The research, led by Gasto Lyakurwa of the College of African Wildlife Management in Mweka, Tanzania, and Pennsylvania State University, together with Edwin Sabuhoro of Pennsylvania State University, applied an expectancy-value model of attitude formation to survey data collected from households adjacent to the park between June and August 2023. The expectancy-value framework, rooted in classic social psychology, holds that a person’s attitude toward an object is a product of their beliefs about that object’s attributes and the value they place on those attributes. In this context, the model suggests that attitudes toward wildlife conservation emerge from the interplay of demographic characteristics, socio-economic conditions, perceptions of benefits, and lived experiences of human-wildlife conflict.

To capture these dynamics, the team randomly selected respondents from village register books across eight villages and conducted face-to-face interviews in Kiswahili, administered by trained research assistants fluent in both Kiswahili and English. Out of 267 respondents initially interviewed, 212 provided complete responses that could be analyzed. The surveyed population was dominated by farmers, who outnumbered pastoralists roughly six to one, and included people from diverse ethnic backgrounds, primarily Pare and Maasai. Slightly more female-headed households participated than male-headed ones. The average age of household heads was 49 years, and average annual household income was approximately 1,115 US dollars. Notably, 75.9 percent of respondents had completed only primary education, while 17.5 percent had no formal education at all.

The statistical heart of the study was a multiple logistic regression model that treated conservation attitude as a binary outcome, coding positive attitudes as one and negative attitudes as zero. The model tested the influence of gender, education level, household income, household size, land ownership, livestock numbers, health insurance status, food sufficiency, distance to health centers, and perception of conservation benefits. The researchers validated their approach with a Box-Tidwell test, confirming that the assumption of linearity in the logit was satisfied for all continuous predictors. The final model was statistically significant, with a chi-square value of 33.776 and a p-value of 0.001, and it correctly classified 69.3 percent of cases.

Three predictors stood out as statistically significant. First, household heads with formal education, defined as completion of primary school or above, were associated with a 539 percent increase in the likelihood of reporting a positive conservation attitude compared with those lacking formal education. Second, respondents who perceived benefits from wildlife were 300 percent more likely to express positive conservation attitudes than those who did not. Third, and perhaps most counterintuitively, households reporting higher food sufficiency were actually 41 percent less likely to hold positive attitudes toward conservation, a finding that inverts the intuitive expectation that food security and conservation support go hand in hand.

The food security result deserves particular attention because it reveals a tension at the heart of conservation policy. The authors suggest that people who work hard to produce enough food may be more anxious about the threat that wildlife poses to their crops and livestock than those with less at stake. In communities where elephants raid maize fields and carnivores take cattle, the very act of achieving food security can heighten vulnerability to wildlife damage, and therefore breed resentment toward the animals and the park that protects them. This interpretation aligns with prior research from Zimbabwe’s Save Valley Conservancy, where residents felt unsafe when wildlife affected their livelihoods, and with studies showing that crop damage and livestock depredation significantly undermine household food security in Tanzanian communities near protected areas.

Beyond the regression model, the study employed chi-square tests to examine the relationship between human-wildlife conflict and conservation attitudes. The results showed a significant association between the perceived trajectory of conflict, whether increasing, decreasing, or stable, and conservation attitude, with a p-value of 0.009. Respondents who felt that conflict was decreasing were far more likely to report positive attitudes. However, there was no significant association between simply being aware of human-wildlife conflict and holding positive attitudes, nor between the raw frequency of conflict encounters and attitude. This distinction matters because it suggests that it is not the mere occurrence of conflict but rather the perceived direction of change that shapes how people feel about conservation. When residents believe the situation is improving, their support for conservation strengthens accordingly.

The overall picture that emerges from the survey is sobering. Respondents were significantly more likely to report negative conservation attitudes than positive ones, with 136 expressing negative views and only 76 expressing positive views, a difference that was statistically significant at p less than 0.001. This skew toward negativity likely reflects the cumulative burden of living alongside wildlife without commensurate benefits. Compared to famous destinations like Serengeti and Kilimanjaro National Parks, Mkomazi has not yet attracted enough tourism to generate meaningful economic returns for surrounding communities, leaving residents to bear the costs of conservation without enjoying its rewards.

The findings carry important implications for conservation policy and practice. The authors argue that to foster positive wildlife conservation attitudes, development policies should be designed to provide knowledge through education, ensure food security through effective human-wildlife conflict mitigation, and enhance the opportunities and ecosystem services that conservation activities can deliver, including ecotourism potential and employment in the wildlife sector. The education finding is particularly consequential, echoing research from Kenya and from Tanzania’s own Ngorongoro region showing that formal schooling and school-based environmental clubs can positively shape young people’s perceptions of wildlife and conservation.

At the same time, the authors caution that the relationship between education and conservation behavior is not always straightforward, and that attitudes are highly context-specific and temporally dynamic. Conservation benefits alone do not guarantee positive attitudes, as prior work near the Selous Game Reserve demonstrated when communities accepted material benefits but still held negative perceptions of the reserve itself. What this study ultimately makes clear is that winning local support for conservation requires more than handing out benefits or building fences. It requires understanding the specific social, economic, and psychological landscape of each community, and designing interventions that address the real drivers of attitude formation, from classroom education to conflict mitigation to equitable benefit sharing. As human populations grow and pressure on protected areas intensifies, this granular, place-based understanding may prove to be the most valuable conservation tool of all.

Subject of Research: Demographic and socio-economic drivers of wildlife conservation attitudes among communities adjacent to Mkomazi National Park in Tanzania

Article Title: Demographic and socio-economic influences on self-reported wildlife conservation attitude: a case study of communities adjacent to Mkomazi National Park in Tanzania

Article References: Lyakurwa, G., & Sabuhoro, E. (2026). Demographic and socio-economic influences on self-reported wildlife conservation attitude: a case study of communities adjacent to Mkomazi National Park in Tanzania. Frontiers in Sustainable Resource Management, 5, Article 1891779. https://doi.org/10.3389/fsrma.2026.1891779

Image Credits: AI Generated

DOI: 10.3389/fsrma.2026.1891779

Keywords: wildlife conservation, human-wildlife conflict, Mkomazi National Park, Tanzania, expectancy-value model, food security, conservation attitudes, protected areas, community-based conservation, ecotourism, logistic regression, environmental education

The research, conducted from a recreation ecology perspective, analyzed 684 documents retrieved from the Scopus database using carefully constructed search equations. The findings paint a picture of a rapidly maturing scientific field—one that has grown at an annual rate of 15.25% since 2019—as researchers worldwide grapple with the paradox of tourism that damages the natural areas it depends upon.

Recreation ecology, the scientific discipline underpinning the analysis, focuses on understanding how recreational activities affect ecosystem structure and function. The reviewed literature consistently documents a range of environmental impacts associated with ecotourism, including wildlife disturbance, soil degradation and erosion, carbon dioxide emissions, waste generation, and mounting pressure on natural resources. These effects are particularly pronounced along trails, campsites, picnic areas, and wildlife observation sites where visitor concentrations are highest.

The geographic distribution of research output reveals a striking imbalance. The United States leads with 119 publications, followed by Australia with 78, China with 78, the United Kingdom with 62, and Canada with 50. This concentration in developed Anglophone nations raises concerns about what researchers call epistemic asymmetry—where developing economies are relegated to the role of subjects rather than generators of knowledge. The authors warn that management models designed for temperate zones of the Global North may be poorly suited to the socioeconomic vulnerabilities and megadiverse ecosystems of the Global South.

Institutional leadership mirrors this geographic pattern. Griffith University in Australia tops the list with 16 publications, followed by Murdoch University with 14, the Chinese Academy of Sciences with 13, the University of Johannesburg with 12, and Universidade do Algarve with 12. The co-authorship network centers on researchers including Newsome D., Dearden P., and Cong R., suggesting that a relatively small group of influential scientists shapes the direction of the global research agenda.

The thematic structure of the field reveals an interdisciplinary evolution. Environmental Science accounts for 26.3% of publications, Social Sciences for 19.7%, and Agricultural and Biological Sciences for 15.7%. Keyword co-occurrence analysis identified 263 terms organized into six thematic clusters, with ecotourism serving as the central node linked through 1,099 structural connections. The dominance of biodiversity, sustainability, recreation, and environmental conservation as research themes underscores the field’s shift from purely biophysical diagnosis toward integrated socio-environmental frameworks.

Despite this growing sophistication, the analysis identifies a critical methodological gap. The relationship between social and ecological dimensions has traditionally been neglected in recreation ecology, with few interdisciplinary studies effectively connecting both aspects. The authors argue that when ecotourism planning lacks interdisciplinary preventive design, ecological problems—such as water pollution or inadequate waste management—can cascade into social crises, governance tensions, and public health issues for local communities.

The reviewed literature also highlights tensions between visitor experience and climate sustainability. Research on tourist itinerary design demonstrates that efforts to maximize economic benefit frequently conflict with the goal of minimizing carbon dioxide emissions from transportation. While optimization models and fuzzy logic approaches have been developed to mitigate the transportation footprint, these tools reduce but do not eliminate the underlying ecological pressures associated with visitor mobility.

Acoustic and landscape disturbances represent another underappreciated dimension of ecotourism’s impact. Studies in highly sensitive national parks, including those in Alaska and Hawaii, reveal that tourist transportation patterns—both ground routes and low-altitude overflights—drastically alter natural soundscapes. These acoustic intrusions disrupt ecological integrity and destabilize wildlife behavior patterns, effects that persist long after visitors depart.

The study concludes that ecotourism, when it exceeds the actual ecological limits of protected sites, intensifies adverse impacts across physical, biological, and environmental dimensions. The authors call for integrating rigorous economic valuation tools and carrying capacity indicators into decision-making, while urging future research to develop predictive simulation models that effectively connect numerical indicators of recreation ecology with social well-being and participatory governance of local and Indigenous communities. As nature-based tourism continues its global expansion, the findings serve as a timely reminder that sustainability claims must be backed by science, not just marketing.

Beyond the headline findings, the methodological choices of the study itself offer useful context for interpreting its conclusions. By relying exclusively on Scopus and restricting the corpus to English-language publications, the analysis captures a substantial share of the indexed literature but inevitably underrepresents research produced in non-Anglophone journals and grey literature, particularly studies conducted by practitioners and government agencies in biodiversity-rich developing countries. The authors acknowledge this constraint implicitly through their discussion of epistemic asymmetry, noting that the dominance of the United States, Australia, China, the United Kingdom, and Canada in publication output does not necessarily reflect where the most severe visitor impacts occur, but rather where the capacity to publish in indexed venues is concentrated.

The temporal filter applied in the study is also instructive. The initial search returned 839 documents, but restricting the window to the ten full years from 2016 to 2025 reduced the final dataset to 684 publications, with 2026 excluded because indexing for that year was incomplete. This decision reflects a common challenge in bibliometric research: partial years can distort growth-rate calculations and citation statistics, since recently published papers have had little time to accumulate citations. The reported annual growth rate of 15.25% is therefore a conservative estimate of the field’s expansion, and the acceleration observed from 2019 onward coincides with rising global attention to sustainable tourism in the wake of overtourism debates and, subsequently, the post-pandemic surge in demand for outdoor recreation in natural settings.

The composition of the corpus provides additional insight into how knowledge in this field is produced and disseminated. Research articles accounted for 77.8% of the 684 documents, with reviews contributing 8.0%, book chapters 6.9%, and books 6.1%, while conference papers and editorial notes made up less than 1.5% combined. This profile suggests that the field is primarily driven by primary empirical research rather than synthesis, which the authors imply is both a strength and a limitation: a wealth of case studies exists, but comparatively few efforts have consolidated them into generalizable frameworks or predictive models that managers can apply across different protected area contexts.

The disciplinary spread of the literature also deserves attention. Although Environmental Science leads with 26.3% of publications, the substantial shares of Social Sciences at 19.7% and Business and Management at 15.5% indicate that research on ecotourism’s negative effects has never been purely ecological. Agricultural and Biological Sciences contributed 15.7% of the corpus, reflecting the discipline’s roots in studying vegetation trampling, soil compaction, and wildlife responses along trails and at campsites. The presence of economics and management-oriented publications underscores the tension the study identifies between visitor experience, revenue generation, and ecological integrity, a tension that optimization models for tourist itineraries have only partially resolved.

Recreation ecology itself has a longer intellectual history than the ten-year window of this analysis suggests. Classic studies cited in the article, including work on trail and campsite impacts dating back decades, established foundational principles that recent research has refined with new tools such as spatial modeling, remote sensing, and visitor-use monitoring. The bibliometric evidence that keyword co-occurrence clusters around ecotourism, protected areas, biodiversity, sustainability, recreation, and environmental conservation indicates that the field has broadened from its original biophysical focus toward socio-ecological integration, though the authors caution that genuinely interdisciplinary studies linking social and ecological dimensions remain scarce.

The findings carry practical implications for protected area managers. Because impacts concentrate at predictable hotspots—trails, campsites, picnic areas, and wildlife viewing sites—targeted interventions such as hardened surfaces, spatial zoning, visitor dispersal, and seasonal closures can address much of the documented damage without necessarily reducing overall visitation. The literature reviewed in the study also emphasizes that visitor use can facilitate the introduction of invasive species and alter wildlife habitat use and population densities, effects that require ongoing monitoring rather than one-time assessments. Adaptive management, in which monitoring data continuously inform visitor-use decisions, emerges as a recurring recommendation across the reviewed publications.

Finally, the study’s call for predictive simulation models connecting recreation ecology indicators with social well-being and participatory governance points toward the field’s likely trajectory. As climate change alters visitation patterns, species distributions, and ecosystem resilience in protected areas, the interactions between tourism pressure and environmental change will become increasingly complex. The authors argue that carrying capacity indicators and rigorous economic valuation tools must be embedded in decision-making processes, and that local and Indigenous communities should be treated as partners in knowledge production rather than merely as stakeholders affected by research conducted elsewhere. Whether the next decade of research can overcome the geographic and disciplinary imbalances documented in this analysis will largely determine whether ecotourism can fulfill its conservation promise in the world’s most biodiverse regions.

Subject of Research: Environmental impacts of ecotourism in protected areas from a recreation ecology perspective

Article Title: Negative effects of ecotourism in protected areas: a bibliometric analysis from a recreation ecology perspective

Article References: Huepa Briñez, J. A., Florez Yepes, G. Y., & Ospino, J. O. (2026). Negative effects of ecotourism in protected areas: a bibliometric analysis from a recreation ecology perspective. Nature Conservation, 64, 213-228. https://doi.org/10.3897/natureconservation.64.193686

Image Credits: AI Generated

DOI: 10.3897/natureconservation.64.193686

Keywords: ecotourism, protected areas, recreation ecology, biodiversity, sustainability, environmental conservation, bibliometric analysis, soil degradation, wildlife disturbance, carbon emissions, visitor management, Global South

The research team surveyed 96 households across four kebeles, the lowest administrative units in the district, chosen in consultation with biodiversity experts for accessibility and security. Respondents were interviewed at home using a pre-tested, semi-structured questionnaire translated into the local language, Hameregna, with a native-speaking interpreter present. The survey was complemented by three focus group discussions involving community leaders, youths, women, administrators, and agricultural experts, as well as direct field observations by the researchers. Sample size was calculated with the Yamane formula at a 95 percent confidence level, and the data were analyzed with descriptive statistics, Pearson chi-squared tests, and multiple linear regression using SPSS version 25.

The demographic profile of the respondents underscores how closely tied the district’s fortunes are to traditional livelihoods. Nearly 88.5 percent of those interviewed were pastoralists or agro-pastoralists, herding cattle, goats, and sheep while cultivating sorghum and selling honey. Almost 91 percent had no formal education, and most were married and between 40 and 60 years old. Every respondent was a native of the area, many with decades of residence, giving them a long-term view of ecological change that few external observers could match.

That view is bleak. About 82.3 percent of respondents said wildlife populations in their vicinity are decreasing, and only 1.1 percent reported any increase. These perceptions differed significantly across the four sampled kebeles, with a chi-squared statistic of 44.152 on three degrees of freedom and a p-value below 0.001, suggesting that the intensity of pressure varies locally even within a single district. The remaining respondents either saw no change or were uncertain, but the overall signal of decline was unmistakable.

When asked to rank ten principal threats, respondents placed deforestation, driven by settlement expansion and charcoal production, at the top: 63.5 percent named it the primary threat. Overgrazing came second, ranked first by 30.2 percent and second by 50 percent. Agricultural land expansion and illegal hunting followed in the middle tiers, while human-wildlife conflict was ranked only seventh by most respondents, at 65.6 percent. Human-induced fire, observed directly by the researchers during field visits, also emerged as a significant pressure, with communities deliberately burning habitat to clear land for farming, settlement, and honey collection.

The ecological logic behind these rankings is well supported by broader research. Agriculture is responsible for roughly 96 percent of deforestation in Ethiopia, and long-term data from western Ethiopian agroecosystems show forest cover shrinking at an annual rate of 2.05 percent. In pastoralist regions, the problem intensifies during drought, when herders may travel more than 45 kilometers in search of fodder and water, pushing livestock deep into wildlife habitat. In Hamer District, grazing lands are not allocated seasonally and enclosures are not established for dry periods, so competition between cattle and wild herbivores for forage and water becomes acute precisely when animals are most vulnerable.

Hunting for bush meat, traditional medicine, ivory, and decorative materials adds a further layer of pressure. Respondents listed a range of species used by local communities, including buffalo, warthog, dik-dik, elephant, leopard, lesser kudu, guinea fowl, and francolins, with colobus monkey skins used for decoration and python parts for traditional medicine. Alarmingly, the local existence of giraffes and ostriches is now uncertain. The pattern mirrors national and regional trends: Ethiopia has lost an estimated 90 percent of its elephants since the 1980s, largely to ivory poaching, while the illegal trade in live cheetah cubs remains a major threat across the Horn of Africa. Notably, no respondent identified zoonotic diseases as a threat, despite earlier studies documenting bovine tuberculosis and trypanosomiasis in the area, a gap the authors attribute to distance from the Murule Controlled Hunting Area or limited local awareness.

Against this backdrop of decline, the willingness of the community to act is the study’s most hopeful finding. Fully 90.6 percent of respondents said they would participate in wildlife conservation activities, and 99 percent agreed that conservation work in the district is urgent. Nearly 97 percent said local communities themselves bear responsibility for protecting wildlife, with only a handful pointing to NGOs or broader coalitions of stakeholders. The regression models revealed that willingness was significantly predicted by kebele of residence, sex, marital status, and years of residency, with standardized coefficients of -0.102, -0.544, -0.378, and 0.324 respectively. Education and occupation, by contrast, were not significant predictors, suggesting that long attachment to the land and lived experience of ecological change matter more than formal schooling in shaping conservation attitudes.

The authors attribute this goodwill to the tangible benefits communities derive from the landscape. The Buska Forest, one of the last substantial forest patches in the district, serves as a drought refuge for livestock and supplies medicinal, cultural, and economic resources, from honey to materials used in ceremonies such as bull jumping and the Ivangadi dance. Comparable studies around Awash National Park in Ethiopia and Kruger National Park in South Africa have found similar positive attitudes where protected habitats provide dry-season grazing, water, and non-consumptive value. Retaining that goodwill, the researchers argue, depends on delivering practical alternatives: managed livestock husbandry, alternative grazing lands, and artificial water ponds that can serve both cattle and wildlife during droughts.

The study’s recommendations extend to community-based conservation, ecotourism development, and stronger law enforcement to halt illegal hunting and habitat clearing. The authors also call for additional seasonal research to document the district’s wildlife diversity and assess its ecotourism potential, noting that most biodiversity exists outside protected areas where conservation attention is thinnest. With extinction rates currently estimated at a thousand times the background rate recorded in the fossil record, the Hamer District offers a case study in both the scale of the challenge and the potential of community-led solutions. If the pastoralists who share this landscape with its last giraffes, kudu, and colobus monkeys are equipped to manage grazing and benefit from standing wildlife, the survey suggests they are ready to become its most committed guardians.

Subject of Research: Anthropogenic threats to wildlife and community conservation willingness in Hamer District, South Omo, Ethiopia

Article Title: Anthropogenic threats and challenges of wildlife conservation in Hamer District, South Omo, Ethiopia

Article References: Temesgen, Z., Biru, Y., & Gibru, A. (2026). Anthropogenic threats and challenges of wildlife conservation in Hamer District, South Omo, Ethiopia. BioRisk, 24, 69-84. https://doi.org/10.3897/biorisk.24.175449

Image Credits: AI Generated

DOI: 10.3897/biorisk.24.175449

Keywords: Ethiopia, wildlife conservation, deforestation, overgrazing, pastoralism, South Omo, bushmeat, human-wildlife conflict, community conservation, Buska Forest, ecotourism, habitat loss

The research, published in Frontiers in Sustainable Resource Management, was conducted by Oliver Simeon Lihuluku, Deo D. Shirima, and Agnes A. Sirima of Sokoine University of Agriculture in Morogoro. The team focused on three strategically contrasting destinations in Tanzania’s Pwani and Dar es Salaam regions: Pugu Kazimzumbwi Nature Forest Reserve, a protected coastal forest about 25 kilometers southwest of Dar es Salaam; Mbudya Island Marine Reserve, a coral reef and seagrass site within the Dar es Salaam Marine Reserve System; and the Kaole Ruins near Bagamoyo, a thirteenth-to-sixteenth-century archaeological site fringed by mangroves dominated by Avicennia marina and Rhizophora mucronata. Together, the sites span forest, marine, and cultural heritage conservation contexts, allowing the researchers to test whether the ecotourism-restoration link holds across fundamentally different ecosystem types.

The study’s architecture combined three data streams. First, the authors compiled official visitor statistics from the Tanzania Forest Services Agency and marine park records covering 2012 through 2022, cleaning and encoding eleven years of annual visitation data for statistical analysis. Second, they acquired multitemporal Landsat satellite imagery from the USGS archive, selecting scenes from Path 166/Row 065 with less than ten percent cloud cover. Raw digital numbers were converted to top-of-atmosphere reflectance using metadata-derived gain and bias coefficients, and thermal bands were translated into brightness temperature. Land use and land cover classes—vegetation, built-up area, bare land, water bodies, farmland, and rangeland—were then mapped with a Random Forest machine-learning classifier implemented through the Dzetsaka plugin in QGIS, with accuracy assessed via confusion matrices, user’s and producer’s accuracy, overall accuracy, and the Kappa coefficient. Third, the team conducted semi-structured interviews with fifteen key informants drawn from district councils, village elders, agency representatives, and tour operators, recruited through purposive sampling until data saturation was reached.

The visitation data alone tell a story of explosive growth. Over the eleven-year window, the three sites collectively recorded 353,492 visits. Kaole Ruins led with 170,971 visitors despite drawing nobody until 2016, when promotion and infrastructure finally arrived; its numbers then climbed steeply to 36,640 in 2022, the highest single-site annual total in the dataset. Mbudya Island maintained a steady marine tourism presence, totaling 130,000 visits anchored by snorkeling, diving, and beach activities. Pugu Kazimzumbwi was the late bloomer: essentially unvisited until 2017, it recorded an exponential rise after 2019, reaching 22,491 visitors in 2022—a 45 percent growth in visitation over the study period at the site level. Across all activities, worshiping at Kaole Ruins attracted the most participants, followed by hiking in the Pugu forest. Men outnumbered women overall, 179,515 to 173,977.

When the researchers correlated annual visitation with restoration indicators, the associations were remarkably strong. Pearson correlation coefficients between visitation rate and restoration index reached 0.992 at Pugu Kazimzumbwi, 0.971 at Kaole Ruins, and 0.858 at Mbudya Island, all statistically significant at the p < 0.01 level. The authors are careful to stress that these are exploratory correlations based on only eleven annual observations per site, and cannot establish causality. Still, the pattern is consistent with a plausible mechanism: entrance fees and tourism revenues are reinvested into restoration, so more visitors mean more money for conservation. Interview testimony reinforced this reading. As one key informant from Pugu Kazimzumbwi put it, “When we started the reforestation program, more visitors came to see the young trees and participate in planting activities,” while a Mbudya tour guide observed that “tourists are more willing to visit when they can see healthy coral reefs during snorkeling”—a feedback loop the researchers describe as a virtuous cycle.

The restoration portfolio documented at the sites is tangible. At Pugu Kazimzumbwi and Kaole, tree planting initiatives funded substantially by ecotourism revenues are organized by the Tanzania Forest Services Agency, the Tanzania Wildlife Management Authority, local government authorities, and WWF, with the Jane Goodall project supporting tree nurseries. Communities supply labor for nursery establishment and planting, and native species are prioritized for restoring degraded forest. At Mbudya Island, local government authorities coordinate regular beach clean-ups and coral reef restoration with tour operators and volunteers, with reef funding correlating positively with visitor growth. The interviews suggest these efforts are not incidental: tourism income appears to be a genuine financing engine for coastal restoration in the region.

Yet the satellite record complicates the optimistic narrative. Pugu Kazimzumbwi lost ten percent of its vegetation cover between 2012 and 2022, dropping from 49 to 39 percent of its area, while built-up land nearly doubled from 11 to 20 percent, bare land expanded from 25 to 35 percent, and water bodies shrank from 15 to 6 percent. Kaole Ruins saw farmland plummet from 35 to 10 percent of its area as built-up land surged from 5 to 22 percent, reflecting conversion of agricultural land to tourism-related and residential development—including new accommodation such as the Kaole homestay. Only Mbudya Island showed unambiguous ecological gain, with vegetation rising from 25 to 35 percent through mangrove recovery and bare land halving, though its open water declined from 60 to 50 percent, a shift the authors attribute to sedimentation, tidal changes, and managed restoration. Classification reliability improved over time, with Kappa coefficients rising from roughly 0.85 to above 0.91 by 2022, lending confidence to these trends.

The tension between rising restoration funding and degrading land cover points to what the authors frame as a threshold effect in newer destinations. Sites that recently opened to tourism lacked the refined management protocols, visitor impact mitigation strategies, and carrying capacity controls that mature destinations typically develop over time, so rapid growth in visitation can outpace the ability of management systems to absorb the pressure. Institutional history matters here too. Pugu Kazimzumbwi, established in 1936, suffered decades of logging and agricultural encroachment before the Tanzania Forest Services Agency assumed management in 2010 and introduced specialized patrols and community engagement through Village Forest Committees—though the authors note that benefit-sharing remains limited and community relations strained. Mbudya’s dedicated Marine Parks and Reserves Unit, with its wardens, rangers, and Beach Management Units, exemplifies the specialized institutional capacity the study links to better outcomes.

The study also surfaced an unexpected demographic pattern. At Mbudya Island, male visitation correlated strongly with restoration indicators, while female visitation did not—a disparity the researchers link to the physical demands and perceived safety risks of water-based activities such as diving and snorkeling. Field observations showed men more often engaging directly with visible marine conservation, while women tended to prefer beach-based activities less connected to restoration outcomes. At the land-based sites of Pugu Kazimzumbwi and Kaole Ruins, both genders showed similarly strong correlations, suggesting that activity type and safety perceptions, rather than restoration itself, shape who responds to ecosystem recovery.

The authors’ recommendations follow directly from the mixed evidence: strengthen site-specific management strategies, enforce carrying capacity limits before visitor growth overwhelms sensitive habitats, and ensure tourism revenues are systematically reinvested in restoration. They caution that the cross-sectional, correlational design cannot definitively prove that tourism drives restoration, and call for longitudinal and experimental research to clarify causal pathways. What the study does establish is that ecotourism in coastal Tanzania is neither a conservation panacea nor a false promise. Managed well, visitor spending can bankroll tree planting, reef rehabilitation, and habitat protection while drawing communities into stewardship. Managed poorly, the same foot traffic—and the development it attracts—can erode the very ecosystems the business model depends on. The net ecological balance, the findings suggest, is decided not by the tourists themselves but by the institutions charged with managing them.

Subject of Research: The relationship between ecotourism visitation rates and coastal ecosystem restoration outcomes at three protected sites in Tanzania

Article Title: Ecotourists' visitation rates and coastal ecosystem restoration: evidence from Tanzania

Article References: Lihuluku, O. S., Shirima, D. D., & Sirima, A. A. (2026). Ecotourists' visitation rates and coastal ecosystem restoration: evidence from Tanzania. Frontiers in Sustainable Resource Management, 5, Article 1918143. https://doi.org/10.3389/fsrma.2026.1918143

Image Credits: AI Generated

DOI: 10.3389/fsrma.2026.1918143

Keywords: ecotourism, coastal ecosystems, Tanzania, ecosystem restoration, land cover change, satellite imagery, Random Forest classification, conservation financing, mangroves, coral reefs, carrying capacity, Pugu Kazimzumbwi

Fire emerged as one of the study’s central themes. Researchers mapped vegetation fires across five seasons, from July 2019 through 2024, revealing dramatic year-to-year swings. The largest burned areas occurred in the 2021–2022 season, when a single March peak consumed 35,425 hectares of reed-dominated landscape, while the most recent season recorded the smallest burned extent. Monthly analysis showed that fires cluster sharply in March, when winter-dried reed beds act as vast stands of standing fuel. Because reeds dominate the delta’s aquatic-to-terrestrial transition zones, fire there is not a marginal disturbance but a landscape-scale force that reshapes habitat availability for birds, mammals, and plants alike. The team selected 90 observation points based on composite fire maps from the last three seasons to track how repeated burning affects bird communities, and field surveys in April 2024 across 31 points recorded 279 birds from 36 species in and around burned zones.

The comparison between burning and harvesting as management tools produced some of the most actionable results. Working across reed beds subjected to 2024 interventions, three-year-old unmanaged stands, and ten-year-old unmanaged stands, the researchers recorded 66 plant species, of which only 22 appeared in every treatment type. Indicator species analysis revealed clear signatures: water-plantain was associated with burned reeds, sedges and bindweed with harvested reeds, and bittersweet, water hemlock, and reed canary-grass with the oldest beds. The effects shifted with the seasons—species richness peaked in three-year reeds in June, while Shannon diversity was highest in freshly harvested and ten-year-old stands by September. Burning reduced light values in the understory by shading out coexisting plants, whereas harvesting increased them, opening the canopy to companions. Stem density was highest in harvested plots, but stems there were shorter and thinner. The authors conclude that no single treatment is optimal: the biodiversity payoff comes from managing reed beds as mosaics of different ages and treatments, alternating fire and harvest to maximize habitat heterogeneity, with the benefits of any single harvest lasting only a few years.

Beyond fire and reeds, the study inventoried the delta’s harvestable natural wealth with striking precision. Three legally harvestable edible mushroom species—field mushroom, wood mushroom, and meadow mushroom—occupy roughly 9,200 hectares of suitable habitat, yielding an estimated 75 tons annually. Eighty-seven medicinal herb species contribute about 22 tons of harvestable dried biomass. Beekeeping potential was assessed through 64 key nectar plants, with standardized honey yields ranging from 5 kilograms per hectare in wetland pastures to 50 kilograms per hectare in natural wetland meadows, and eight optimized beekeeping routes were proposed. Natural pastures covering about 27,500 hectares support 195 grazing-tolerant plant species at moderate quality, while the delta’s roughly 170,000 hectares of reed beds, managed through 19 concession areas, hold an annual harvestable potential of about 88,000 tons. These figures feed a database intended to guide the reserve administration in issuing permits that balance local livelihoods against ecosystem stability.

The biodiversity surveys delivered a mixed picture. Invertebrate monitoring across dozens of sites, from the delta’s western edge to the Black Sea coast at Vadu and Histria, found a strikingly species-poor fauna: the only invertebrate of Community interest reliably observed all year was the large copper butterfly, Lycaena dispar, recorded at Zebil and Saon. The researchers attribute this scarcity to extreme weather—a cold spring followed by sudden, prolonged summer heat—and caution that the picture is incomplete because surveys focused on peripheral areas. Fourteen years of monitoring have revealed persistent knowledge gaps in the distribution of protected invertebrates, complicating both the implementation of the reserve’s management plan and the Natura 2000 standard data sheets, which themselves require periodic revision.

Fish surveys told a richer story. Sampling in the Danube River, its branches, the Somova–Parcheș lake complex, and interfluvial lakes captured 61 species, including 14 of Community interest, plus the Ukrainian brook lamprey, Eudontomyzon mariae, and the cryptic gudgeon Romanogobio kessleri antipai, both absent from the official standard form. Official 2023 catch data showed freshwater species dominating at 91.5 percent of total fishery resources, led overwhelmingly by gibel carp at 40.4 percent, followed by Asian carps, breams, and common carp. Migratory fish accounted for 8.3 percent, almost entirely Pontic shad. More troubling were the trends among predators: pike, after a continuous decline from 5.7 percent of catches in 2019 to 2 percent in 2022, showed a slight recovery to 2.9 percent, but zander continued to fall, representing just 1.6 percent of catches—under 41 tons. Climate change compounds the pressure, with rising temperatures, reduced precipitation, and falling water levels desiccating delta ponds.

Herpetofauna and bird surveys filled in other pieces of the puzzle. Four of the five reptile and amphibian species of Community interest were detected in 2024, with the European pond turtle the most frequently encountered at 11 points and the steppe viper, Vipera ursinii moldavica, reduced to a single observation inside a fish farm; the Dobrogean crested newt went undetected. Among birds, the reserve’s 26 colonial waterbird species formed 102 colonies holding an estimated 61,834 to 63,915 pairs—one of the highest population levels of the past five years. Medium to high water levels favored black terns, black-headed gulls, and common terns, but in Lake Sinoie, insufficient drainage flooded colonies on Ceaplace and Prundul cu Păsări Islands, shrinking nesting habitat. Predation took its own toll: brown rats caused high chick mortality in the Musura colony of Caspian gulls, and falling water levels at Sărături Murighiol let predators reach nests. The great cormorant, an exclusively piscivorous species, declined over the past two years—most likely, the authors suggest, because persistently high water limited its access to food, making it a living bioindicator of fish resources.

Mammal monitoring across the ROSCI0065 and ROSCI0066 Natura 2000 sites recorded seven of nine species of Community interest. The Eurasian otter reigned as the most frequently encountered, present at 57 of 130 surveyed locations with 175 individuals. The European mink appeared at eight locations with nine individuals—and, notably, was observed in summer for the first time, having previously been recorded only in autumn, winter, and spring. The beaver expanded to three areas, up from one in previous years. The Dobrogean hamster, a Balkan endemic restricted to Romania and Bulgaria, clung to its known stronghold at Cape Iancina, while the European ground squirrel numbered 957 individuals concentrated in northwestern steppe habitats. At sea, 46 harbor porpoises and 46 bottlenose dolphins were recorded along the coast, but three dead porpoises found on Sulina Beach pointed to a grim pattern: interviews with reserve staff indicate that two to three porpoises drown daily in fishing nets, most often during the mackerel season. The steppe polecat and marbled polecat were not observed at all.

Invasive aquatic plants added another warning sign. Two species—Vallisneria spiralis and Elodea nuttallii—were identified across the delta’s lakes through phytocoenological surveys of 93 plots and cluster analysis of vegetation data. Vallisneria spread through eight lakes, thriving especially in Furtuna, Uzlina, and Parcheș, while Elodea appeared in four lakes, with significant coverage in Lake Uzlina ranging from 0.5 to 62.5 percent. The analysis classified lake vegetation into 12 community types, from stonewort meadows to waterlily-dominated stands, providing a baseline for tracking how these invaders reshape native submerged communities under warming conditions.

The human dimension rounded out the assessment. In Jurilovca commune, a case-study community whose population has fallen 15.2 percent since 2011, questionnaires administered to 125 residents revealed a cautiously contented but anxious population: 52 percent saw their living conditions as unchanged and expected stagnation, fewer than half considered their incomes sufficient for a decent standard of living, and 67.2 percent identified job creation for young people as the single most important need. Yet tourism is booming. The Jurilovca area recorded 4,721 tourists and 10,435 overnight stays in 2024, with visitors averaging three days—drawn by the Gura Portiței beach, a new tourist port, and the cultural traditions of the Lipovan community, earning Jurilovca fourth place among Romania’s top destinations from FIJET. The Murighiol area, easier to reach, attracted 31,765 tourists for 78,500 overnight stays, its accommodation capacity more than doubling since 2019, and was certified as Tulcea County’s first official tourist resort. Together, the two areas illustrate the delta’s emerging model: tailored ecotourism that balances visitor experience with conservation and local prosperity—provided the fires, the falling water, the invasive plants, and the porpoises in the nets are managed with the same rigor that this year’s research has now established.

Subject of Research: Integrated 2024 assessment of biodiversity, reed habitat management, resource use, invasive species, and socio-economic trends in the Danube Delta Biosphere Reserve

Article Title: Status of biodiversity, reed habitats, sustainable exploitation of natural resources, invasive species, and socio-economic implications in Danube Delta Biosphere Reserve in 2024

Article References: Lupu, G., Covaliov, S., Doroftei, M., Mierlă, M., Simionov, M., Năstase, A., Cenișencu, I., Fomici, I., Chirilă, S.-D., Naum, A., Doroșencu, A., Bolboacă, L., Marinov, M., Alexe, V., Sicrieru, F., & Ene, L. (2025). Status of biodiversity, reed habitats, sustainable exploitation of natural resources, invasive species, and socio-economic implications in Danube Delta Biosphere Reserve in 2024. Scientific Annals of the Danube Delta Institute, 30, 141-162. https://doi.org/10.3897/saddi.30.163613

Image Credits: AI Generated

DOI: 10.3897/saddi.30.163613

Keywords: Danube Delta, biosphere reserve, reed beds, wildfire ecology, biodiversity monitoring, invasive species, colonial waterbirds, fisheries, Natura 2000, ecotourism, sustainable resource use, Black Sea porpoise

The review, published in Frontiers in Forests and Global Change, was conducted by Samuel Adeyanju of the University of British Columbia and colleagues in Canada, the Netherlands, and Indonesia. The team searched Scopus, Web of Science, and Google Scholar in November 2025, retrieving 667 peer-reviewed articles and screening them through a three-stage process managed in the systematic review platform Covidence. After deduplication and full-text assessment, 34 articles met the inclusion criteria, comprising 75 case studies across 24 CREMAs. The researchers deliberately avoided AI tools for the interpretive work, manually reviewing 111 full-text articles, and used Microsoft Copilot only to streamline data retrieval and visualization, following the RAISE recommendations for responsible AI use in evidence synthesis.

What they found first was a problem with the evidence itself. Research attention is heavily concentrated: the Wechiau and Zukpiri CREMAs alone account for 38 percent of all documented cases, with 16 and 13 case studies respectively, while 18 of the 24 CREMAs were represented by one to three studies each. Most well-studied sites were established before the formal CREMA framework and later folded into it, becoming internationally recognized model sites that attract disproportionate research funding and scrutiny. The authors warn that this geographic bias risks misrepresenting the state of community conservation nationally, potentially leading to flawed policy decisions built on the experiences of a handful of exceptional sites.

A second methodological weakness compounds the first. The reviewed literature is dominated by qualitative and mixed-methods designs, with interviews, surveys, and focus groups as the primary data collection tools. Remote sensing and ecological sampling appeared in only a small minority of studies. Most articles relied on people’s perceptions as evidence of CREMA outcomes, which the authors note makes it difficult to establish causal links between the conservation intervention and the reported results. Only a few cases, most notably Wechiau, provided baseline or longitudinal monitoring data. Publication years ranged from 2009 to 2025, with output peaking sharply in 2024 and 2025, each year accounting for one-third of the included articles.

On the environmental front, the picture is genuinely mixed. The most consistently reported positive outcome is increased conservation awareness, documented across 16 CREMAs, driven by environmental education, awareness campaigns, and local bylaw enforcement. Communities in CREMAs participate in wildfire prevention, tree planting, and reporting illegal encroachment, and several studies link these efforts to reductions in bushfires and illegal logging. Comparative evidence suggests greater awareness in CREMA communities than in non-CREMA areas, particularly where sustained engagement with NGOs and park authorities has occurred. At some sites, vegetation cover, soil quality, and biodiversity improved relative to adjacent areas, and firebreak construction reduced wildfire incidence. In the well-documented Wechiau case, conservation measures have been associated with increases in wildlife populations and species diversity, supported by ranger patrols and reinvestment of tourism revenues into conservation.

Yet the same review documents persistent deforestation, forest degradation, and land-use change across nine CREMAs. Research around the Ankasa Conservation Area showed continued conversion of forest land to cash crop plantations alongside significant declines in forest cover, and similar trends of expanding built-up areas and agricultural land were observed elsewhere. CREMAs have also struggled to control hunting and illegal logging in several contexts, with encroachment, poaching, and trade in flora and fauna persisting despite formal conservation rules. The authors conclude that while CREMAs can reduce certain pressures, they are not always sufficient to fully halt biodiversity decline, and only a few sites, most notably Wechiau and Zukpiri, demonstrate consistent evidence of conservation success.

Economically, the benefits are real but modest. Livelihood interventions such as beekeeping, agro-processing, small livestock rearing, and non-timber forest product commercialization provide supplemental income rather than substantial income transformation. Women are key beneficiaries, particularly in shea nut processing and groundnut farming. In Wechiau, the women’s shea group benefited 1,010 women across all 17 communities in 2009, with a projected harvest value of USD 34,564, and certified organic shea nuts earn women about GHS 30 more per bag than uncertified sales. Ecotourism has generated meaningful employment in only a few CREMAs: Wechiau’s visitor numbers leveled off at 1,500 to 1,800 annually, and by 2017 over 2,000 individuals, representing 22 percent of the estimated 9,415 residents, were employed on a salaried or commissioned basis. But these gains are typically incremental, vulnerable to market fluctuations, and constrained by limited access to conservation financing and continued reliance on external donor support.

The review also documents genuine costs. In some cases, CREMA leaders paid office utility bills and motorbike repairs from personal funds to keep operations running. Restrictions on hunting and forest product access reduced income opportunities for some households, and in one striking case, 30,000 hectares of private and communal land previously used for sugarcane farming were appropriated for the Avu Lagoon CREMA, with less than 30 percent of the original 1980s land area remaining available for cultivation. Such restrictions can adversely affect agricultural production and even drive youth out-migration in search of alternative income.

Socioculturally, the outcomes reflect a complex interplay between empowerment and exclusion. CREMAs can reinforce cultural norms, protect culturally significant species, and strengthen community cohesion through collective action platforms. Wechiau received the USD 20,000 UN Equator Prize in 2008, and infrastructure within the CREMA advanced faster than outside it: large settlements received eight times as many new amenities per community as those outside, and solar lighting programmes installed 60 times more lights per community within the CREMA between 2005 and 2009. By 2009 and 2017, 20 and 42 scholarship recipients respectively had completed secondary schooling, with some later returning as skilled tour guides and research assistants. But the review also documents elite capture, with a small group of individuals dominating leadership and decision-making power in eight CREMAs, fewer women occupying top management positions, and conflicts over ecotourism revenue distribution and alleged embezzlement eroding trust in four sites.

The authors conclude that CREMAs remain highly relevant due to their geographic scope, strategic locations outside formal protected areas, integration with rural development, and growing policy significance, particularly following the Wildlife Resources Management Act of 2023, which gave them statutory recognition. But they argue that CREMAs should be understood as a complementary conservation approach alongside multiple-use protected areas and other effective area-based conservation measures, not a standalone solution. Their recommendations are pointed: avoid overpromising livelihood benefits during establishment, mobilize long-term diversified financing including village savings and loan associations, empower local actors to enforce rules while ensuring fair compensation and inclusive participation, and establish robust monitoring systems to support adaptive management. For researchers, the call is to expand evaluation beyond northern Ghana and move beyond perception-based assessments toward remote sensing and longitudinal studies. As the broader community-based conservation literature has long suggested, the approach sometimes works and sometimes does not, and the difference lies in design, governance, financing, and context.

Subject of Research: Environmental outcomes of Ghana's CREMA initiative vary widely, with increased conservation awareness in some sites but persistent deforestation in others, while economic benefits remain modest and unevenly distributed.

Article Title: Environmental, economic, and sociocultural outcomes of Community Resource Management Area in Ghana: a systematic review

Article References: Adeyanju, S., Pienaah, C. K. A., Bulkan, J., Ros-Tonen, M. A. F., & Sunderland, T. (2026). Environmental, economic, and sociocultural outcomes of Community Resource Management Area in Ghana: a systematic review. Frontiers in Forests and Global Change, 9, Article 1961815. https://doi.org/10.3389/ffgc.2026.1961815

Image Credits: AI Generated

DOI: 10.3389/ffgc.2026.1961815

Keywords: CREMA, Ghana, community-based conservation, CBNRM, systematic review, biodiversity, livelihoods, ecotourism, Frontiers in Forests and Global Change, Environmental, economic, sociocultural

The study, authored by Kun Zhao of Luoyang Normal University, introduces a model called Green Anaconda Optimized Long Short-Term Transformer Memory, or GAO-LSTR. At its core sits a hybrid neural architecture that many natural language processing researchers already know well: the LSTM-Transformer. Long Short-Term Memory networks are a family of recurrent neural networks built with gated memory cells — forget, input, and output gates — that let them retain relevant information across a sequence while discarding noise. This makes them well suited to the way opinions unfold in a paragraph of tourist prose. Transformer components, by contrast, rely on self-attention to weigh relationships between distant words, capturing long-range dependencies that recurrent networks can struggle with. Zhao’s model chains the two together, feeding LSTM outputs into Transformer encoder blocks and fusing both representations before a final classification layer.

The ‘Green Anaconda’ part of the name refers to a nature-inspired metaheuristic that mimics the searching behavior of its namesake snake. Crucially, the algorithm is not used to train the network’s millions of weights directly — that would be computationally impractical for any population-based optimizer. Instead, GAO performs a global search over the hyperparameter space: learning rate, number of hidden units, batch size, dropout rate, and sequence length. Each candidate ‘anaconda’ in the population represents one complete hyperparameter configuration. Once GAO settles on promising settings, the actual network weights are learned conventionally, through backpropagation driven by the Adam optimizer. The author describes this as a hierarchical strategy: a global, gradient-free search over a low-dimensional space, paired with efficient gradient descent inside the network itself.

On the input side, the framework is deliberately multi-source. Text reviews are cleaned through tokenization and stop-word removal, then represented two ways at once. Term frequency–inverse document frequency, the classic statistical weighting scheme, highlights discriminative words while suppressing common ones. BERT, the pretrained bidirectional encoder, produces contextual embeddings that encode how meaning shifts with context — the difference, say, between a ‘small’ crowd and a ‘crushing’ one. Numerical attributes such as visitor age, travel distance, eco-rating, and service quality scores are normalized with StandardScaler so no feature dominates training simply by virtue of its range. All of this is fused before the hybrid network makes its predictions.

The experiments used the Ecotourism Sentiment and Satisfaction Dataset, a benchmark of 2,400 records containing text reviews, rating scores, and structured visitor attributes, split 70-20-10 into training, test, and validation sets. Each record carries a sentiment label — positive, neutral, or negative — and a satisfaction level of high, medium, or low. The headline results are striking: GAO-LSTR reached 98.13 percent accuracy on sentiment classification and 79 percent on satisfaction prediction, with precision of 97.5 percent, recall of 98.2 percent, and an F-measure of 97.8 percent. When retrained side by side under identical conditions, the proposed model outperformed Voting, SieBERT-Marrakech, VADER, GPT-4o, Support Vector Machines, Naive Bayes, and Random Forest baselines. VADER fared worst, hampered by a fixed lexicon that cannot parse domain-specific terms like ‘eco-lodge’ or ‘eco-education.’

The optimization component earned its keep in controlled comparisons. A plain LSTR trained with Adam alone achieved 95.1 percent accuracy; with SGD, 93.8 percent. GAO-LSTR hit 98.13 percent and converged in just 20 epochs, versus 28 for Adam and 35 for SGD. An ablation study broke the gains down piece by piece: standalone LSTM performed worst, adding the Transformer improved contextual learning, and GAO tuning pushed the full system to its best scores. Five-fold cross-validation confirmed stability, with the model averaging 0.9813 accuracy and an exceptionally tight confidence interval of [0.9794, 0.9831].

But the study is unusually candid about the caveats behind those numbers. The benchmark dataset is synthetic and semi-structured, with sentiment labels derived from predefined heuristic rules tied to the ratings themselves. That partial dependency between the rating attribute and the target label can inflate performance, and the author says so explicitly, framing the results as validation under controlled experimental conditions rather than proof of real-world generalization. The remarkably low variance across cross-validation folds reflects the homogeneity of the data, not a guarantee of similar stability on noisy, naturally annotated reviews. The satisfaction prediction accuracy of 79 percent, while solid, also lags well behind the sentiment figures, suggesting that mapping reviews onto discrete satisfaction tiers remains a genuinely hard problem.

Qualitative error analysis highlights where the system still stumbles. Misclassifications cluster around reviews expressing mixed sentiments — a visitor who adores the scenery but loathes the service — where the dominant positive context can drown out weaker negative cues. Sarcasm and implicit emotion remain difficult for even well-engineered transformers. The author also flags deployment realities: the stacked complexity of BERT, LSTM, Transformer, and metaheuristic tuning demands GPU-class hardware for real-time inference, and the black-box nature of the hybrid model complicates the explanations that tourism managers and stakeholders would want before acting on its outputs.

Those practical concerns shape the research agenda going forward. The author proposes testing the framework on large, human-annotated datasets drawn from platforms such as TripAdvisor and Google Reviews, extending validation across languages and cultures, and applying model compression techniques — quantization, pruning, knowledge distillation — alongside explainable AI methods to make predictions transparent and deployable on lighter hardware. Incremental learning is another suggested direction, allowing the model to adapt to streaming review data without full retraining. The payoff, if it materializes, would be real: destination managers who can spot dissatisfied visitors early, track sentiment trends around service quality and sustainability, and allocate resources based on evidence rather than intuition. For now, the green anaconda has demonstrated it can slither through a controlled benchmark with impressive precision. Whether it can handle the tangled, messy undergrowth of genuine human opinion is the test that comes next.

Subject of Research: A multi-source deep learning framework for sentiment analysis and satisfaction prediction in ecotourism consumption

Article Title: Sentiment analysis and satisfaction evaluation of ecotourism consumption under a multi-source deep learning framework

Article References: Zhao, K. (2026). Sentiment analysis and satisfaction evaluation of ecotourism consumption under a multi-source deep learning framework. Discover Artificial Intelligence, 6(1), Article 1420. https://doi.org/10.1007/s44163-026-01946-1

Image Credits: AI Generated

DOI: 10.1007/s44163-026-01946-1

Keywords: ecotourism, sentiment analysis, deep learning, LSTM, Transformer, BERT, TF-IDF, hyperparameter optimization, Green Anaconda Optimization, satisfaction prediction, sustainable tourism, natural language processing

The research team, led by Ayu Dewi Karuniawati of Universitas Gadjah Mada and Politeknik Bintan Cakrawala together with colleagues including Muhammad Ali Imron, applied a single-season occupancy model to the Bintan Resorts Tourism Area, a roughly 230 square kilometre stretch of the island’s northern coast. This zone is a mosaic of secondary lowland forest, coastal and mangrove forest, shrubland and built-up areas, containing 21 resorts and hotels within about four square kilometres of development. The area was designed around an eco-island concept, but expanding infrastructure continues to fragment the natural habitat on which the langur depends. Because the species’ entire range overlaps with this tourism landscape, understanding where it survives is not an academic exercise; it is the foundation for any plan that hopes to keep the monkey alive.

Between December 2023 and January 2024, the researchers walked the forest in a systematic grid-based survey. They overlaid a one by one kilometre grid across the study area, producing 232 sampling cells, and within each surveyed grid they traversed at least ten 100 metre transect segments with a 20 metre effective width, amounting to at least one kilometre of survey effort per cell. Over 53 days they covered 182 of the 232 grids, with the remainder treated as missing data because of inaccessibility, time constraints or dense shrubland. Because each grid was visited only once, the team used a clever statistical substitution: they treated the individual transect segments as spatial replicates, a recognised approach for occupancy modelling when repeat visits to remote sites are impractical. Every segment was scored simply as detection or non-detection, generating the detection histories the model required.

The raw numbers were sobering. Langurs were detected at 40 distinct locations, and they turned up in only 26 of the 232 grids, a naive occupancy of just 11.8 percent. But raw detections understate reality, because a shy, arboreal primate is easy to miss even when it is present. The intercept-only model, which ignores environmental covariates, pushed the estimated occupancy probability up to 39.2 percent, with a constant detection probability of just 4.32 percent. In other words, in any given survey segment where a langur group was actually present, the observers had less than a one in twenty chance of seeing it. That figure alone illustrates why occupancy modelling has become an indispensable tool in primate conservation: without correcting for imperfect detection, conservation planners would dramatically underestimate how much habitat the species still uses.

The team then built a suite of candidate models incorporating seven environmental covariates: food availability measured in the field, the Normalized Difference Vegetation Index, the Normalized Difference Building Index, the Normalized Difference Water Index, precipitation from CHIRPS satellite data, Land Surface Temperature derived from Landsat 8 thermal imagery, and distance to roads calculated from official topographic maps. A Spearman correlation test revealed that NDBI and NDVI were strongly negatively correlated, so NDBI was dropped to avoid multicollinearity. Detection was modelled as a function of NDVI and sampling effort. Using Akaike’s Information Criterion with a delta-AIC threshold of two, the researchers identified five statistically indistinguishable top models and averaged them, weighting parameter estimates by each model’s support to guard against selection bias. The best-fitting model, combining food availability, road proximity and vegetation cover, predicted an occupancy probability of 30.2 percent, with a 95 percent confidence interval running from 11.7 to 55.8 percent, and a detection probability of 4.31 percent.

The covariate relationships tell a coherent ecological story. Food availability showed a positive effect on occupancy, confirming the basic principle that folivorous primates cluster where their preferred plants grow. Field observations during preliminary surveys documented the langurs feeding on young leaves from species such as Adinandra dumosa, Syzygium grande and Rhizophora apiculata, flowers from Durio and Hibiscus tiliaceus, and seeds from Adenanthera pavonina, Acacia mangium and Hevea brasiliensis. Land Surface Temperature and NDWI, an index of vegetation water content and health, were also positively associated with occupancy, likely because warmer, well-hydrated vegetation supports the productive growth that herbivores need. Intriguingly, NDVI itself correlated negatively with occupancy, possibly because very dense vegetation reduces access to preferred food resources, even though NDVI showed a positive relationship with detection, perhaps because continuous canopy makes the arboreal langurs easier to spot.

One of the most consequential findings concerns roads. Occupancy was negatively associated with distance from roads, meaning the langurs are disproportionately found close to the island’s vehicle corridors. Roads can generate edge effects that promote the growth of pioneer plants whose young leaves and fruits are exactly what folivorous primates eat, so the pattern makes ecological sense. But it is also a deadly trap. Roadkill incidents have been frequently reported in the Bintan Resorts Tourism Area, and the tourism roads effectively bisect the forest, splitting habitat into left and right sides that arboreal animals must cross on the ground or through the canopy at perilous heights. The study’s overlay of the predicted occupancy map onto the resort masterplan revealed that planned future development coincides with areas of high food plant availability, threatening to destroy precisely the habitat the langurs favour most.

The authors translate these findings into a concrete conservation toolkit. They propose buffer zones around remaining langur habitat to limit encroachment and maintain connectivity, and canopy bridges that would allow langurs and other arboreal wildlife to cross roads safely without descending into traffic, an intervention proven elsewhere to reduce wildlife-vehicle collisions. They also identify an unexpected opportunity: the areas designated as Water Catchment under the master plan currently show low predicted occupancy because of limited canopy cover and open water, but their legal protection from intensive development means they could serve as refugia for langurs displaced by expanding infrastructure, provided that enrichment planting of key food species improves canopy continuity. Because the species lacks legal protection in Indonesia despite its IUCN endangered listing, such landscape-level planning by tourism managers may be the only realistic safeguard it has.

The study is honest about its limits. A single season of data cannot capture seasonal shifts in food availability, breeding cycles or patterns of human activity, and the authors caution that the estimates represent a snapshot rather than a year-round picture. They call for long-term monitoring, greater survey effort and the integration of additional covariates such as seasonal food phenology and predator presence. Still, the broader message resonates far beyond one Indonesian island. Small-island primates are exceptionally vulnerable to anthropogenic pressure, and tourism, which can fund conservation or destroy it, is expanding across exactly these fragile ranges. By showing that spatially explicit occupancy data can be overlaid directly onto development masterplans, the Bintan langur study offers a template for evidence-based wildlife tourism worldwide: find out where the endangered species lives, protect those places before the bulldozers arrive, and build the bridges, literal and institutional, that let development and conservation share the same map.

Subject of Research: Occupancy modelling of the endangered Bintan langur in a tourism landscape on Bintan Island, Indonesia

Article Title: Occupancy modelling for wildlife tourism and conservation planning of endangered Bintan langur (Presbytis siamensis rhionis)

Article References: Karuniawati, A. D., Putra, H. S. A., Sudarmadji, Priyambada, D., & Imron, M. A. (2026). Occupancy modelling for wildlife tourism and conservation planning of endangered Bintan langur (Presbytis siamensis rhionis). Discover Conservation, 3(1), Article 4. https://doi.org/10.1007/s44353-026-00074-2

Image Credits: AI Generated

DOI: 10.1007/s44353-026-00074-2

Keywords: Bintan langur, Presbytis siamensis rhionis, occupancy modelling, primate conservation, wildlife tourism, Indonesia, habitat fragmentation, road mortality, canopy bridges, endangered species, ecotourism, small island ecology

The research team, led by Xu Yuhui with Batunacun and colleagues, focused on Jiziwan, a region spanning roughly 790,000 square kilometres across Inner Mongolia, Gansu, Ningxia, Shaanxi and Shanxi. Home to about 48 million permanent residents, the area drew approximately 627.8 million tourists and 53.4 billion yuan in tourism revenue in 2023 alone. Yet it is also one of the most ecologically fragile corners of the Yellow River Basin, beset by soil erosion, land desertification and acute water scarcity. That combination of rich natural and cultural attractions and a delicate ecosystem makes Jiziwan an ideal natural laboratory for asking a question that has long dogged tourism scholarship: when a destination’s tourism economy bounces back from shocks, is that resilience anchored primarily in the environment, or in a broader constellation of economic, social and technological forces?

To answer it, the researchers first had to measure resilience itself. They built a comprehensive evaluation system grounded in Martin’s economic resilience theory, capturing four dimensions: resistance ability, the capacity to absorb shocks; recovery ability, the capacity to return to a dynamic equilibrium; reconstruction ability, the capacity to reorganise internal structures under stress; and update ability, the capacity to break path dependence through knowledge and innovation. Twenty indicators fed into this framework, ranging from the density of A-level tourist attractions and tourist arrivals to the number of days of good air quality, green coverage rates, highway passenger volumes, university enrolment and tourism research and development funding. Rather than weighting these indicators subjectively, the team applied a genetic algorithm-optimised projection pursuit model, a technique that projects high-dimensional data into a lower-dimensional space and identifies optimal weightings while sidestepping multicollinearity. An entropy weight method used as a cross-check produced broadly consistent rankings, strengthening confidence in the results.

The measurement revealed a tourism economy on the rise, but an uneven one. The composite resilience index for Jiziwan climbed from 0.175 in 2010 to 0.298 in 2023, dipping in 2020 under the impact of COVID-19 before rebounding to its peak. Kernel density estimation showed the entire distribution shifting rightward over time, but also exposed a persistent double peak: a large peak of cities stuck at low resilience and a smaller peak of high performers, with a right trailing tail signalling entrenched regional disparity. Spatially, resilience declined steadily from the centre of the region to its periphery in every year examined. Cities such as Hohhot, Ordos, Yulin and Taiyuan formed a resilient core, with Taiyuan standing alone at the high-resilience level by 2023, while five western cities, including Alxa, Bayannur, Wuhai, Zhongwei and Baiyin, remained trapped at low levels throughout the study period. Global Moran’s I statistics confirmed weak positive spatial autocorrelation, significant in 2015 and 2020 but not in 2010 or 2023, pointing to localised clustering rather than region-wide spatial dependence.

The methodological heart of the study, and its most novel contribution, lies in how the drivers were identified. Conventional tools each carry well-known limitations: geographic detectors struggle to rank drivers against one another, regression analysis leans on linear assumptions, structural equation models typically yield a single global optimum, and qualitative approaches are subjective and hard to visualise. The team instead turned to partial order theory and the Hasse diagram technique, mathematical tools previously applied to chemical risk assessment, groundwater quality and land degradation, but never before to tourism. Partial order theory ranks comparable elements across multiple criteria without requiring complete ordering, ignores multicollinearity within driver groups, and preserves the information of every indicator. The Hasse diagram technique then renders those partial orders as directed graphs in which maximal elements mark the most strongly influenced cities and chains trace the descending ranking of influence. Fifteen indicators across four driver groups, regional economy, social environment, ecological environment and technological innovation, were compared across 21 cities in three periods: 2010 to 2015, 2015 to 2020 and 2020 to 2023, generating twelve partially ordered sets and twelve Hasse diagrams.

The verdict was unambiguous. In the first period, the ecological environment dominated in 15 of the 21 cities, ahead of social environment with 11, regional economy with 9 and technological innovation with 8. In the second period, the ecological environment again led with 15 cities. By the third period, its reach had expanded to 18 cities, while technological innovation surged from 5 cities in the first period to 11, overtaking the regional economy to claim second place. The regional economy and social environment remained relatively stable influences, with the economy weakening in the northwest and strengthening in the centre-south, and the social environment showing the reverse pattern. The overall picture, the authors conclude, is a multi-dimensional synergistic driving pattern operating under the relative dominance of the ecological environment, a finding confirmed as statistically significant by fixed-effects panel regression in which all four driver groups and nearly all individual indicators showed significant associations with resilience.

Why does the environment loom so large here? The explanation lies in what Jiziwan’s tourism actually sells. Unlike heavily urbanised destinations that trade on cultural landscapes, service infrastructure and urban consumption, this region’s products rest on pleasant climates, vast grasslands, clean rivers, deserts, waterfalls and biodiversity. The quality of the ecological landscape is not one input among many; it is the raw material of the entire tourism economy. Technological innovation, the study argues, functions mainly as a tool for amplifying the value of those ecological resources and improving management efficiency, and cannot substitute for the environmental foundation that shapes attractiveness and resource supply. At the same time, tourism offers ecologically fragile regions a pathway to development that traditional resource-intensive industries cannot, converting ecological advantages into economic benefits through ecotourism, rural tourism and Yellow River cultural tourism, while channelling revenue back into biodiversity protection and environmental education.

Policy has amplified this natural advantage. The researchers identified twelve major ecological policies shaping the region, spanning three tiers: overarching national strategies such as the Three-North shelterbelt project and desertification control; Yellow River Basin-specific programmes linking ecological protection with cultural inheritance and tourism; and local initiatives including wetland restoration, ecological corridors and the development of the Yellow River cultural tourism belt. This cascade of macro-strategic guidance, river-basin support and local implementation has systematically improved environmental quality, raised tourism attractiveness and reinforced the ecological environment’s dominant position in the resilience system. The study also flags the exceptions that prove the rule: in Yulin, Yan’an and Shuozhou, cities at the intersection of the Loess Plateau and the Mu Us Sandy Land whose economies centre on coal and petroleum extraction, mining has so degraded sewage treatment capacity, surface stability and vegetation that the ecological environment exerted no positive influence on resilience in the most recent period.

The practical payoff is a differentiated policy map. The team classified the 21 cities into five strategic categories: ecological restoration-oriented cities such as Alxa, Bayannur, Ulanqab and Linfen, which should prioritise conservation and control development intensity; technology empowerment-oriented cities such as Hohhot, Taiyuan, Yan’an and Luliang, encouraged to deploy virtual and augmented reality and strengthen real-time risk monitoring; economic structure optimisation-oriented cities such as Baiyin, Luliang and Baotou, which can deepen industrial chains and cultivate new tourism business models; infrastructure improvement-oriented cities such as Ordos, Yulin, Baiyin and Wuhai, which need stronger transport, accommodation and public health systems; and regional coordination-oriented cities such as Hohhot, Baotou and Taiyuan, which should build cross-regional governance mechanisms, linked tourism routes and shared markets. Former mining cities, the authors suggest, could pursue green transformation through industrial heritage tourism.

The authors are candid about the limits of their analysis. Because the study area contains only 21 prefecture-level cities, spatial econometric models were not applied, and the results should be read as direct local associations rather than a full account of spatial spillovers between neighbouring cities. Future work with larger samples, longer panels and spatial lag, error or Durbin models could sharpen the picture. Even so, the study delivers a rare quantitative answer to a question usually left to qualitative argument, and its message resonates far beyond the Yellow River bend: in ecologically fragile destinations, protecting the environment is not a constraint on tourism resilience but its very foundation, and the first application of partial order theory to tourism research offers a transferable template for proving it.

Subject of Research: Drivers of tourism economic resilience in the ecologically fragile Jiziwan region of China's Yellow River Basin

Article Title: Ecological environment dominant or multi-dimensional drivers? A study on the drivers of tourism economic resilience in Jiziwan, China

Article References: Yuhui, X., Batunacun, Changan, Kaixin, L., Yufeng, Yongmei, & Dandan, Z. (2026). Ecological environment dominant or multi-dimensional drivers? A study on the drivers of tourism economic resilience in Jiziwan, China. Environmental and Sustainability Indicators, 32, Article 101505. https://doi.org/10.1016/j.indic.2026.101505

Image Credits: AI Generated

DOI: 10.1016/j.indic.2026.101505

Keywords: tourism economic resilience, ecological environment, Jiziwan, Yellow River Basin, partial order theory, Hasse diagram, GA-PP model, ecologically fragile regions, technological innovation, spatial heterogeneity, ecotourism, China