. MODIS-based NPP values for Kiruna are lower than expected under conditions of elevated atmospheric CO₂ and climate warming, which typically increase productivity. This discrepancy suggests potential errors in MODIS data [36]. The need for real-time validation of HANPP remains open to debate. Unlike rapidly fluctuating variables such as aerosol concentrations or soil moisture, HANPP is primarily governed by slow-changing land-system dynamics. As such, indicators like land-use change and seasonal soil moisture may serve as indirect but useful proxies for identifying deviations. Despite these limitations, long-term datasets, such as those provided by Matej et al. (2024) [35], offer a robust baseline for assessing functional ecosystem integrity over time. In the case of Kiruna, MODIS-derived NPP values fall significantly below the historical estimates derived from both ISIMIP2b and Matej et al. [35,36,113,114,115,116,117], suggesting a possible transgression in the region. However, additional data uncertainty persists, as the dataset by Matej et al. reports even higher baseline values than ISIMIP2b, further complicating the interpretation of observed deviations [35,37,118,119,120,121,122,123,124].

Genetic diversity, as expressed in extinctions per million species years, also illustrates the tension between global indicators and local realities. Although the data suggest a clear breach of the planetary boundary for genetic diversity, they rest on species distribution maps that may be outdated or incomplete. Arctic and sub-Arctic regions are undergoing rapid ecological shifts due to climate change, yet many biodiversity records do not reflect these dynamics. Some species classified as regionally extinct may have simply shifted ranges rather than disappeared entirely [32,33,34]. Moreover, the spatial scope of the analysis introduces an additional layer of uncertainty to the assessment. The exclusive focus on the urban core of Kiruna excludes ecologically and hydrologically connected areas from the assessment. Interregional feedbacks, such as cross-boundary water flows or species migration corridors, are therefore not captured. As a result, cumulative impacts and system-level interactions are likely underestimated, which limits the explanatory power and generalizability of the AESA framework in this particular setting. In the case of Kiruna, these methodological questions are closely linked to mining as a dominant socioeconomic driver. At the same time, this raises the question of burden shifting. Some of the regional transgressions identified here may reflect environmental costs associated with the extraction of materials that support industrial production and sustainability transitions elsewhere. From this perspective, Kiruna’s environmental pressures cannot be interpreted solely as local sustainability failures, but also as part of a broader spatial division of environmental burdens and benefits. This highlights the need to complement regional AESA with supply-chain-, consumption-, and life-cycle-oriented perspectives in future research.

Ultimately, the feasibility of regional AESA depends not only on scientific and technical resources, but also on governance capacity, institutional coordination, and local expertise. This study demonstrates the promise of regionalized approaches but also highlights the urgent need for better data integration, spatial refinement, and standardized assessment protocols. Only through these advancements can regional AESA become a credible and effective tool for managing sustainability within planetary boundaries.

5. Conclusions

This study set out to examine whether planetary-boundary-based Absolute Environmental Sustainability Assessment (AESA) can be meaningfully translated to the regional level using historical reference conditions, Earth observation, and spatially explicit datasets. Using the urban core of Kiruna as a proof-of-concept case study, the results show that such regionalization is methodologically feasible, but also highly sensitive to data quality, spatial delimitation, and indicator-specific assumptions.

Across the five selected indicators, the assessment identified substantial pressures on the allocated regional safe operating space. Particularly strong transgressions were found for genetic diversity and land system change, while green water, atmospheric aerosol loading, and functional integrity also indicated considerable deviations from the derived thresholds.

The main contribution of this study lies in advancing a transparent framework for regional AESA that combines global planetary-boundary logic with locally observable environmental conditions. Rather than treating regional assessment as a simple downscaling exercise, this study highlights the importance of validating regional thresholds against empirical observations and of making uncertainty explicit. In this sense, the Kiruna case illustrates both the analytical potential and the limitations of applying planetary boundaries below the global scale. Several limitations define the agenda for future research. The most critical data gaps concern higher-resolution biodiversity monitoring, improved seasonal and spatial calibration of Earth observation products, more robust historical baselines for regional comparison, and finer-grained socioeconomic data that would allow equity-sensitive allocation approaches to be operationalized beyond simple proportional proxies.

A key next step is to connect regional AESA more systematically with consumption-based, supply-chain, and life-cycle-oriented approaches. Strengthening these linkages would improve the ability of regional AESA to capture not only where environmental burdens occur, but also how they are driven, distributed, and potentially shifted across scales. Overall, the study shows that regional AESA can provide valuable insights for place-based sustainability assessment, but only if it is applied with methodological transparency, contextual sensitivity, and explicit recognition of uncertainty. Under these conditions, regionalized planetary-boundary assessments can become a useful tool for linking global environmental limits to regional planning and decision making.