Boreal and hemiboreal forests constitute one of the Earth’s largest terrestrial biomes, spanning over 1.2 billion hectares across the circumboreal zone and providing critical ecosystem services, including global climate regulation through massive carbon sequestration (Gauthier et al. 2015; Pan et al. 2011). The sustainability of this vast biome is, however, increasingly threatened by accelerating climate change, which is shifting the boundaries of suitable habitats poleward at velocities unprecedented in the Holocene (Aitken & Bemmels 2016; Loarie et al. 2009). A central challenge in contemporary environmental geography is, therefore, to predict the long-term response of these forest ecosystems to novel climatic regimes and to develop spatially explicit tools for ranking their relative adaptive capacity (Khoroshev 2022).

Sustainability, defined here as the capacity of an ecosystem to maintain its structure, functions, and adaptive potential over multi-decadal timescales, is not a static property of a biome. Rather, it is deeply spatially heterogeneous, having been shaped by the prolonged co-adaptation of local populations to their historical climatic envelopes (Davis & Shaw 2001; Savolainen et al. 2007). As climate zones shift, populations of foundation tree species – the ecological engineers that define the structure and function of the entire ecosystem (Ellison et al. 2005) –increasingly find themselves in environments far removed from their evolutionary optima. This is particularly acute for species with vast continuous ranges, such as Scots pine (Pinus sylvestris L.), which extends from the Atlantic coast of Western Europe to the Sea of Okhotsk in the Russian Far East, encompassing an extraordinary diversity of climatic and edaphic conditions (Naydenov et al. 2007). The critical knowledge gap that persists is this: how does the geographical origin of such a foundation species determine the long-term sustainability of the ecosystem it supports, and can we quantify this capacity as a spatially explicit rank?

Common garden experiments, particularly long-term provenance trials, are uniquely powerful tools for addressing this question (Mátyás 1996). By growing populations from diverse geographical origins in a uniform environment, these trials effectively simulate a spatial and temporal analogue of the future climate that the local population will face. Unlike short-term physiological or genomic studies, a proper assessment of ecosystem sustainability demands a holistic, whole-life-cycle perspective. It must integrate metrics of long-term survival, structural integrity, and functional resilience to episodic stress – parameters that can only be obtained from multi-decadal field observation (Nielsen et al. 2020). Recent advances in thermodynamic geography have proposed exergy and entropy as robust, process-based indicators of ecosystem maturity and adaptive capacity at large spatial scales (Lisitsyn 2024; Lisitsyn et al. 2023; Sandlerskiy et al. 2023). However, these approaches often rely on modeled or remotely sensed data, and there is a pressing need for direct, ground-truth empirical benchmarks to validate and calibrate such models.

The Negorelskaya provenance trial of Scots pine, established in 1959 in the Republic of Belarus (53.6° N, 26.8° E), offers an exceptional, and perhaps unrepeatable, opportunity for such an empirical assessment. Located at the bioclimatic transition zone between the hemiboreal and nemoral forest belts, the trial comprises 44 climatic ecotypes originating from a vast geographical range (48°–62° N, 22°–111° E), effectively covering a significant portion of the species’ Eurasian distribution. With a continuous 66-year observation series, the experiment captures the full life-cycle ecological performance of a keystone boreal species at the mature stand stage –from establishment through the competitive exclusion phase to the onset of senescence. Earlier studies at this site have focused on traditional silvicultural evaluations, such as productivity ranking and the selection of optimal provenances for timber production (Rabko et al. 2026; Mikhailova et al. 2023). In parallel, recent advances in seed science and nursery technology (Sokolov et al. 2019; Novikov et al. 2021) have opened new possibilities for linking macroscale provenance effects to individual-level seedling performance. For instance, Novikova et al. (2023) demonstrated that the root collar diameter growth of juvenile Scots pine trees is strongly correlated with height growth and that this relationship can be traced back to spectrometric features of individual seeds. Such findings imply that the adaptive capacity observed at the provenance level in long-term trials may ultimately be expressed and potentially predicted from properties measurable at the single-seed stage. Yet, the fundamental geographical question remains unanswered: what is the spatially explicit pattern of the long-term sustainability of these model ecosystems, and which quantifiable biological metrics can serve as its reliable proxies?

We hypothesize that a significant, quantifiable geographical gradient in ecosystem sustainability exists across the East European Plain, reflecting the differential degree of maladaptation of each provenance to the trial site’s novel, shifting climate. Our specific objectives were: (1) to develop and apply a multi-criteria sustainability ranking framework that reinterprets traditional mensuration metrics through the analytical lens of ecosystem resilience and adaptive capacity; (2) to map and interpret the geographical pattern of this sustainability gradient across the sampled portion of the Eurasian range; and (3) to identify ecotypes exhibiting divergent functional traits – such as high resilience but moderate productivity –that could serve as critical genetic resources for climate adaptation strategies.

The investigation was conducted at the Negorelskaya provenance trial of Pinus sylvestris L., established in 1959 within the Negorelsky Educational and Experimental Forestry Enterprise (Dzerzhinsky district, Minsk region, Republic of Belarus; 53.6° N, 26.8° E). The site is situated in the Berezinsko-Predpolessky geobotanical district of the hornbeam-oak-dark-coniferous forest subzone. The trial initially comprised 44 climatic ecotypes spanning a latitudinal range of 48° to 62° N and a longitudinal range of 22° to 111° E. Southeastern provenances from Kazakhstan perished within the first few years, leaving a subset of viable ecotypes for long-term monitoring. By 2025, the trial occupied 4.4 ha. It stands as one of the oldest and most extensive geographical experiments for Scots pine in Eastern Europe.

A full mensurational census of all surviving ecotypes was conducted in 2025 at a stand age of 66 years. For each ecotype, the following parameters were determined according to standard forest inventory protocols (Anuchin 1977): mean stand height (H, m), mean diameter at breast height (DBH, cm), absolute stand density (N, trees ha-¹), stand basal area (G, m² ha-¹), growing stock volume (M, m³ ha-¹), and relative stocking (a dimensionless density index relative to a normative standard). Survival rate (S, %) was calculated as the ratio of the current number of living stems to the initial planting density.

For the purposes of a sustainability-oriented interpretation, these metrics were not treated merely as productivity indices. Survival rate was taken as a direct, unambiguous measure of long-term ecological persistence. Relative stocking and current stand density served as indicators of structural and functional ecosystem integrity, with low values denoting a disrupted, unstable state. The rate of biomass accumulation (proxied by mean stand height and growing stock volume) was interpreted as a measure of functional efficiency under the emerging climate normal.

For a subset of 17 representative ecotypes, increment cores were extracted at breast height from ten dominant trees per plot using a Pressler increment borer. Tree-ring widths were measured to the nearest 0.01 mm on high-resolution scans (1200 dpi) using QGIS software. To isolate the climatic signal from age-related trends, tree-ring indices (TRI) were calculated as the ratio of raw ring width to a 5-year moving average (Rumyantsev & Vorobeva 2023). The analysis focused on the most recent 15-year period (46–60 years, i.e., 2005–2025) as a direct indicator of contemporary adaptive performance.

To transition from individual metrics towards an integral, spatially comparable assessment, a composite Geographical Sustainability Rank was developed. For each of five key indicators – mean stand height, mean diameter, survival rate, relative stocking, and growing stock volume – a 6-point ranking scale was constructed. In contrast to the earlier silvicultural approach (Rabko et a, 2026), the scale intervals were explicitly designed to reflect functional thresholds of stand condition derived from the 66-year empirical distributions. A rank of ‘1’ signifies the highest sustainability (e.g., high survival, high structural integrity), whereas a rank of ‘6’ represents a state of critical dysfunction (e.g., severely depleted survival, stand collapse). The final Sustainability Rank for each ecotype was calculated as the simple arithmetic mean of its five individual indicator ranks. Consequently, lower scores correspond to higher long-term sustainability.

Twenty-four ecotypes possessing complete datasets for all five metrics were grouped into three macro-regions to test the hypothesis of a continental-scale geographical gradient. Region A – Central European Russia and Belarus (approx. 50°–55° N, central sector of the East European Plain); Region B – Northern and Northwestern Europe (Baltic States, northern European Russia, up to 61° N); and Region C – Volga-Ural and Western Siberia (east of 40° E). The mean Sustainability Rank and its standard deviation were calculated for each macro-region. The relationship between geographical origin and ecosystem sustainability was visualized using a scatter plot. For a subset of 18 ecotypes for which both sustainability ranks and recent radial growth data were available, Pearson’s correlation coefficient was calculated to assess the degree to which short-term productivity reflects long-term ecological stability.

The composite Sustainability Rank revealed a pronounced, non-linear geographical pattern (Figs. 1 and 2). The highest long-term sustainability – i.e., the lowest rank – was achieved not by the local provenance but by the Bryansk ecotype (Rank = 1.4) originating from the Central Russian Upland. It was followed by a group of southern and southwestern ecotypes (Khmelnitsky, Donetsk, Rostov, Sumy) with ranks ranging from 2.2 to 2.6. The local Minsk and Vitebsk provenances exhibited only intermediate performance, with ranks of 3.6 and 3.2, respectively.

Scatter plot displaying the Sustainability Rank (dimensionless; 1 = highest sustainability) for 24 ecotypes of Pinus sylvestris L. as a function of their longitude of origin. The quadratic fit (red dashed line) and the shaded green band (30–37° E) delineate a “sustainability optimum” on the Central Russian Upland. Ecotypes originating far north of the trial site (Region B, orange circles) and those east of the Urals (Region C, green triangles) show critically low sustainability, providing an empirical model of climate-driven maladaptation. Key ecotypes (Bryansk, Vitebsk, Arkhangelsk) are annotated.

Left panel (a): Bar chart comparing the mean Sustainability Rank (± SD) for the three macro-regions. A lower rank equates to higher long-term sustainability. The red arrow emphasizes the significant west-to-east decline. Right panel (b): Box-and-whisker plot combined with jittered individual data points. The plot reveals not only the difference in central tendencies (horizontal dashed lines indicate group means) but also the markedly higher within-group variability in macro-regions B and C, underscoring the unpredictability of adaptive outcomes for peripheral provenances. Region A – Central European Russia and Belarus; Region B – Northern and Northwestern Europe; Region C – Volga-Ural and Western Siberia.

When aggregated into macro-regions (Fig. 2), a clear west-to-east – and center-to-periphery – decline in sustainability became evident. Ecotypes from Region A (Central European Russia and Belarus) were characterized by high average sustainability (mean rank = 2.8 ± 0.9). Region B (Northern and Northwestern provenances) showed markedly lower and highly variable performance (mean rank = 4.2 ± 0.7). Notably, several stands within this region – specifically those of Arkhangelsk and Leningrad origin – were in a state of terminal collapse. Region C (Volga-Ural and Siberia) exhibited the greatest variability, ranging from the relatively sustainable Tomsk ecotype (Rank = 2.4) to the severely destabilized Chelyabinsk and Kurgan provenances (Ranks of 3.8 and 4.4, respectively), yielding a low average rank (3.8 ± 0.8).

Analysis of recent radial growth (2005–2025) uncovered divergent functional pathways towards the observed sustainability ranks. A bivariate analysis for 18 ecotypes demonstrated a moderate positive, yet statistically non-significant, linear correlation between Sustainability Rank and mean recent radial growth (Pearson r = 0.42, p = 0.08). This finding explicitly confirms that high short-term productivity does not reliably translate into high long-term ecosystem stability.

The Bryansk ecotype, the highest-ranked in terms of sustainability, exhibited a high but not dominant radial growth rate. In stark contrast, the northern Arkhangelsk and Leningrad ecotypes suffered a catastrophic decline in radial increment over the last two decades. Their mean tree-ring index for the 46–60 year period plummeted to 0.31 and 0.34, respectively, perfectly congruent with their state of functional collapse and minimal sustainability ranks. Remarkably, the local Vitebsk ecotype – despite its intermediate sustainability rank of 3.2 – displayed the single highest mean absolute radial growth over the past twenty years (0.292 cm yr-¹). This suggests a specific adaptive strategy of high metabolic investment that has not yet conferred demographic stability, highlighting a high-risk, high-reward functional syndrome.

The results of this study provide robust empirical evidence for the central hypothesis: the long-term sustainability of forest ecosystems exposed to novel climatic conditions is fundamentally governed by the geographical provenance of their foundation species. The 66-year Negorelskaya trial, situated at the bioclimatic transition between hemiboreal and nemoral zones, effectively acts as a time-machine for the central East European Plain. It demonstrates that the resilience and adaptive capacity of a stand are not predicated on contemporary productivity alone but on the complex, historically determined match between its genotype and a dynamic, non-analogue environment.

The identification of the Bryansk ecotype as the most sustainable configuration, outperforming even the local Minsk provenance, strongly suggests that the trial site is entering a new climatic envelope. Populations from slightly more southern, more continental areas appear better pre-adapted to this emerging regime. This concept of a requisite “assisted migration” distance is in close agreement with climate velocity models, which predict a northward and eastward shift of suitable habitats for temperate tree species (Aitken & Bemmels 2016). The “sustainability optimum” we observed in the 30–35° E longitudinal band corresponds to a climatic space that now best matches the projected future conditions at the trial location. This renders these central Russian ecotypes superior candidates for maintaining functional ecosystem integrity in the coming decades.

The dramatic collapse of the northern provenances (Arkhangelsk, Leningrad) serves as a stark, real-world validation of maladaptation theory. For these populations, originating at the northern edge of the species’ range (61–62° N), the “sink” effect of the new environment has exhausted their adaptive capacity, triggering a rapid disintegration of stand structure over the past decade. These failing stands no longer function as persistent carbon sinks or structurally stable ecosystems; they are transitioning into early-seral or gap-phase states. This observed collapse provides a direct empirical analogue for the fate of boreal forest stands that are unable to track their shifting climatic niche (Loarie et al. 2009). The high variability observed in Region C (Siberia) underscores the substantial risk associated with deploying material from vastly different photoperiodic and climatic regimes.

A critical finding is the decoupling of current radial growth from integral sustainability, most strikingly illustrated by the Vitebsk ecotype. Its high growth rate but only moderate demographic stability can be interpreted as a high-risk, high-reward functional strategy: robust physiological performance coexists with elevated sensitivity to episodic stressors, resulting in comparatively higher mortality. This finding serves as a cautionary note against employing simple productivity proxies – such as NDVI or modelled net primary production – as standalone indicators of ecosystem sustainability. It strongly validates our multi-criteria ranking framework, which explicitly integrates persistence and structural integrity – core components of ecosystem resilience (Nielsen et al. 2020).

Although a detailed analysis of natural regeneration lies beyond the scope of the present study, it is noteworthy that the documented patterns of mortality and structural decline have direct implications for regeneration dynamics. The progressive canopy opening in collapsing provenances (e.g., Arkhangelsk, Leningrad) could, in principle, facilitate understory recruitment of both Scots pine and associated species. However, whether such regeneration would originate from the local, potentially maladapted gene pool or from more climate-adapted sources remains an open question of both fundamental ecological and practical significance. A systematic assessment of natural regeneration across all trial variants is planned as a dedicated follow-up investigation.

The implications for spatial planning and climate adaptation are both direct and actionable. Our geographical sustainability ranks translate the outcomes of a complex, 66-year experiment into a practical, spatially explicit decision-support tool. For the hemiboreal and nemoral zones of the East European Plain, ecotypes from the Bryansk-Kursk region can be prioritized for restoration and assisted migration programs aimed at creating climate-resilient “future-proof” forests. The methodology itself provides a template for a “thermodynamic geography” (Lisitsyn, 2024) grounded in direct, whole-life-cycle biological evidence rather than solely in biophysical models. It allows policymakers and forest managers to move beyond generic prescriptions towards spatially targeted, provenance-informed conservation strategies that explicitly account for the geographical dimension of adaptive capacity.

Several limitations of the present study should be explicitly acknowledged.

First, the provenances tested are maintained as artificial, regularly spaced plantations established in 1959. The resulting stand structures may not fully replicate the dynamics of naturally regenerated, uneven-aged forest ecosystems. Consequently, the sustainability ranks we derived are most directly applicable to managed forest landscapes undergoing active restoration or assisted migration, rather than to old-growth or strictly protected forest areas.

Second, our composite index does not yet incorporate data on natural regeneration within the trial plots – a critical component of long-term ecosystem self-maintenance. Preliminary observations indicate substantial variation in the presence and vigour of understory seedlings among provenances, a pattern that warrants dedicated investigation.

Third, the ranking framework was developed for a single keystone species at a single geographical location within the boreo-nemoral ecotone. While the methodology is explicitly designed to be scalable, its transferability to other species, biomes, and bioclimatic zones requires independent validation.

Future work will focus on integrating natural regeneration dynamics into the sustainability ranking and extending the approach to other foundation tree species of the circumboreal zone.