Transformer-based approaches have shown promising performance in image restoration tasks due to their ability to model long-range dependencies, which are essential for recovering clear images. Although various efficient attention mechanisms have been proposed to address the intensive computational loads of transformers, they often suffer from redundant information and noisy interactions from irrelevant regions, as they consider all available tokens. In this work, we propose an Adaptive Sparse Transformer (AST-v2) to mitigate these issues by reducing noisy interactions in irrelevant areas and removing feature redundancy along channel dimension. AST-v2 incorporates two core components: an Adaptive Sparse Self-Attention (ASSA) block and a Feature Refinement Feed-forward Network (FRFN). ASSA adopts a dual-branch design, where the sparse branch guides the modulation of standard dense attention weights. This paradigm reduces the negative impact of irrelevant token interactions while preserving the important ones. Meanwhile, FRFN utilizes an enhance-and-ease scheme to eliminate feature redundancy across channels, enhancing the restoration of clear images. Experimental results on commonly used benchmarks show the competitive performance of our method for 6 restoration tasks, including rain streak removal, haze removal, shadow removal, snow removal, blur removal, and low-light enhancement. The code is available in the supplementary materials.

Snowfall and road surface freezing cause traffic disruptions and skidding accidents. When widespread extreme cold events or sudden heavy snowfalls occur, the continuous monitoring and management of extensive road networks until the restoration of traffic operations is constrained by the limited personnel and resources available to road authorities. Consequently, road surface condition prediction models have become increasingly necessary to enable timely and sustainable decision-making. This study proposes a road surface condition prediction model based on CCTV images collected from roadside cameras. Three databases were constructed based on different definitions of moisture-related surface classes, and models with the same architecture were trained and evaluated. The results showed that the best performance was achieved when ice and snow were combined into a single class rather than treated separately. The proposed model was designed with a simplified structure to ensure applicability in practical operations requiring computational efficiency. Compared with transfer learning using deeper and more complex pre-trained models, the proposed model achieved comparable prediction accuracy while requiring less training time and computational resources. These findings demonstrate the reliability and practical utility of the developed model, indicating that its application can support sustainable snow removal decision-making across extensive road networks.

To explore ways to protect dogs' paw pads from hot asphalt in the summer and snow removal chemicals in winter, a film was created by adding erythritol, a sugar alcohol that has a moisturizing effect and is a phase-changing material that undergoes an endothermic reaction when it changes from a solid to a liquid, and coconut oil, which is often used as a moisturizer, to a pullulan solution. When 15g of erythritol was evenly mixed with 50 mL of water, the temperature of the water decreased by 8 ℃, and when the coconut oil melted, the temperature decreased by 2.63 ℃. By utilizing the properties of these substances, a film was created. First, when a film was created using a ratio of 1/10 of erythritol (g):1% pullulan solution (mL), due to crystallization, it was difficult to maintain the shape of the film. When this film was dropped in water, it reduced the temperature by 8% more than the pullulan film, but when it was mixed with coconut oil, the temperature reduction wasn’t much different. To stabilize the film, the erythritol content was reduced, and the pullulan content was increased. The results of adding 1g and 3g of erythritol to 1% and 3% of 100mL pullulan solution showed that the 1g erythritol film was much more stabilized than the 3g erythritol film. When the films were dissolved in water, the 1g erythritol film didn’t show much effect, but the 3g erythritol film had a 2% temperature reduction in comparison to the pullulan film. Moreover, because the film dissolved more slowly if the density of the pullulan solution increased, the 3% pullulan film with erythritol dissolved more slowly than the 1% film, resulting in the 1% film showing a greater reduction in temperature. When coconut oil and sodium alginate were added to these films, the films were stable but showed no cooling effect. Because the 1% pullulan solution with 1g of erythritol was the most stable film, it was used as a model for the furtherance of other films. Through experimenting with the surface temperature of pig skin by comparing the temperature increase with the control, which was with no film, the film made with erythritol and coconut oil showed a 6% decrease in temperature compared to the control, and the moisture content increased by 70%. Therefore, using erythritol and coconut oil can effectively moisturize and reduce temperatures. Because it is a type of film, it is convenient and easy to use and can reduce the surface temperature of dogs’ paws, something that prior moisturizers couldn’t do. Therefore, this film can be seen to effectively treat the damages inflicted upon dogs due to outside activity in extreme conditions.

We propose an Event-Based Snow Removal algorithm called EBSnoR. We developed a technique to measure the dwell time of snowflakes on a pixel using event-based camera data, which is used to carry out a statistically optimal dwell time thresholding to partition event stream into snowflake and background events. The effectiveness of the proposed EBSnoR was verified qualitatively on a new dataset called UDayton25EBSnow comprised of front-facing event-based camera in a car driving through snow with manually annotated bounding boxes around surrounding vehicles, as well as a quantitatively using new snowflake event simulator called EBSnoGen. Qualitatively, EBSnoR correctly identifies events corresponding to snowflakes; and quantitatively, EBSnoR showed accuracy of 96.19%. Additional experiments showed that snow removal improved event-based object detection performance.

Urban snow and ice removal and storage: A systematic literature review

In this study, the thermal effectiveness of thermally conductive concrete pavements (TCPs) using silicon carbide (SiC) as a fine aggregate replacement was investigated, compared with that of ordinary Portland cement pavements (OPCPs). The most important purpose of this study is to improve the thermal performance of concrete pavement. Additionally, this study utilized improved thermal properties to enhance the efficiency of pavement heating to prevent icing and snow stacking. Both mixtures met the Korean standards for air content (4.5-6%) and slump (80-150 mm), demonstrating adequate workability. TCP exhibited a higher mechanical performance, with average compressive and flexural strengths of 42.88 MPa and 7.35 MPa, respectively, exceeding the required targets of a 30 MPa compressive strength and a 4.5 MPa flexural strength. The improved strength was mainly attributed to the filler effect and partly due to the van der Waals interactions of the SiC particles. Thermal conductivity tests showed a significant improvement in the TCP (3.20 W/mK), which was approximately twice that of OPCP (1.59 W/mK), indicating an enhanced heat transfer efficiency. In winter field tests, TCP effectively maintained high surface temperatures, overcoming heat loss and outperforming the OPCP. In the site experiment, thermal efficiency was clearly shown in the temperature at the center of the TCP, which was 3.5 °C higher than at the center of the OPCP at the coldest time. These improvements suggest that SiC-reinforced concrete pavements can be practically utilized for effective snow removal and ice mitigation in road systems.

Abstract The purpose of this study is to investigate the various external elements that have an effect on photovoltaic panels and lead to a decline in the efficiency of those panels. Accumulation of snow on photovoltaic panels lowers power generation. Moreover, it can be generated efficiently between 40% and 50%, depending on tilt angle. Dust buildup can reduce solar panel efficiency by 20%–30%, with the effect varying by altitude due to differences in dust levels. Bird droppings increase power loss and generate hot spots on the panel surface. This causes a reduction in performance by 20% to 30% and permanently damages the affected area of the solar cell. Variation in relative humidity influences the power output of photovoltaic panels, with increased moisture levels often leading to performance losses. Additionally, the rainy season further impacts efficiency due to reduced solar irradiance. 100% shading over photovoltaic panels may reduce approximately 60% of their total power loss, and partial shading may reduce approximately 10%–15% power loss. Whether manufacturing or installation, hardware failure leads to significant power loss in photovoltaic panels. A proper cleaning method is required for the removal of snow and dust from its surface, which leads to minimum power loss. Material properties need to be examined under mechanical and thermal stresses while manufacturing photovoltaic panels.

BackgroundGlobal climate change has led to dynamic changes in snow and ice cover, exerting a profound impact on the functional traits and reproductive adaptation strategies of perennial plants in alpine meadows. However, the underlying response mechanisms remain unclear. Through the field simulation experiments with different snowpack thicknesses in the alpine meadows of Qinghai-Tibet Plateau, we systematically analyzed the response patterns and regulatory mechanisms of the organ functional traits and reproductive allocation of four perennial plants: Saussurea nigrescens, Anaphalis lactea, Elymus nutans, and Koeleria macrantha.ResultsChanges in snowpack significantly affected resource allocation strategies of plants. In snow removal condition, S. nigrescens and A. lactea significantly reduced the biomass of vegetative organs, while K. macrantha significantly decreased its reproductive allocation under multi-fold snowpack. E. nutans maintained stable reproductive allocation under different snowpack treatments, demonstrating its high adaptability. Compared with the stability of carbon (TC) content, the nitrogen (TN) and phosphorus (TP) contents in various plant organs were more sensitive to changes in snowpack, and there were significant differences in nutrient responses among different organs and species. We found that Apart from E. nutans, the reproductive allocation of the other three plant species showed significant responses to soil temperature and/or moisture. The impact of changes in snowpack on reproductive traits was partially dependent on individual size, suggesting that plants had diverse adaptation strategies in resource allocation.ConclusionSignificant changes in snowpack profoundly influence plant resource allocation patterns among different functional organs, highlighting the diverse adaptation strategies plants employ under resource scarcity and environmental stress. Soil temperature and moisture are key factors influencing reproductive allocation in four plant species. The relationship between reproductive allocation and individual size varies across species. This study provides a scientific basis for understanding ecosystem functions and plant population maintenance mechanisms in alpine meadow ecosystems under climate change.

Spatial-Temporal Sequence Attention Based Efficient Transformer for Video Snow Removal

Face masks used during the COVID-19 pandemic are composed of polymers which when broken down release microplastics to the environment. As part of the most extensive monitoring program of COVID face mask littering ever undertaken for a coastal community, 50 parking lots in the town of Truro, Nova Scotia were surveyed for half a year (November 2021-April 2022). A total of 3,036 discarded or lost face masks were retrieved, with abandonment being consistent through time but for the notable exception of a period of rapid melting that released masks which had been damaged by snow removal maintenance. Qualitative observations and mark-and-recapture experiments indicated parking lots to be sources of litter dispersed to the wider environment. Interpolating these data on loss rates suggests that each of the 25,583 residents of Truro are estimated to have abandoned five face masks in parking lots during the 20 months of the peak pandemic. Expanding these results to the population of a quarter of a million people living around the Bay of Fundy, and using the known material composition of face masks, produces an estimate of 2,822 kg of pandemic plastic waste being generated with the potential to decompose and release microplastics over subsequent years into the Bay of Fundy, a designated World Heritage Site.

Introduction Rising global temperatures lead to milder winters with higher minimum temperatures and shorter periods of sub-zero conditions, but more frequent freezing and thawing in many boreal regions. Additionally, snow cover duration has decreased in recent decades in the northern hemisphere, and the overall amount of seasonal snow has declined (Pulliainen et al. 2020). This impacts the relationship between soil and air temperatures since snow acts as an insulator, helping to prevent deep soil freezing. Soil temperature and frost depth are crucial for the functioning of high-latitude ecosystems, as they influence tree water uptake, photosynthesis, and growth (e.g. Jyske et al. (2011), Lintunen et al. (2020)). For dwarf shrubs on the forest floor, snow’s insulating effect is even more critical in cold climates (Campbell et al. 2005). Soil temperature experiments are usually carried out in controlled environments such as laboratories or greenhouses, often focusing on shrubs or tree seedlings. However, some previous field studies have been conducted in mature boreal forests (e.g. Bergh and Linder (2001), Jyske et al. (2011)). In this study, we blocked snow from reaching the forest floor for two consecutive winters in a mature boreal forest to examine its effects on soil temperature, forest floor respiration, tree hydraulics, stem and root growth, as well as the growth of dwarf shrubs. Methods The experiment was conducted at SMEAR II station (Hari and Kulmala 2005) in southern Finland. The core of the SMEAR II stand with long-term tree-measurements was our control site and the snow removal treatment was done in its close proximity by preventing snow from falling to the ground with four shelters with total area of 360 m2. Each shelter was constructed around 2-3 study trees so that the tree canopies reached above the shelters. The shelters were removed for the snowless periods. The studied species were Scots pine, Norway spruce, silver birch, and a dwarf shrub bilberry. We measured air temperature, snow depth, soil temperature, soil moisture and root temperature at the control and treatment sites. Forest floor respiration was measured monthly with a dark manual chamber. Tree sap flow was measured with constant heat-dissipation probes and sap pressure from birches with pressure transducers. We measured stem diameter growth from tree cores and pine fine root growth with root ingrowth bags. We also measured coarse root hydraulic conductivity, bark osmolality and frost damages. Regarding dwarf shrubs, we measured diameter growth of the stem, elongation growth, fine root growth, leaf mass, leaf area, and the percentage of dead ramets. Results and discussion The absence of snow cover led to lower soil temperatures, which in turn reduced forest floor respiration during winter and spring. At the same time, the sensitivity of respiration to temperature appeared to increase, possibly due to the exposure of forest floor vegetation to cold air. The lack of snow also caused bilberry mortality, but the surviving plants grew taller and developed larger leaves, likely as a compensatory response to biomass loss. Tree hydraulics were also affected, with reduced water uptake in spring and a delayed start of the sap pressure season in birch. Pine and birch showed a tendency for reduced growth under snow exclusion, whereas spruce exhibited increased growth. However, previous studies have shown that the growth responses may have a time lag (Repo et al. 2021). Coarse root traits, such as water content and cellular frost damage, remained unaffected by the treatment. This study adds to our understanding of how changing snow cover influences springtime tree and forest floor processes in mature boreal forests. However, it also highlights the need for further research on mature trees to fully grasp the long-term impacts.

Due to the diversity and semi-transparency of snowflakes, accurately locating and reconstructing background information during image restoration poses a significant challenge. Snowflakes obscure image details, thereby affecting downstream tasks such as object recognition and image segmentation. Although Convolutional Neural Networks (CNNs) and Transformers have achieved promising results in snow removal through local or global feature processing, residual snowflakes or shadows persist in restored images. Inspired by the recent popularity of State Space Models (SSMs), this paper proposes a Mamba-based multi-scale desnowing network (SnowMamba), which effectively models the long-range dependencies of snowflakes. This enables the precise localization and removal of snow particles, addressing the issue of residual snowflakes and shadows in images. Specifically, we design a four-stage encoder–decoder network that incorporates Snow Caption Mamba (SCM) and SE modules to extract comprehensive snowflake and background information. The extracted multi-scale snow and background features are then fed into the proposed Multi-Scale Residual Interaction Network (MRNet) to learn and reconstruct clear, snow-free background images. Extensive experiments demonstrate that the proposed method outperforms other mainstream desnowing approaches in both qualitative and quantitative evaluations on three standard image desnowing datasets.

Abstract: This article examines the coverage of snow removal campaigns in the city of Ürümchi by the Chinese-language Xinjiang Daily ( Xinjiang ribao ) between 1937 and 1949. According to the discourse articulated by the state-controlled Daily , winter snow and resulting spring muds threatened the modern future toward which the newspaper and its editors sought to advance the city. During both the Sheng Shicai era (1933–1942) and the subsequent Nationalist era (1943–1949), snow removal was thus portrayed by the newspaper as requiring the mobilization of city residents. The Daily presented such mobilization as requiring a shifting combination of citizen participation and supervision by police and neighborhood headmen. The Daily ’s coverage of snow removal thus provides a striking example of the manner in which Ürümchi’s natural environment was used to construct and justify expansive state apparatus and citizen mass mobilization in the late Republican period in a manner that was also remarkably congruent with pre-and post-1949 efforts in China Proper.

The successful integration of photovoltaic (PV) generators in cities requires careful planning that accounts for possible factors influencing their operation. Numerous authors have extensively studied these factors; however, the urban environment has its unique characteristics. This study aims to conduct a narrative review of the most common and influential urban factors that impact the operation of PV modules and explore potential mitigation strategies. Based on preliminary knowledge on the topic, a methodology was proposed according to which they are classified into two categories: those enhanced by the urban environment and those specific to it. A total of 97 studies, mostly from the last decade, were selected based on the relevance and impact criteria. Shading, soiling, and snow were analyzed in an urban context, followed by different urban-specific factors, such as the urban landscape, pollution, and the limitations of PV mounting spots, which can lead to more than 50% performance losses. The performed review also identified the key and most promising approaches for mitigation of the abovementioned factors, such as electrostatic dust cleaning and forward bias current snow removal. Furthermore, recommendations for urban landscape planning were made in the context of PV integration. This review could also be useful for designers and operators of urban PV facilities by providing them with basic guidelines for their optimization.

Towards marine snow removal with fusing fourier information

Research on snow removal method of ultra high voltage DC composite bushing based on hot water jet

The paper considers the place of green structures in the blue infrastructure of cities. A scheme of integrated rainwater management using green structures is built. The combination of different green structures allows to creation of a unified and effective rainwater management system. The impact of green building structures on their supporting structures plays an important role. The loads from green roofs have two components: the load from structural elements and plants, including wind loads, and the load from precipitation-retained rainwater and snow. The first group of loads is constant, except for periodic wind loads, but its peak values vary little during rain and snow. It is impractical to consider snow load management. It can be reduced by snow removal. However, this will lead to a high risk of improper performance of snow removal duties with overloading of the supporting structures. Therefore, for safety reasons, the calculation is based on the maximum load. The load from rainwater depends on the runoff coefficient, which can be changed. Therefore, the paper underestimates the snow load for different snowy regions and average recurrence periods. The critical water retention of rainwater with the same load as the snow cover was determined. In the worst-case scenario of the first snow region and an average recurrence period of 10 years, we have a critical water retention of 56.2 dm3/m2, which is significantly higher than the intensity of precipitation. This means that the load of retained rainwater will be less than that of snow. Therefore, it’s necessary to ensure maximum water retention by the amount of precipitation. This cannot affect the bearing capacity of structures, which will be determined by the snow load. The possibility of utilising melt water for household needs is shown. The tasks for future research have been set.

Effective snow removal devices for road pavement using geothermal heat pipe

Study on snow removal characteristics and safety of the high-speed train operating through a snowdrift based on SPH-FEM coupling method

Snow accumulation on solar panels presents a significant challenge to energy generation in snowy regions, reducing the efficiency of solar photovoltaic (PV) systems and impacting economic viability. While prior studies have explored snow detection using fixed-camera setups, these methods suffer from scalability limitations, stationary viewpoints, and the need for reference images. This study introduces an automated deep-learning framework that leverages drone-captured imagery to detect and quantify snow coverage on solar panels, aiming to enhance power forecasting and optimize snow removal strategies in winter conditions. We developed and evaluated two approaches using YOLO-based models: Approach 1, a high-precision method utilizing a two-class detection model, and Approach 2, a real-time single-class detection model optimized for fast inference. While Approach 1 demonstrated superior accuracy, achieving an overall precision of 89% and recall of 82%, it is computationally expensive, making it more suitable for strategic decision making. Approach 2, with a precision of 93% and a recall of 75%, provides a lightweight and efficient alternative for real-time monitoring but is sensitive to lighting variations. The proposed framework calculates snow coverage percentages (SCP) to support snow removal planning, minimize downtime, and optimize power generation. Compared to fixed-camera-based snow detection models, our approach leverages drone imagery to improve detection precision while offering greater scalability to be adopted for large solar farms. Qualitative and quantitative analysis of both approaches is presented in this paper, highlighting their strengths and weaknesses in different environmental conditions.