The shells of tortoises function as living archives of the environments they inhabit. Unlike mammals or birds, chelonians carry their skeleton externally, with the carapace and plastron formed from fused dermal bone covered by keratinous scutes. This armor is remarkably durable, but it is not indestructible: scratches, bite marks, broken marginal plates, and charred, peeling surfaces all persist for years after the injury occurs. Because shell damage accumulates and rarely heals to an unblemished state, the proportion of individuals bearing intact shells in a population can serve as an integrated, long-term indicator of the physical hazards an animal has encountered over its lifetime. This is precisely what makes the Serbian dataset so revealing — each damaged shell represents an event that the animal survived, and the rising frequency of such events points to escalating pressures on the population as a whole.

The monitoring protocol underlying these findings was deliberately conservative. Each tortoise encountered during the twice-yearly surveys was individually marked by notching the marginal plates of the carapace, a standard technique in chelonian field studies that allows researchers to distinguish new individuals from previously recorded ones. Only the first record of each tortoise in a given year entered the analysis, which prevents frequently encountered animals from being counted multiple times and thereby skewing annual estimates of shell condition. The high and stable proportion of recaptures — averaging roughly three-quarters of each year’s sample — indicates that the population was sampled thoroughly and consistently, lending confidence to the observed temporal trends rather than reflecting mere artifacts of shifting survey effort.

The categorization scheme used to score shell condition deserves attention because it distinguishes injuries of different severity and, implicitly, different origins. Minor damage, defined as affecting less than thirty percent of the shell surface with no broken plates and few bite marks, is consistent with the everyday wear of a tortoise navigating dense vegetation, scraping under rocks, or surviving brief encounters with predators. Major damage, involving extensive scratching or broken peripheral plate sections, and severe damage, including stubs and completely broken plates, suggest more violent events such as vehicle strikes, machinery contact, prolonged predator attacks, or intense fire exposure. Separating wildfire effects as an independent variable — identified by charring and peeling of the dermal plates, a signature described in earlier Mediterranean studies — allowed the researchers to disentangle fire-related injuries from the broader spectrum of shell trauma.

Wildfire deserves particular emphasis in this context. Fire is a well-documented threat to tortoise populations across the Mediterranean basin, where it can kill animals outright or leave survivors with burned, peeling shells that compromise thermoregulation, water balance, and protection from future injury. The eastern Hermann’s tortoise, with its range extending across the Balkan Peninsula, has repeatedly been reported as vulnerable to fire events in Greece, Montenegro, Bulgaria, and Romania. What distinguishes the Serbian study is the temporal resolution: rather than documenting a single catastrophic burn, the researchers recorded the annual proportion of tortoises with identifiable fire marks, which fluctuated between less than one percent and nearly eighty-two percent of the annual sample, and averaged almost fifty-four percent across the final three years of monitoring. Five wildfire events were documented within the Nature Park between 2016 and 2025, most of them burning areas exceeding a tenth of a hectare. This pattern suggests that fire has shifted from an occasional disturbance to a recurrent and intensifying pressure on the local population.

The mechanisms linking fire to shell damage are worth unpacking. Tortoises caught in the path of a fast-moving grassland or understory fire may survive if they retreat into crevices or remain within their shells, but the heat can scorch the keratinous scutes and underlying dermal plates. Survivors carry charred, peeling surfaces that may take years to be replaced through normal growth, meaning that a single fire season can leave a lasting mark on the population’s shell-condition profile for many subsequent years. When fires recur at short intervals, the cumulative proportion of fire-marked individuals rises steeply, even if each individual event is modest in extent. This ratchet-like accumulation helps explain why the fire-related figures climbed so dramatically in the last three years of the study even though only five fire events were recorded across the entire decade and a half.

Anthropogenic pressures beyond fire also feature prominently in the researchers’ interpretation. The intensification of regional and local transport, together with a general increase in human presence in and around the study area, plausibly contributes to the rising frequency of shell injuries. Roads and vehicle traffic pose direct collision risks to slow-moving reptiles, while agricultural machinery operating in vineyards, orchards, and gardens — land uses that are interspersed throughout the tortoise habitat in the gorge — can inflict severe shell trauma on animals sheltering in vegetation or crossing open ground. Human presence also brings indirect effects: increased foot traffic can disturb nesting females, alter predator-prey dynamics by subsidizing generalist predators, and raise the likelihood of intentional fires used for pasture management escaping into surrounding woodland. The study area’s patchwork of deciduous forest, grassland, vineyards, and gardens makes it a mosaic where natural and human-modulated hazards overlap.

Natural pressures have not been ruled out either. The authors note that increasing predator population sizes may contribute to the observed decline in intact shells. Tortoises face predation pressure throughout their lives — eggs and hatchlings are vulnerable to mammals and birds, while adults, though armored, can be attacked by carnivores capable of biting or breaking shell margins. Bite marks on the carapace are a telltale signature of such encounters, and the scoring scheme explicitly counted them. If predator communities in the region have grown in abundance or shifted in composition, the frequency of non-lethal attacks on adult tortoises could rise accordingly, adding to the burden of injuries recorded each year.

It is notable that mortality within the marked population remained remarkably low over the study period: fewer than one percent of the 746 marked tortoises were found dead, with causes unknown in every case. This low recorded mortality, combined with the high recapture rate, suggests that the population itself persisted throughout the monitoring window even as shell condition deteriorated sharply. The injuries being tallied are, by definition, non-lethal — the animals survived the events that marked them. Yet this apparent resilience should not be mistaken for population health. A rising injury burden can carry sublethal costs, including reduced mobility, impaired thermoregulation, increased susceptibility to infection at wound sites, and diminished reproductive output, any of which could erode population viability over time even without an immediate spike in deaths.

The demographic structure of the population adds another layer of interpretation. In long-lived species such as Hermann’s tortoises, which can survive for many decades, populations with high nest predation and low juvenile recruitment can become dominated by older age classes. If the adults observed in the Serbian study hatched decades ago, the current cohort may reflect conditions that prevailed long before the monitoring began, while the injuries they now bear reflect recent environmental deterioration. This lag between cause and observable effect is a hallmark of long-lived vertebrate ecology and underscores why sustained monitoring over many years — as undertaken here — is essential for detecting genuine population trends rather than short-term fluctuations.

Statistically, the study relied on nonparametric methods appropriate for count and categorical data that deviated from normal distributions. The strong negative rank correlation between year and the proportion of intact shells, together with significant variation in annual sample sizes, indicates that the decline in shell condition was not a product of chance variation. The stability of the recapture proportion across years, in contrast, suggests that detectability remained roughly constant, which strengthens the inference that the trend in shell condition reflects real change in the population rather than shifting survey conditions.

Placed in a broader context, the findings from Sićevo Gorge resonate with concerns expressed across the Mediterranean and Balkan regions, where Hermann’s tortoise populations face habitat loss, fragmentation, illegal collection, and increasingly frequent fire regimes. The species is already considered imperiled across parts of its range, and the eastern subspecies, while more widely distributed than its western counterpart, is not immune to the pressures reshaping Mediterranean and sub-Mediterranean landscapes. Shell condition offers a practical, non-invasive metric that can be incorporated into existing monitoring programs without specialized equipment, making it a valuable early-warning signal for conservation practitioners.

The study’s design — repeated annual surveys across a fixed area, individual marking, standardized injury scoring, and separation of fire-related damage — provides a template that could be replicated elsewhere. Because shell injuries persist and accumulate, populations in other regions experiencing rising fire frequency, expanding transport networks, or changing predator dynamics could be assessed with comparable methods, allowing researchers to identify emerging threats before they translate into measurable population declines. The Serbian data, spanning fifteen years, demonstrate how much information such a sustained effort can yield, and they make a compelling case for continued and expanded monitoring of this and other tortoise populations facing rapidly changing environments.

Subject of Research: Long-term monitoring of non-lethal shell injuries in an eastern Hermann's tortoise population

Article Title: Long-term variation in frequency of non-lethal shell injuries in a population of eastern Hermann’s tortoise, Testudo hermanni boettgeri

Article References: Stojadinović, D., Nikolić, M., Mijatović, J., Ćorović, J., Jovanović, B., Ćosić, N., & Crnobrnja-Isailović, J. (2026). Long-term variation in frequency of non-lethal shell injuries in a population of eastern Hermann’s tortoise, Testudo hermanni boettgeri. Herpetozoa, 39, 303-311. https://doi.org/10.3897/herpetozoa.39.e191964

Image Credits: AI Generated

DOI: 10.3897/herpetozoa.39.e191964

Keywords: Hermann's tortoise, shell injuries, wildfires, Serbia, conservation, herpetology, Testudo hermanni boettgeri, anthropogenic pressure, predation, long-term monitoring, Sićevo Gorge, chelonians