The study, published in Discover Ecology, focuses on the Klinské rašelinisko peatland in the Orava region of northern Slovakia, a small bog nestled in the foothills of the Western Carpathians. Slovakia sits at the southern edge of Europe’s peatland distribution, which makes its bogs naturally small, isolated and fragmented — and exceptionally vulnerable to climate change and human activity. Historical maps from the Second Military Survey of the nineteenth century show that the core of Klinské rašelinisko has survived the centuries largely intact, but its surrounding zones have disappeared. Past melioration, or drainage for agriculture and forestry, degraded a considerable portion of its habitats. Today, the State Nature Conservancy of the Slovak Republic is carrying out revitalisation work there, and knowing exactly where degraded and healthy habitat remains is crucial for judging whether those efforts are succeeding.

To get that information, the researchers turned to an uncrewed aerial vehicle. In August 2024, a DJI Phantom 4 Pro V2.0 drone flew at roughly 108 metres above the 0.418 square kilometre site, capturing 437 aerial images with a ground sampling distance of 2.74 centimetres per pixel. The imagery was processed in Agisoft Metashape to produce an orthomosaic and a digital surface model at five-centimetre resolution, georeferenced with five ground control points to a total error of just 2.40 centimetres. August was chosen deliberately: during the late-summer vegetation period, wetland plants reach peak biomass and their most distinctive phenophases, giving the clearest possible spectral separation between plant communities.

But colour alone was not enough. The team added a fourth data channel: vegetation height, calculated as the difference between the drone-derived digital surface model and a digital terrain model drawn from a nationwide airborne LiDAR survey conducted between 2017 and 2019. Height matters enormously in peatlands, where mosaics of forest, shrub and herb layers, along with pools, hummocks and hollows, encode the moisture gradients that define each habitat. The researchers estimate the computed vegetation height is accurate to within 10 to 15 centimetres — good enough to distinguish a mossy fen surface from a shrub thicket, and a crucial advantage over satellite products such as Sentinel-2, whose 10-metre resolution blurs away the fine-scale patches that dominate these small Slovak mires.

With the data assembled, the team applied an algorithm called the Natural Numerical Network, or NatNet, developed at the Slovak University of Technology in Bratislava and implemented in the NaturaSat software platform. Unlike the black-box reputation of conventional deep learning, NatNet is mathematically interpretable: the influence of representative samples, the graph structure and the classification parameters can all be directly inspected and understood. That transparency is not a luxury in conservation work, where managers must justify decisions about protected habitats. NatNet also retrains efficiently when capture conditions change, preserving representative habitat samples and classification logic across campaigns.

Training data came from the ground. Botanists walked the peatland, identified homogeneous areas of four Natura 2000 habitat types, documented their boundaries with GPS, and recorded species composition on Tansley’s cover scale. The habitat types themselves are defined by subtle floristic and structural criteria that coarser land-cover classes cannot capture: 91D0 Bog woodlands dominated by species such as downy birch and Norway spruce; 7120 Degraded raised bogs still capable of natural regeneration; 7230 Alkaline fens, the base-rich small-sedge and brown-moss wetlands that are the main restoration target at the site; and KRO06 Mire willow scrub, which forms mosaics at the peatland’s margins. Field polygons were refined with the semi-automatic segmentation tools of NaturaSat, yielding fourteen training polygons and seven independent validation polygons.

The researchers ran two experiments testing how the size of the training window shapes the result. In the first, NatNet learned from 150 representative squares of 11 by 11 pixels — 30 per habitat plus a background cluster of non-peatland meadow — and reached 94.6 percent validation accuracy across 167 independent validation squares. Bog woodlands and mire willow scrub were classified nearly perfectly, with F1-scores of 0.99 and 0.98. Most confusion occurred between the two spectrally similar wetland habitats, 7120 and 7230, which share comparable vegetation height, canopy density and surface texture. In the second experiment, larger 21 by 21 pixel squares — 63 in total — gave the model a broader spatial context. Training accuracy hit 100 percent, validation accuracy was 94.0 percent, and the 7120-7230 confusion dropped markedly, suggesting the wider window better captures the characteristic structure of alkaline fens, though it occasionally overestimated shrub occurrence.

The most striking result came from the relevancy maps — grayscale images in which brighter pixels indicate stronger resemblance to a target habitat. In the map for 7230 alkaline fens, bright areas appeared not only in the field-validated polygons but also inside the degraded raised bog interior, marking small fragments of high-quality fen vegetation that may represent early-stage habitat recovery. Several previously unlabelled violet validation areas were distinctly flagged as white on the map and were subsequently confirmed by botanical field surveys to contain 7230 habitat. From a restoration perspective, this is the study’s headline finding: the algorithm can spotlight candidate zones where revitalisation may already be working, precisely the information managers need to direct field verification and long-term monitoring. The 11 by 11 pixel model caught these fine-scale patches that the larger-window model, better suited to mapping broad, continuous habitats, deliberately smoothed over.

The performance figures compare favourably with the wider literature. Studies classifying broad wetland ecosystem types typically report overall accuracies of 79 to 94 percent, while plant-community-level classifications generally fall between 77 and 90 percent. A multi-sensor airborne study combining LiDAR, hyperspectral and thermal data across 22 peatland classes reached 79 percent; species-level drone classification of peatland vegetation has managed only 69 percent in the best case. The 94 percent achieved here with a simple RGB camera and a height layer, at just four habitat classes and a single site, underscores how much a well-chosen interpretable algorithm and high-resolution imagery can accomplish — though the authors caution that their validation took place within the same peatland, a less demanding test than the multi-site transferability evaluations other studies attempt.

Limitations remain. The study rests on a single drone acquisition during one phenological window, and seasonal variation in vegetation, hydrology and illumination could affect reproducibility, meaning expert botanists must time future flights. Both training and validation data came from the same locality, so the method still needs testing across additional peatlands, seasons and acquisition conditions. Yet a comparable drone-NatNet workflow has already succeeded at the Čiližská Radvaň wetland, hinting at transferability, and the model can be retrained on new representative samples without full re-acquisition campaigns. The authors see the framework as a foundation for long-term peatland monitoring in Central Europe, potentially fused with satellite imagery for scalability or drone LiDAR for richer structural information. As the EU Biodiversity Strategy for 2030 and the Nature Restoration Law demand concrete, verifiable evidence that ecosystems are genuinely recovering, a small drone circling above a Slovak bog may be quietly delivering exactly the proof that policymakers need.

Subject of Research: Drone-based machine learning classification of Natura 2000 peatland habitats in the Klinské rašelinisko peatland, Slovakia

Article Title: Drone-based classification of peatland habitats in the Klinské rašelinisko peatland (Slovakia) using Natural Numerical Networks

Article References: Ožvat, A. A., Šibíková, M., Šibík, J., Papčo, J., & Mikula, K. (2026). Drone-based classification of peatland habitats in the Klinské rašelinisko peatland (Slovakia) using Natural Numerical Networks. Discover Ecology, 2(1), Article 26. https://doi.org/10.1007/s44396-026-00044-x

Image Credits: AI Generated

DOI: 10.1007/s44396-026-00044-x

Keywords: peatlands, drones, remote sensing, habitat classification, Natural Numerical Network, Natura 2000, alkaline fens, bog restoration, machine learning, vegetation height, Slovakia, conservation

The research team, led by Anastasia Christopoulou of the University of the Aegean together with colleagues from the National and Kapodistrian University of Athens, Aristotle University of Thessaloniki, and the Academy of Athens, focused on the two most significant Macedonian pine areas in Greece: Mount Voras on the border with North Macedonia and the Chaidou forest in the Rhodope mountains near the Bulgarian frontier. Both sites lie within the Natura 2000 network of protected areas, yet while Mount Voras hosts a formally mapped 76 hectares of the protected habitat type 95A0, the High oro-Mediterranean pine forests, the presence of Macedonian pine in Chaidou had never even been mapped in the site’s official documentation. Reaching these stands required permits from the Ministry of Environment and Energy and close coordination with local forest services, whose staff guided researchers through terrain where brown bears are common and previously unmapped pine groves hide among the ridges.

The methodological approach combined classical dendrochronology with modern forest inventory and vegetation science. The team established five sampling plots of 0.1 hectares each, measuring the diameter at breast height of every tree exceeding 8 centimeters, and using laser hypsometry to capture tree heights. From 29 individual Macedonian pines, they extracted 34 increment cores, preparing each sample in the laboratory and measuring ring widths to the hundredth of a millimeter using a LINTAB measuring table and TSAP-Win software. Cross-dating relied on both visual comparison and statistical indices, including the Gleichläufigkeit coefficient and the Baillie-Pilcher and Hollstein t-values, ensuring that each ring could be assigned to its exact calendar year. Where cores missed the pith, the researchers applied standard growth-rate curves to estimate the missing rings to the tree’s center.

The headline finding is one of youth. The oldest tree sampled on Mount Voras reached just 104 years, with an estimated maximum of 112 years once missing rings were accounted for. At Chaidou, the maximum estimated age was a mere 72 years. This stands in stark contrast to Macedonian pine populations elsewhere in the Balkans, particularly in Bulgaria’s Pirin Mountains, where the species can live up to 600 years, and to its compatriot the Bosnian pine, Pinus heldreichii, which has produced multi-century and even millennia-old trees in the Greek Pindos range. The Greek Macedonian pine forests, it turns out, are essentially new arrivals on the landscape, dominated by trees established within living memory.

What caused this demographic reset remains an open question. The authors consider several plausible explanations, including past exploitation for timber, fires set by shepherds to improve grazing land, and broader land-use change. Historical records do not specifically document heavy use of Macedonian pine in the study area, although the species has been valued for carpentry and furniture elsewhere. Intriguingly, the young and expanding stands observed on Mount Voras, where saplings are colonizing former grasslands, point to land abandonment and the reduction of livestock grazing in recent decades as drivers of natural afforestation. Warming winter temperatures may further favor the species’ expansion toward the treeline, a pattern already documented in Bulgaria. Whatever the precise cause, the absence of very old individuals suggests these forests are recovering cohorts rather than ancient survivors, though the researchers caution that old trees could still persist in inaccessible ravines and cliffs beyond the reach of systematic survey.

Despite their youth, the trees are recording climate history with remarkable fidelity. The two independent chronologies, one from each mountain, correlated strongly with each other despite their limited overlap, and marker year analysis identified 36 years of exceptionally high or low growth. Four marker years were common to both sites, and each corresponds to documented climatic extremes in Greece. The year 1970, the sole shared positive marker year, aligns with the cooling trend that gripped Greece from the early 1960s to the mid-1970s. The negative marker years tell a darker story: 1987 marks the historic heatwave when July temperatures exceeded 40 degrees Celsius for nine consecutive days and reached 44 degrees in some regions; 2000 brought a parched August and sustained summer heat; and 2020 delivered another early-summer heatwave. Density fluctuations detected in the Chaidou samples, present in more than 60 percent of cores and consistently dated to 2006 and 2007, correspond to two years of extreme temperatures and unusually mild winters in Greece.

The anatomical analysis yielded an unexpected bonus. Four trees from Chaidou displayed indented growth rings, known as hazelwood, a feature caused when external injuries to the cambium, whether from fire, rockfall, herbivores, or human activity, deform the wood structure. While experts have anecdotally noted hazelwood in Macedonian pine, formal documentation in the scientific literature has been sparse. Its prevalence in the younger portions of trees aged 59 to 63 years suggests that even recently established forests are sensitive to episodic stressors capable of leaving permanent anatomical imprints, underscoring the value of dendrochronology for reconstructing disturbance histories that leave no other trace.

Structurally, the two forests could hardly be more different. Mount Voras supports a heterogeneous, uneven-aged stand shaped by prolonged regeneration and episodic disturbance, with lower densities and smaller trees. Chaidou, by contrast, shows a uniform age structure indicative of synchronous establishment, combined with higher stand density and intense competition. These differences produced a striking divergence in the relationship between size and age: on Mount Voras, diameter and height correlated moderately with tree age, but at Chaidou the relationships were weak and non-significant, with diameter and height even trending in opposite directions. The practical lesson is that the largest trees are not necessarily the oldest, a rule the data confirm emphatically. Fitted diameter distributions reinforced the contrast, with a lognormal distribution best describing Mount Voras and a gamma distribution fitting Chaidou, both revealing positively skewed structures dominated by small and intermediate trees.

The floristic survey recorded 55 plant taxa across the two sites, with 42 on Mount Voras and 22 at Chaidou, including understory species such as Juniperus communis, Vaccinium myrtillus, and Fragaria vesca. Only Pinus peuce and Calamagrostis arundinacea were typical species of the habitat type found at both sites. Worryingly, despite minimal human pressure and evidence of active expansion, both forests were assessed as having an Unfavourable-Inadequate conservation status, mirroring the national assessment for habitat type 95A0. The culprit is inadequate regeneration of the dominant pine at certain sites, compounded by the species’ limited and fragmented distribution in Greece. Remoteness, the authors warn, is a double-edged sword: it shields the forests from everyday human pressures but hampers active protection against large-scale disturbances, particularly wildfires, which are increasingly affecting high-altitude conifer forests in Greece and are expected to intensify under climate change.

For a species classified as Near Threatened on the IUCN Red List, and for a habitat type whose Greek conservation status is formally assessed as Inadequate-Bad, these findings carry real weight. The young Macedonian pine forests of Greece may lack the romantic gravitas of millennia-old Bosnian pines, but they represent unique ecosystems at the southernmost edge of the species’ range, with demonstrated sensitivity to regional climate variability and clear signs of dynamic expansion. The researchers argue that long-term monitoring of stand dynamics, regeneration, and disturbance responses is essential for adaptive management within the Natura 2000 sites that host them. As climate and land use continue to reshape mountain landscapes across the Balkans, these slender, half-century-old pines may prove to be not remnants of a lost past but pioneers of an uncertain future, their rings faithfully inscribing whatever comes next.

Subject of Research: Ecology, age structure, and conservation of Pinus peuce forests in Greece assessed through tree-ring analysis and plant diversity surveys

Article Title: Revealing the ecology of Pinus peuce forests in Greece through tree rings forest dynamics and plant diversity

Article References: Christopoulou, A., Lasut-Zmudzka, D., Zikos, A., Chrysafis, I., Kapsomenakis, I., Paidi, C., Ntagkounakis, G., Zevgolis, Y. G., Zerefos, C., & Arianoutsou, M. (2026). Revealing the ecology of Pinus peuce forests in Greece through tree rings forest dynamics and plant diversity. Discover Conservation, 3(1), Article 31. https://doi.org/10.1007/s44353-026-00095-x

Image Credits: AI Generated

DOI: 10.1007/s44353-026-00095-x

Keywords: Macedonian pine, Pinus peuce, dendrochronology, tree rings, forest ecology, Natura 2000, habitat 95A0, Balkan endemic species, climate change, forest conservation, plant diversity, Greece