Sustainable Forest Management and Natural Hazard Prevention

1. Introduction

Globally, forest ecosystems are shrinking at an alarming rate of 47,000 km² per year, with deforestation progressing at 100,000 km² per year [1]. Inadequate forest management and practices can damage or destroy forest resources and services, negatively altering the conditions of forest habitats and species. Additionally, forest ecosystems are being increasingly subject to environmental hazards, such as fires, storms, changes in rainfall patterns, droughts, extreme temperatures (heat and cold waves, extreme winter conditions, etc.), and pathological outbreaks [2,3,4,5]. Moreover, implementing and monitoring sustainable forest management remain challenging in an era of increasing types and intensities of environmental hazards [6,7]. While local and national governments might address anthropogenic threats locally, many environmental hazards require timely global responses to sustain forest resources. Sustainable forest management, combined with risk and vulnerability assessments, facilitates decision making and the development of management strategies to mitigate adverse effects and protect ecosystems. This Special Issue, “Sustainable Forest Management and Natural Hazards Prevention”, was launched with the aim to gather studies exploring interdisciplinary perspectives on forest sustainability to address management challenges posed by natural hazards. The scientific community was invited to contribute novel and original research addressing at least one of the following topics:

• Innovatively using state-of-the-art strategies, technologies, and methods/models for sustainable forest management;
• Innovatively using state-of-the-art strategies, technologies, and methods/models for environmental hazard prevention in forest ecosystems;
• Developing new approaches to support risk and vulnerability assessment in forest ecosystems;
• Adopting adaptation programs to enhance the resilience of forest ecosystems to environmental hazards;
• Determining the level of forest ecosystems vulnerability to environmental hazards;
• Determining the historical impact of environmental hazards on the sustainability of forest ecosystems;
• Predicting the future vulnerability of forest ecosystems to different climate change scenarios;
• Predicting the future species distributions in response to different climate change scenarios.

This Special Issue was also open to studies modeling the probability of the occurrence of any type of natural hazards (e.g., fire, wind, drought, deforestation, land degradation, landslide, flood, extreme temperature, earthquake, sea-level rise, and volcano) and man-made hazards (e.g., logging operations, road construction, oil spill, gas flare, and heavy metal contamination) at any spatial level (e.g., state/provincial, national, or international) and temporal scale (e.g., month, year, decade, and century).

After rigorous peer review, only submissions with high-quality scientific content and clear, cutting-edge contributions were accepted. Out of all the submissions, 14 articles (13 research articles and 1 review article) were accepted. The articles were from 74 authors across 18 countries: Austria, Bangladesh, China, Czech Republic, Finland, Germany, Iran, Italy, Poland, Saudi Arabia, Serbia, Republic of Korea, Sweden, Syria, Turkey, Ukraine, USA, and Vietnam.

2. An Overview of the Published Articles

The article by Salehnasab et al. (contribution 1) discussed that estimating the diameter increment of forests is crucial for management and planning. They explored two machine learning methods to develop diameter increment models for the Hyrcanian forests: the multilayer perceptron artificial neural network (MLP) and the adaptive neuro-fuzzy inference system (ANFIS). They first recorded the diameters at breast height (DBH) of seven tree species during two inventory periods (2003 and 2012) and categorized them into four groups: beech, chestnut-leaved oak, hornbeam, and other species. Then, they developed separate models for each group using the k-fold strategy for evaluation. The models were assessed using Pearson correlation coefficient, coefficient of determination (R²), root mean square error (RMSE), Akaike information criterion (AIC), and Bayesian information criterion (BIC). Despite low R² values, the correlation tests yielded significant results at a level of 0.01 for all groups. Among their models, ANFIS performed better for beech and chestnut-leaved oak, indicating a strong relationship between modeling techniques and tree species. This study was published online on 14 March 2022, in this Special Issue; as of 5 August 2024, it has received 10 citations according to Google Scholar.

The article by Park et al. (contribution 2) explored landcreep, a common natural hazard in the Republic of Korea. They highlighted that landcreep is often classified as a type of landslide despite the significant differences between the two. This classification persists due to the lack of verification on whether national landslide vulnerability criteria are applicable to areas prone to landcreep. Their study aimed to assess the applicability of these criteria specifically for landcreep. They conducted a correlation analysis of seven geomorphological environment criteria used for landslide-vulnerable areas on 57 landcreep areas. They found positive correlations only in the slope type and parent rock. When they applied the landslide vulnerability criteria to landcreep areas, they discovered that 61.4% had low or no possibility of landslides. Their overlapping analysis of the landslide hazard map and landcreep areas revealed that 67.6% were Level 3 or lower, excluding high hazard areas that were Levels 1 and 2, and 21.5% were landcreep areas not included in hazard levels. The authors concluded that applying landslide vulnerability criteria to landcreep-vulnerable areas is inappropriate, highlighting the urgent need for specific landcreep vulnerability criteria. Since its publication on 8 April 2022, in this Special Issue, this study has been cited once as of 5 August 2024, according to Google Scholar.

The article by Abdo et al. (contribution 3) addressed forest fires in the Al-Draikich region of western Syria. This study compared two techniques for mapping forest fire susceptibility: the frequency ratio (FR) and the analytic hierarchy process (AHP). The authors utilized an inventory map of 32 historical forest fire events from the summers of 2019, 2020, and 2021. They also considered 13 forest fire driving factors, including slope, aspect, curvature, elevation, normalized difference vegetation index (NDVI), normalized difference moisture index (NDMI), topographic wetness index (TWI), rainfall, temperature, wind speed, and proximity to settlements, rivers, and roads. Their results indicated that the FR technique outperformed the AHP method in mapping forest fire susceptibility. The authors concluded that their findings provided valuable spatial insights that could inform and enhance forest management strategies, ultimately contributing to better forest fire prevention and mitigation efforts. Since its publication on 13 April 2022, in this Special Issue, this study has received 39 citations as of 5 August 2024, according to Google Scholar.

The article by Salehnasab et al. (contribution 4) argued that the Hyrcanian forests of Iran are primarily managed using the single-selection silvicultural technique. While selection cutting offers significant ecological benefits, it presents challenges in maintaining and practicing successful forestry compared to even-aged systems. Therefore, the authors set out to provide relevant management tools in the form of models to estimate low growth levels in the Hyrcanian forests. They focused on estimating the population growth and allowable cut rates using a matrix model in the Gorazbon district and utilized data from 256 permanent sample plots measured between 2003 and 2012, along with records of harvested trees according to the forestry plan, to achieve this. The most frequently occurring tree species were categorized into four groups: beech, hornbeam, chestnut-leaved oak, and other species. The district’s compartments were divided into logged and unlogged groups to estimate the allowable cut and compare it with observed and predicted volumes from forestry plans. The results revealed that the total operated allowable cut (OAC) in logged compartments exceeded the estimated allowable cut (EAC), while the total predicted allowable cut (PAC) was higher than the EAC in unlogged compartments. A comparison of EAC and OAC showed that hornbeam was harvested beyond its potential, whereas chestnut-leaved oak and other species exhibited opposite trends. The authors concluded that their models could offer significant advancements for estimating allowable cuts, thereby enhancing sustainable forestry practices. This study was published on 1 June 2022, in this Special Issue, and has received four citations by 5 August 2024, according to Google Scholar.

The article by Sresto et al. (contribution 5) investigated changes in vegetation patterns and heat-island zones in Dhaka, Bangladesh, before and after the COVID-19 lockdown. This study aimed to understand how the lockdown affected environmental pollutants and the urban heat island across different land use and cover types. They collected Landsat-8 images to determine vegetation patterns and heat island zones. They derived the normalized difference vegetation index (NDVI) and the modified soil-adjusted vegetation index (MSAVI12) to analyze vegetation patterns. The NDVI results showed that the health of the vegetation improved after one month of lockdown. For MSAVI12, the highest coverages in March of 2019, 2020, and 2021 (ranging from 0.45 to 0.70) were 22.15%, 21.8%, and 20.4%, respectively. In May of those years, dense MSAVI12 values accounted for 23.8%, 25.5%, and 18.4%, respectively. At the beginning of the lockdown, the calculated land surface temperature (LST) for March 2020 was higher than in March 2019 and 2021. However, after more than a month of lockdown, the LST decreased in May 2020. Following the lockdown in May 2020, the highest urban heat island (UHI) values, ranging from 3.80 to 5.00, covered smaller land-cover regions and reduced from 22.5% to 19.13%. However, after the lockdown period ended, industries, markets, and transportation resumed, leading to the expansion of heat island zones. The authors demonstrated strong negative correlations between the LST and vegetation indices. Since its publication on 22 June 2022, in this Special Issue, this study has received seven citations as of 5 August 2024, according to Google Scholar.

The article by Varamesh et al. (contribution 6) addressed the question of “How do different land uses/covers contribute to land surface temperature and albedo?” and explored the spatiotemporal variability in land surface temperature (LST) and land surface albedo (LSA) across different land use/cover (LULC) classes in northwest Iran. They began by applying an object-oriented algorithm to 10 m resolution Sentinel-2 images from the summer of 2019, creating a detailed LULC map for a 3284 km² region. Next, they calculated the LST and LSA for each LULC class using the SEBAL algorithm, which was applied to Landsat-8 images from the summer of 2019 and the winter of 2020. The findings revealed that during the summer, barren land exhibited the highest LSA value (0.33), while water bodies had the lowest (0.11). In winter, the highest LSA values were found in farmland and snow cover, whereas forest areas had the lowest (0.21). As for LST, rangeland recorded the highest temperature in summer (37 °C), and water bodies the lowest (24 °C). In winter, forests had the highest LST (4.14 °C), while snow cover had the lowest (−21.36 °C). The authors concluded that barren land and residential areas could somewhat mitigate heating effects with their high LSA in summer. Conversely, forest areas contributed more to regional warming than other LULC classes due to their low LSA and high LST, especially in winter. This suggests that forests might not always mitigate the effects of global warming as much as previously thought. This study was published on 17 December 2022, in this Special Issue, and has been cited twice as of 5 August 2024, according to Google Scholar.

The article by Li et al. (contribution 7) evaluated five machine learning methods combined with Sentinel-2A data band features to identify and extract mangrove forests in Dongzhai Harbor, northeast Hainan Province, China. They found that the XGBoost algorithm had the highest accuracy and used it to identify and extract information on mangrove forests across Hainan Island, focusing on five periods from 2000 to 2020. They integrated the landscape pattern index, dynamic attitude model, and mathematical statistics to analyze trends over these 20 years. The results revealed several key findings: (i) The total mangrove landscape area on Hainan Island initially decreased and then increased between 2000 and 2020. Over the past 20 years, the mangrove area grew by 1315.75 hectares, with an annual change rate of 65.79 hectares per year. (ii) From 2000 to 2020, the mangroves on Hainan Island experienced increased fragmentation, heterogeneity, and richness and decreased connectivity. The proportion of each landscape type tended to become more balanced. (iii) Natural factors, such as the annual average temperature in the study area, were the primary drivers of the large-scale reduction in mangroves and the deepening of landscape fragmentation. Human factors and the impact of macro-policies also played significant roles. The findings of this study provide valuable insights for future remote sensing data extraction from mangrove forests and their ecological protection and restoration on Hainan Island. This study was published on 31 December 2022, in this Special Issue, and has received eight citations as of 5 August 2024, according to Google Scholar.

The article by Imbrenda et al. (contribution 8) was a comprehensive review of recent dynamics in the European forestry sector, highlighting the complexity of the environmental–economic nexus. The extensive use of wood has social implications for local districts adapting to ecological changes, such as climate warming and landscape transformations. The technical–economic dimension of forestry plays a vital role in sustainable development, affecting economic dynamics, sector growth, supply chain organization, company interconnections, and investment strategies. A major issue is the low reliability of official statistics on forest resources. Forestry practices are crucial for maintaining habitats and species while increasing sustainable timber production. The European Commission encourages public and private sectors to adopt circular economy practices, leading to job creation, recycled materials, reduced CO₂ emissions, and increased community value. Forestry should holistically contribute to sustainable development, focusing on economic and environmental targets and fostering cooperation between member countries and regional authorities. This study was published on 28 June 2023, in this Special Issue, and has received 10 citations as of 5 August 2024, according to Google Scholar.

The article by Milanović et al. (contribution 9) evaluated the contribution of variables obtained from open-source datasets (such as MODIS, OpenStreetMap, and WorldClim) for modeling fire ignition probability at the country levels of Austria (AT) and Czech Republic (CZ) and investigated the performance of the Random Forest (RF) method across different countries. They assessed the importance of the predictors using the Gini impurity method and employed the RF method to evaluate the ROC-AUC and confusion matrix. The topographic wetness index emerged as the most important variable in the AT model, and slope was the most significant in the CZ model. The AUC values in the validation sets were 0.848 for the AT model and 0.717 for the CZ model. When applied to the entire dataset, the models achieved accuracies of 82.5% (AT model) and 66.4% (CZ model). Cross-comparison revealed that the CZ model could be successfully applied to the AT dataset (AUC = 0.808, Acc = 82.5%), while the AT model showed poor explanatory power when applied to the CZ dataset (AUC = 0.582, Acc = 13.6%). This study provided insights into how the accuracy and completeness of open-source data affect the reliability of national-level forest fire probability assessments. This study was published on 16 March 2023, in this Special Issue, and has received one citation as of 5 August 2024, according to Google Scholar.

The article by Sheng et al. (contribution 10) addressed the illegal logging trade (ILT), as a major cause of global deforestation and ecological unsustainability, analyzed wood trade data to assess the current status of ILT in China, summarized the efforts and shortcomings of China’s response actions, and discussed potential strategies and sustainable development prospects for combating ILT in the future. The authors found that the volume of ILT in China gradually increased from 2013 to 2020, potentially contributing to the slowdown in global ecological sustainability. The Chinese government and NGOs have taken numerous actions to address ILT through legislation, industry supervision, and international cooperation, achieving some positive results. However, further efforts are needed to limit and manage ILT to ensure the sustainable development of forest resources. Strengthening legislation, particularly restricting ILT clauses, is crucial as a mandatory policy to solve the ILT problem, providing a legal basis for other actions. Economic incentives to encourage the import of legal wood can also help reduce illegal wood trade. Promoting an international certification system for wood and standardizing logging practices are effective ways to mitigate ILT. Improving the management of wood imports into China will address the critical gap in global efforts to combat ILT and have positive impacts on reducing global ILT. This study was published on 10 August 2023, in this Special Issue, and has not yet been cited as of 5 August 2024.

The article by Küçükarslan et al. (contribution 11) prioritized occupational health and safety (OHS) achievements across ten forest management directorates in a specific province of Turkey, using multi-criteria decision-making techniques (MCDMT). A key focus of this research was the development and application of a sophisticated OHS performance model to critically assess the OHS performance within these directorates. The evaluation considered static variables such as land use and cover, slope, vegetation type, soil characteristics, and proximity to highways and human settlements. Dynamic variables like temperature, wind speed and direction, and humidity were also included. The findings highlighted the significant tangible and intangible impacts of workplace accidents, emphasizing the need for extensive, cross-industry research initiatives aimed at effective accident prevention. Using the Analytical Hierarchy Process (AHP), Fuzzy Analytical Hierarchy Process (F-AHP), and the Preference Ranking Organization Method for Enrichment Evaluations (PROMETHEE), this study demonstrates the effectiveness of MCDMTs in assessing and categorizing OHS performance. This study was published on 11 October 2023, in this Special Issue, and has been cited twice as of 5 August 2024, according to Google Scholar.

The article by Wu et al. (contribution 12) argued that clearing diseased wood is a common practice to reduce the spread of pine wilt disease and prevent it from infecting other pines. However, the impact of this practice on the physicochemical properties of soil remains unclear. To address this, the authors conducted a series of soil experiments and observations in Changdao, China, focusing on cut and uncut black pine and the inter-forest zone. The results revealed several key findings: (i) The soil beneath the forest began to exhibit characteristics of both forest and grassland after the diseased wood was cleared, indicating a potential shift in the ecosystem’s structure and function. (ii) Clearing diseased wood led to an increase in soil pH by 0.15, suggesting it helped in the recovery of acidic soil. (iii) The practice also coarsened the soil’s texture and effectively reduced the surface soil temperature (0–20 cm) in summer by 1.52 °C. Additionally, it significantly decreased surface soil moisture (0–20 cm) in spring and summer by 1.3% and 2.43%, respectively. (iv) Clearing diseased wood altered the content of essential nutrients in the soil. It reduced available nitrogen by 26.86 mg·kg⁻¹, increased available phosphorus by 0.57 mg·kg⁻¹, and decreased available potassium by 1.68 mg·kg⁻¹. However, it also exacerbated soil salinization, increasing the soil’s salt content by 0.70 g·kg⁻¹. These findings provide scientific insights for the sustainable ecological development of black pine forests in Changdao. This study was published on 15 November 2023, in this Special Issue, and has not yet been cited as of 5 August 2024.

The article by Thai et al. (contribution 13) reported that various activities were undertaken to sustain the mangrove forest on peat soil remnants in the Mekong Delta region following the largest forest fire in Vietnam in 2002. These activities included promoting natural regeneration, afforestation, and rapid forest restoration measures, along with protective measures such as rainwater retention to maintain moisture levels for fire prevention. However, two critical challenges emerged: allowing the forest to naturally regenerate would lead to annual forest fires, while maintaining a constant water level through year-round water retention would harm biodiversity. The authors conducted a study in U Minh Thuong National Park to address forest regeneration. After the major forest fire, various measures were taken to promote forest regeneration, including afforestation, silvicultural solutions, and hydrological techniques such as rainwater storage to maintain humidity and prevent future fires. The authors used a hand drill to collect samples and set up 15 plots to survey the growth of the forest at three-peat thickness levels. At each of the three collection sites, samples of one kilogram were collected and labeled as UTM1, UTM2, and UTM3. The authors found chemical composition of peat water changed significantly due to the rainy and dry seasons, affecting forest growth. The study highlighted the importance of considering multiple factors when developing effective forest restoration strategies. This study was published on 10 January 2024, in the Special Issue and, as of 5 August 2024, has not yet been cited.

The article by Tavankar et al. (contribution 14) investigated the growth characteristics and architecture of beech (Fagus orientalis Lipsky) seedlings grown in three different microenvironments based on canopy and soil conditions. The experimental treatments included skid trails (removal of canopy and compacted soil), winching corridors (natural canopy and compacted soil), felling gaps (removal of canopy and natural soil), and a control area (canopy and soil in natural state). The results showed that many growth and architectural indicators of seedlings were significantly less favorable in skid trails and winching corridors compared to the control area. These indicators included the length and biomass above and below ground and the ratio of root length to stem length. However, the status of these indicators was more favorable in felling gaps than in the control area. The seedling quality index decreased by 12.2% and 4.9% in skid trails and winching corridors, respectively, but increased by 2.4% in felling gaps compared to the control area. The growth characteristics and biomass of seedlings had a significant negative correlation with soil bulk density and penetration resistance and a significant positive correlation with soil porosity, moisture, and organic matter content. These results indicated that creating a gap in the stand canopy from cutting individual trees created a favorable microenvironment for seedling growth, while soil compaction caused by logging operations created an unfavorable microenvironment. The authors concluded that it was necessary to plan and execute the extraction of cut trees in a manner that reduced the extent and severity of soil compaction, to preserve and maintain the stability of the forest ecosystem. This study was published on 23 May 2024, in this Special Issue and has not yet been cited as of 5 August 2024.

3. Conclusions

Fourteen articles published in this Special Issue collectively highlighted the critical importance of advanced techniques, reliable data, and tailored approaches in addressing sustainable forest management. They emphasized the necessity of continuous, adaptive management practices to achieve long-term sustainability and resilience. The studies underscored the complex interplay between human activities and environmental health, advocating for the integration of advanced methodologies and targeted strategies in forest management, ecological restoration, and occupational health and safety. Additionally, they stressed the need for accurate statistics, strong legislation, economic incentives, and international cooperation to combat issues like illegal logging and enhance safety measures. Overall, these findings advocate for a comprehensive and holistic approach to managing forest resources and ensuring sustainable development.