Trees, Forests and People最新文献

{"title":"Assessing community's wildfire risk awareness: A case study of Mafra, Portugal","authors":"Sidra Ijaz Khan ,&nbsp;Jennifer L. Beverly ,&nbsp;Ana Catarina Sequeira ,&nbsp;Maria Conceição Colaço","doi":"10.1016/j.tfp.2026.101255","DOIUrl":"10.1016/j.tfp.2026.101255","url":null,"abstract":"<div><div>Wildfires pose a huge risk to communities, particularly in wildland-urban interface areas, such as the entire Mafra municipality in Portugal. This study investigates wildfire risk perception among residents of different villages of the Mafra municipality. The perceived risk is compared with the calculated biophysical exposure, which serves as a proxy for risk. A field survey was conducted among 205 participants across eight villages, collecting demographic data, risk perception, preparedness, and awareness of fire mitigation measures. Key trends in risk perception and behavioral responses were identified statistically. Results show that perceived wildfire risk was generally high (78%) but varied across villages and did not always align with biophysical exposure levels, revealing local mismatches between objective hazard and subjective assessment, alongside limited awareness of programs such as Aldeia Segura Pessoas Seguras. Integrating spatial exposure analysis with community-based perception data provides a comprehensive understanding of wildfire risk and highlights the need for targeted, context-specific communication and preparedness strategies in WUI communities. Future work should focus on longitudinal studies to track changes in wildfire perception and evaluate the effectiveness of risk mitigation policies. A holistic approach combining exposure and community assessment will be crucial in mitigating wildfire risks in the Mafra municipality and similar fire-prone regions.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101255"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147798047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Forestry-based carbon financing mechanisms and their role in advancing sustainable forest management: A systematic review","authors":"Jabulani Nyengere ,&nbsp;Harineck Mayamiko Tholo ,&nbsp;Sylvester Kumpolota ,&nbsp;Allena Laura Njala ,&nbsp;Andrew Jamali ,&nbsp;Trisha Chalulu ,&nbsp;James Akai ,&nbsp;Chimwemwe Gremu ,&nbsp;Misozi Kambanje ,&nbsp;Pamela Kuwali ,&nbsp;Precious Masuku ,&nbsp;Chikondi Chisenga ,&nbsp;Enoch Kwame Tham-Agyekum","doi":"10.1016/j.tfp.2026.101270","DOIUrl":"10.1016/j.tfp.2026.101270","url":null,"abstract":"<div><div>Forestry-based carbon financing mechanisms are significantly reshaping forest governance by converting emissions reductions and carbon removals into revenue streams that incentivize sustainable forest management, especially in the Global South. This systematic review synthesized peer-reviewed evidence on the types of forestry-based carbon financing mechanisms documented in the literature, how these mechanisms are designed and implemented, the environmental, social, and governance outcomes they produce, and the challenges and opportunities shaping their scalability. A PRISMA-guided systematic search retrieved 709 records across multiple academic databases, of which 73 studies met the eligibility criteria for final synthesis. The evidence base expanded sharply after 2022, with Asia and Africa representing the most studied regions. Reducing Emissions from Deforestation and Forest Degradation (REDD+) was the most commonly documented mechanism, featured in 41 studies, followed by voluntary carbon markets (18 studies) and Payments for Ecosystem Services (9 studies). Across mechanisms, results-based finance, robust measurement, reporting and verification systems, and community-based governance arrangements were the most commonly reported design features. Environmental outcomes were most consistently reported, particularly reduced deforestation, enhanced carbon sequestration, and biodiversity conservation, while social and governance outcomes were more variable and strongly shaped by tenure security, benefit-sharing equity, and institutional capacity. The most recurrent constraints were insecure land tenure, high transaction costs, weak governance and policy fragmentation, and market integrity risks including leakage and non-permanence. Nevertheless, opportunities were identified in the growing demand for high-integrity credits, integration with national climate policy frameworks, and advances in monitoring technologies. The review concludes that forestry-based carbon finance can contribute to sustainable forest management in the Global South, but its long-term success hinges on credible governance, transparent safeguards, and equitable benefit distribution, not just finance.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101270"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147798055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Understanding of three decade temporal dataset of mangrove distribution on shoreline expansion in Subang Northern Coastal Area, Indonesia","authors":"Nyoto Santoso ,&nbsp;Rizki Praba Nugraha ,&nbsp;Khairani Putri Marfi ,&nbsp;Aulia Ulfa ,&nbsp;Rahmat Asy’Ari","doi":"10.1016/j.tfp.2026.101260","DOIUrl":"10.1016/j.tfp.2026.101260","url":null,"abstract":"<div><div>Mangroves are vital ecosystems with temporal dynamics influenced by tidal fluctuations and anthropogenic coastal activities that cause high erosion of river estuary landscapes, thereby creating new space for mangroves to expand. Unfortunately, the temporal dynamics of coastal morphology and its relationship with mangroves have been understudied in Indonesia. Therefore, this study aims to examine the dynamics of mangrove cover in changing shoreline areas in Subang Regency, using Landsat spatial data. The Random Forest algorithm through the GEE platform plays a role in classifying mangroves annually from 1990 to 2025, obtaining an existing mangrove area of 261 ha in 2025. This area has decreased significantly compared to 1992, which reached 639 ha (Rsquared 0.63 and <em>p</em>-value &lt; 0.001). The <em>Avicennia marina</em> silvofishery system dominates the anthropogenic intervention area in the critical coastal landscape, following mangrove expansion. The shoreline change reached 46 m year^-1 and accretion formed due to sediment accumulation at an accretion rate of 23 ha year^-1, resulting in additional mangrove cover. However, the rate of aquaculture intervention in coastal areas was not proportional to mangrove succession in the accretion area, resulting in discrimination of mangrove areas up to the shoreline. In mangrove conservation efforts, regional and national policy support is urgently needed to regulate anthropogenic activities through the implementation of a mangrove silvofishery system. These efforts are being undertaken in response to the critical condition of mangroves on the northern coast of Java Island, with the aim of preserving their ecological benefits to support the national FOLU Net Sink 2030 target and annual NDC reporting.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101260"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147798636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modelling spatiotemporal dynamics of wattle plantations in northwestern Ethiopia using harmonised PlanetScope and RapidEye imagery","authors":"Celuxolo Michal Dlamini ,&nbsp;Trylee Nyasha Matongera ,&nbsp;Simon Lawson ,&nbsp;Madaline Healey ,&nbsp;Agena Tanga ,&nbsp;Kumela Regasa ,&nbsp;Weldesenbet Kassie ,&nbsp;Brett Phillip Hurley ,&nbsp;Ilaria Germishuizen","doi":"10.1016/j.tfp.2026.101293","DOIUrl":"10.1016/j.tfp.2026.101293","url":null,"abstract":"<div><div>Plantation forests within agroforestry systems play a critical role in supporting rural livelihoods across sub-Saharan Africa, yet their sustainability is increasingly threatened by land-use change, demographic pressure, and emerging biotic threats. In Awi Zone of Ethiopia, wattle (<em>Acacia</em> spp.) plantations are widely cultivated by smallholders, providing timber, fuelwood, charcoal, and income, while also contributing to soil amelioration and reducing pressure on natural forests. However, their invasive tendencies can also drive biodiversity loss and disrupt ecosystem services beyond intended planting sites. Balancing these benefits against ecological risks remains a global challenge. This study assesses the spatiotemporal dynamics of wattle plantations and associated Land Use Land Cover (LULC) changes in the Awi Zone from 2013 to 2025, with particular attention given to recent declines potentially linked to the emergence of wattle rust disease (<em>Uromycladium acaciae</em>). PlanetScope (3–5 m) and RapidEye (5 m) satellite images were harmonised and classified on Google Earth Engine (GEE) using a Random Forest classifier to generate LULC maps for five epochs (2013, 2016, 2019, 2022, and 2025). Classification performance was consistently high [Overall Accuracy (OA) 0.80–0.93; Kappa 0.70–0.91] for all periods. Change detection analysis indicated rapid expansion of wattle plantations between 2013 and 2016 (+181%) followed by a sustained and consistent decline with most pronounced gross loss of 181 km² (24.8%) between 2019 and 2022, largely driven by conversion to Other vegetation and alternative LULC classes. The trend continued until 2025, with a further 27.7% reduction in wattle plantation area to 527 km². Although rotational agroforestry practices partly explain cyclical changes, the timing and magnitude of recent declines may indicate that wattle rust reduced plantation viability and encouraged transitions to alternative land uses. This study highlights the value of remote sensing in monitoring disease risks, land-use pressures, and livelihood sustainability.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101293"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147849680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Landscape visual preferences in urban parks (Case study: Mashhad city)","authors":"Melika Hamedani,&nbsp;Seyed Hassan Taghvaei,&nbsp;Saeid Norouzian-Maleki","doi":"10.1016/j.tfp.2026.101283","DOIUrl":"10.1016/j.tfp.2026.101283","url":null,"abstract":"<div><div>The quality of the urban environment, particularly in parks and green spaces, plays a vital role in enhancing users’ visual experience and satisfaction. Despite its importance, visual preferences have often been overlooked in urban landscape design processes. This study aims to identify the key factors influencing users’ visual preferences in urban parks in Mashhad and to determine the primary objectives from the perspective of experts. The research questions focused on exploring the factors shaping visual preferences from the viewpoints of both urban park users and professionals in related fields. Using the Fundamental Values and Factors of Landscape (FVFL) model, 13 parks were selected from approximately 700 samples across the city, and the main landscape elements were extracted through a combination of field evaluations and comprehensive examinations. A park was then chosen as the primary case study for conducting the user survey. The research employed the Delphi technique with 18 experts and a questionnaire-based survey involving 61 park users. The findings revealed that both groups placed the highest importance on human-made factors. Users expressed the most excellent preference for pathways, while experts emphasized the aesthetic quality and visual appeal of water features. Furthermore, images containing a higher proportion of natural elements were rated more favorably by users. The main innovation of this study lies in its simultaneous assessment of both user and expert perspectives regarding the factors and goals influencing visual preferences—an approach rarely addressed in the previous research. The study faced certain limitations, including a small sample size due to time constraints and the seasonal limitation of the images, which were taken during only two seasons. Nonetheless, the results provide valuable insights into the design of urban parks by highlighting the importance of balancing human-made and natural components in accordance with the visual preferences of both users and professionals.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101283"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147849707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trapped in the middle: The provincial governance trilemma in Indonesia’s recentralized forestry sector","authors":"Muhammad Asgaf ,&nbsp;Ardian Halis ,&nbsp;Andi Sadapotto ,&nbsp;Andang Suryana Soma ,&nbsp;Muhammad Alif K. Sahide","doi":"10.1016/j.tfp.2026.101299","DOIUrl":"10.1016/j.tfp.2026.101299","url":null,"abstract":"<div><div>Recentralization of forest governance in Indonesia has shiftd key authorities from districts to provinces, with significant implications for forest sustainability and forest-dependent communities. While intended to improve coordination and control, its practical implementation and socio-ecological consequences remain insufficiently understood. This study examines how provincial governments become “trapped in the middle” of competing central and local pressures, facing a governance trilemma of coordination, capacity, and accountability. Using a qualitative case study of West Sulawesi, based on 41 in-depth interviews, document analysis, and focus group discussions, the study shows that provincial governance is shaped by three interrelated constraints. Coordination is limited by fragmented authority across central agencies and districts. Capacity is constrained by expanded mandates without sufficient personnel and fiscal resources. Accountability is weakened by unclear responsibility toward forest-dependent communities. Together, these dynamics create a self-reinforcing governance trap that leads to delayed enforcement, prolonged conflict resolution, and livelihood uncertainty. Rather than resolving decentralization failures, recentralization reconfigures them at the provincial level, where responsibility expands without corresponding authority or capacity. These findings show that improving forest governance requires realigning authority, capacity, and accountability, rather than relying on institutional restructuring alone.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101299"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147849809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Allometry and UAV-based estimation of tree dimensions, volume and aboveground biomass of Eucalyptus robusta plantations in the central highlands of Madagascar","authors":"Julien Sarron ,&nbsp;Iaviantsoa Ramanandraibe ,&nbsp;Jean-Pierre Bouillet ,&nbsp;Tantely Maminiaina Razafimbelo ,&nbsp;Bruno Bordron","doi":"10.1016/j.tfp.2026.101279","DOIUrl":"10.1016/j.tfp.2026.101279","url":null,"abstract":"<div><div>In the central highlands of Madagascar, smallholder eucalyptus plantations play a crucial role in meeting the energy needs of households. However, these plantations remain poorly characterized due to a lack of species-specific allometric equations, which hampers the precise assessment of tree volume and aboveground biomass (AGB). Additionally, there is currently no accessible method for estimating biomass at the plot scale. Moreover, volume and biomass measurements at this scale still rely on traditional forest inventories – methods that are both time-consuming and labor-intensive. This study aimed to calibrate species-specific models for estimating tree volume, and stem and total AGB in smallholder <em>Eucalyptus robusta</em> plantations. Allometric models were calibrated using destructive biomass measurements on 40 trees during the first stand rotation (4 to 6 years old), including both fertilized and unfertilized plots. Furthermore, two methods for estimating volume, stem AGB and total AGB using unmanned aerial vehicle (UAV) imagery were developed and compared. Allometric models yielded satisfactory results for predicting volume and biomass, with relative Bias% ranging from 0.9 ± 1.2 % to 4.7 ± 6.0 % and normalized root mean square error (NRMSE) between 11.2 ± 3.5 % and 20.6 ± 14.0 %. UAV-based models also produced promising results for estimating tree height and circumference at breast height (CBH), with R² of 0.93 and 0.76, respectively. The most effective approach combined UAV-derived estimates of height and CBH with species-specific allometric models. This study provides robust species-specific models for estimating <em>E. robusta</em> volume and biomass and demonstrates the potential of UAV imagery for characterizing tree plantations in smallholder contexts.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101279"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147798638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Financial evaluation of converting agricultural land to plantation forests in Ethiopia","authors":"Asabeneh Alemayehu ,&nbsp;Deginet Birhanu ,&nbsp;Tensaye Abate ,&nbsp;Kassa Chanie ,&nbsp;Gemedo Furo ,&nbsp;Mister Abebe","doi":"10.1016/j.tfp.2026.101282","DOIUrl":"10.1016/j.tfp.2026.101282","url":null,"abstract":"<div><div>Land is a fundamental factor of production and plays a critical role in shaping rural livelihoods and economic development. In many parts of Ethiopia, increasing conversion of agricultural land to plantation forestry has raised important questions regarding its financial viability and land-use implications. This study evaluates the financial performance of converting agricultural land to plantation forests, particularly <em>Eucalyptu</em>s and <em>Acacia decurrens</em>, in six districts of Ethiopia: <em>Cheha, Gumer, Jawi, Sinan, Hadero Tunto,</em> and <em>Boloso Sore</em>. Data were collected from 645 households using structured interviews, focus group discussions, and key informant consultations. A multi-stage sampling approach was employed to select study districts, kebeles, and households. Financial viability was assessed using Net Present Value (NPV), Benefit–Cost Ratio (BCR), and Internal Rate of Return (IRR) over a 15-year planning horizon, complemented by sensitivity analysis to evaluate the effects of changes in interest rates and market conditions. The results show that plantation forestry provides substantially higher financial returns than most traditional crops. Crop production generated an average NPV of 89,060.70 ETB per hectare, with a BCR of 1.29 and IRR of 46%, whereas Eucalyptus plantations achieved an NPV of 280,240.14 ETB per hectare and a BCR of 5.23. Similarly, Acacia plantations yielded an NPV of 136,909.21 ETB per hectare and a BCR of 2.26, outperforming crop-based systems in most cases. These findings indicate that plantation forestry represents a highly attractive land-use option for smallholder farmers. However, the results also highlight the importance of balanced land-use planning to ensure that plantation expansion does not compromise food production and long-term environmental sustainability<strong>.</strong></div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101282"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147850250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Balancing wood and food: assessing the drivers and spatiotemporal conflicts of Eucalyptus plantation expansion in Guangxi, China","authors":"Wen H. Cai ,&nbsp;Qiuxia Huang ,&nbsp;Jian Yang ,&nbsp;Le Yu ,&nbsp;Yuanzheng Yang ,&nbsp;Jiaxing Zu ,&nbsp;Jiali Wang ,&nbsp;Zhenxian Li","doi":"10.1016/j.tfp.2026.101292","DOIUrl":"10.1016/j.tfp.2026.101292","url":null,"abstract":"<div><div>Eucalyptus has become one of the most widely cultivated tree genera in subtropical China, generating significant economic returns but raising land-use conflicts with agriculture. Understanding the driving mechanisms behind this expansion is crucial for sustainable land-use management. To inform strategies for balancing wood production and agricultural stability, we investigated the spatiotemporal expansion dynamics and predicted the potential suitable habitat distribution of Eucalyptus plantations using a Boosted Regression Tree (BRT) model. Our analysis revealed three distinct spatiotemporal expansion phases. Crucially, this expansion has progressively shifted from marginal sloped terrain onto flat valley bottoms, intensifying the agriculture-forestry conflict in the region. Climatic conditions consistently emerged as the dominant large-scale regulatory driver across all expansion stages. The BRT model identified critical precipitation thresholds that dictate suitability, demonstrating Eucalyptus's strong sensitivity to the precipitation of the driest quarter (&gt; 110 mm) and mean annual precipitation (1000–1700 mm). Under current climatic scenarios, a large suitable habitat area is predicted in the central and southern parts of the study region. This study provides spatially explicit evidence of how meteorological factors shape the expansion patterns of short-rotation forestry and contribute to land-use tensions. Our findings offer actionable scientific insights crucial for developing climate-informed site selection and management policies to achieve a viable equilibrium between timber economic objectives and the protection of prime agricultural resources in subtropical China.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101292"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147849708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A hybrid deep learning model for cross-seasonal forecasting of forest fire occurrence from long-term survey data","authors":"Weike Li ,&nbsp;Feng Chen ,&nbsp;Liqing Si ,&nbsp;Cheng He ,&nbsp;Nuerlan Jianati ,&nbsp;Longxi Zhu ,&nbsp;Lifu Shu ,&nbsp;Mingyu Wang ,&nbsp;Fengjun Zhao ,&nbsp;Qiuhua Wang ,&nbsp;Guang Yang","doi":"10.1016/j.tfp.2026.101275","DOIUrl":"10.1016/j.tfp.2026.101275","url":null,"abstract":"<div><div>Climate change is exacerbating global forest fire regimes, creating an urgent demand for sophisticated predictive modeling tools. Focusing on the climatically sensitive and ecologically critical Daxing'anling Mountains in China, this study develops a hybrid deep learning framework termed CNN-ATT-SYNERGY. This model integrates Convolutional Neural Networks (CNN) with attention mechanisms to fuse multi-source environmental covariates, encompassing meteorological conditions, topographic features, lightning activity, anthropogenic disturbances, socioeconomic factors and historical fire records, thereby enabling robust forest fire occurrence assessment. Trained on a comprehensive dataset involving 3368 fire incidents across 20 forestry bureaus, the proposed model delivers state-of-the-art performance, with an overall accuracy of 81.22 % and an AUC of 87.97 %, surpassing conventional benchmark models including Support Vector Machine (SVM) and Random Forest. In addition to its superior predictive capacity, the model effectively identifies dynamic spatiotemporal fire risk patterns: fire risks cluster in eastern regions during spring and autumn, migrate westward in summer, and localized high-risk zones are also detected in winter. Accordingly, the CNN-ATT-SYNERGY framework serves as a seasonally adaptive and scalable decision-support instrument for forest fire management, providing a refined technical solution for boreal forest regions grappling with intensifying wildfire risks.</div></div>","PeriodicalId":36104,"journal":{"name":"Trees, Forests and People","volume":"25 ","pages":"Article 101275"},"PeriodicalIF":2.9,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147798637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}