Forest fires, severe natural disasters causing substantial damage, necessitate accurate predictive modeling to guide preventative measures effectively. This study introduces an enhanced window-based Transformer time series forecasting model aimed at improving the precision of forest fire predictions. Leveraging time series data from 2020 to 2021 in Chongli, a myriad of forest fire influencing factors were ascertained using remote sensing satellite and GIS technologies, with their interrelationships estimated through a multicollinearity test. Given the intricate nature of real-world forest fire prediction tasks, we propose a novel window-based Transformer architecture complemented by a dual time series input strategy premised on 13 influential factors. Subsequently, time series data were incorporated into the model to generate a forest fire risk prediction map in Chongli District. The model’s effectiveness was then evaluated using various metrics, including accuracy (ACC), root mean square error (RMSE), and mean absolute error (MAE), and compared with traditional deep learning methods. Our model demonstrated superior predictive performance (ACC = 91.56%, RMSE = 0.37, MAE = 0.05), harnessing spatial background information efficiently and effectively utilizing the periodicity of forest fire factors. Consequently, the study proves this method to be a novel and potent approach for time series fire prediction.

The sustainable development of forests involves a combination of environmental protection and economics. Ecotourism is a new and growing sustainable economic model for forests, compared with the traditional utilization and protection of forests. The purpose of this study was to explore the applicability of the Prism model and SEM model in seeking to understand farmers’ perceptions and decision-making regarding tourism, so as to fully understand farmers’ thinking and behavior around forest tourism, and to truly make them in harmony with the sustainable development of forests. In this study, based on a field investigation of 392 farmers living in 11 national forest parks in Jiangxi Province, China, three hypotheses were constructed using the SEM model; and then six first-level indicators and 15 s-level indicators were constructed based on the Prism model, to verify the cognition and influencing factors for farmers, regarding sustainable forest tourism. The results showed that (1) the formation process of farmers’ willingness to adapt to forest tourism follows the path of “individual cognition → individual behavior”. (2) Economic influences, social culture, environmental cognition, and institutional cognition all have significant positive effects on their adaptive willingness. (3) Farmers’ cognition of economic life, environmental protection, and social culture lead to significant differences in their cognition and adaptability to forest tourism. These findings highlight the importance of the perceived value of tourism in assessing how tourism develops, as well as whether it is sustainable for the forest and its impacts. This study provides a better understanding of the factors that influence farmers’ attitudes towards tourism and highlights the importance of valuing local communities as important players in sustainable development, especially as this boosts the economy in forests.

The Nanwenghe Nature Reserve is located in the Daxing’anling forest region, which is abundant meadow resource which is prone to fire in fire season every year. In this paper, the change in the soil chemical properties of burned stands was emphasized. The results showed that: (1) Soil organic matter (SOM) concentration in burned stands showed an extremely significant difference from that in the control stands (p < 0.01). Alkali-hydrolysable N, available K and total N in the burned stands were significantly lower than those in the control stands (p < 0.05). (2) Total P, total K, pH and available P in the burned stands were not significantly different from the concentrations in the control stands (p > 0.05). (3) Available K, total N, total P and total K concentration showed no significant difference from June to September. The pH values increased with a variance of 8%, and pH values in July, August and September were significantly different from those in June. SOM concentrations obtained in July, August and September increased significantly, compared to the concentrations in June (p < 0.05). Alkali-hydrolysable N showed a decreasing trend for all conditions.

This study aims to develop an interpretable dual-layer information fusion framework driven by the Euclidean Consistency Index (ECI) for UAV flight parameter optimization, which integrates a statistically interpretable linear module with a nonlinear learning layer implemented via support vector regression to enhance 3D reconstruction accuracy in traffic accident scenarios. Experiments based on UAV oblique photogrammetry (27 configurations) were conducted to evaluate reconstruction accuracy using elevation error, horizontal error, and distortion anomaly metrics. The results reveal that flight altitude and image overlap exert coupled effects on reconstruction performance. An optimal parameter range is identified as 20–25 m altitude, 80–85% forward overlap, and 70–75% side overlap, achieving centimeter-level accuracy and stable geometric consistency. The high inter-layer agreement (ECI = 0.9133) confirms the robustness of the proposed fusion strategy. Validation in a real-world accident scene offers preliminary support for its practical applicability under similar conditions. Within the tested experimental scope, this framework provides a quantitatively interpretable and operationally stable foundation for UAV flight parameter selection in accident reconstruction.