Search Results (301)

Accurate short-term weather forecasting plays a vital role in disaster response, agriculture, and energy management, where timely and reliable predictions are essential for decision-making. Graph neural networks (GNNs), known for their ability to model complex spatial structures and relational data, have achieved remarkable success in meteorological forecasting by effectively capturing spatial dependencies among distributed weather stations. However, most existing GNN-based approaches rely on pairwise station connections, limiting their capacity to represent higher-order spatial interactions. Moreover, their dependence on supervised learning makes them vulnerable to spatial heterogeneity and temporal non-stationarity. This paper introduces a novel spatial–temporal pretraining framework, Hypergraph-enhanced Meteorological Pretraining (HyMePre), which combines hypergraph neural networks with self-supervised learning to model high-order spatial dependencies and improve generalization across diverse climate regimes. HyMePre employs a two-stage masking strategy, applying spatial and temporal masking separately, to learn disentangled representations from unlabeled meteorological time series. During forecasting, dynamic hypergraphs group stations based on meteorological similarity, explicitly capturing high-order dependencies. Extensive experiments on large-scale reanalysis datasets show that HyMePre outperforms conventional GNN models in predicting temperature, humidity, and wind speed. The integration of pretraining and hypergraph modeling enhances robustness to noisy data and improves generalization to unseen climate patterns, offering a scalable and effective solution for operational weather forecasting. Full article

Urbanization and cropland irrigation modify land surface water and energy budgets in different ways; however, few observational studies have explicitly quantified their contrasts. Using high-resolution observations from over 2000 surface weather stations and urban and irrigation fraction data, this study investigated the individual and combined effects of urbanization and cropland irrigation on surface air temperature over the Beijing–Tianjin–Hebei (BTH) region in China, where highly urbanized areas and heavily irrigated croplands exist together. The results indicate that (1) the daytime irrigation cooling (with surface air temperature decreasing by ~0.1–0.5 °C at irrigated stations) was non-negligible in late autumn, early winter, and later spring months, when winter wheat irrigation mainly occurred over the BTH region, while a slight warming was observed at many irrigated stations during the nighttime. By contrast, urban warming was most pronounced in the nighttime, especially in winter, and the daytime warming at urban sites was much weaker and comparable to the magnitude of cooling induced by concurrent irrigation for winter wheat. (2) Collectively, the vast stretches of irrigated croplands helped mitigate urban warming, and their combined effect on the daytime surface air temperature over the whole region resulted in a slight cooling of ~0.2 °C in some of the winter wheat-growing months. (3) The contrasting temperature changes due to urbanization and irrigation were spatially variable. Beijing was predominantly characterized by urban warming, while Shijiazhuang, with extensive irrigation, exhibited irrigation cooling (or slight warming) during the daytime (or nighttime) in most of the winter wheat-growing months, which could be a possible contributor to the daytime cooling (or stronger nighttime warming) at urban sites. This work highlights the temperature contrasts between urban areas and surrounding irrigated croplands, as well as the potential role of extensive irrigation in mitigating (or enhancing) daytime (or nighttime) urban warming. Full article

Accurate real-time icing grid fields are critical for preventing ice-related disasters during winter and protecting property. These fields are essential for both mapping ice distribution and predicting icing using physical models combined with numerical weather prediction systems. However, developing precise real-time icing grids is challenging due to the uneven distribution of monitoring stations, data confidentiality restrictions, and the limitations of existing interpolation methods. In this study, we propose a new approach for constructing real-time icing grid fields using 1339 online terminal monitoring datasets provided by the China Southern Power Grid Research Institute Co., Ltd. (CSPGRI) during the winter of 2023. Our method integrates static geographic information, dynamic meteorological factors, and ice_kriging values derived from parameter-optimized Empirical Bayesian Kriging Interpolation (EBKI) to create a spatiotemporally matched, multi-source fused icing thickness grid dataset. We applied five machine learning algorithms—Random Forest, XGBoost, LightGBM, Stacking, and Convolutional Neural Network Transformers (CNNT)—and evaluated their performance using six metrics: R, RMSE, CSI, MAR, FAR, and fbias, on both validation and testing sets. The stacking model performed best, achieving an R-value of 0.634 (0.893), RMSE of 3.424 mm (2.834 mm), CSI of 0.514 (0.774), MAR of 0.309 (0.091), FAR of 0.332 (0.161), and fbias of 1.034 (1.084), respectively, when comparing predicted icing values with actual measurements on pylons. Additionally, we employed the SHAP model to provide a physical interpretation of the stacking model, confirming the independence of selected features. Meteorological factors such as relative humidity (RH), 10 m wind speed (WS₁₀), 2 m temperature (T₂), and precipitation (PRE) demonstrated a range of positive and negative contributions consistent with the observed growth of icing. Thus, our multi-source remote-sensing data-fusion approach, combined with the stacking model, offers a highly accurate and interpretable solution for generating real-time icing grid fields. Full article

Crop water use (ET_c) is typically estimated as the product of crop evapotranspiration (ET_o) and a crop coefficient (K_c). However, the estimation of ET_o requires various meteorological data, which are often unavailable or of poor quality, particularly in countries such as Guinea-Bissau, where the maintenance of weather stations is frequently inadequate. The present study aimed to assess alternative approaches, as outlined in the revised FAO56 guidelines, for estimating ET_o when only temperature data is available. These included the use of various predictors for the missing climatic variables, referred to as the Penman–Monteith temperature (PMT) approach. New approaches were developed, with a particular focus on optimizing the predictors at the cluster level. Furthermore, different gridded weather datasets (AgERA5 and MERRA-2 reanalysis) were evaluated for ET_o estimation to overcome the lack of ground-truth data and upscale ET_o estimates from point to regional and national levels, thereby supporting water management decision-making. The results demonstrate that the PMT is generally accurate, with RMSE not exceeding 26% of the average daily ET_o. With regard to shortwave radiation, using the temperature difference as a predictor in combination with cluster-focused multiple linear regression equations for estimating the radiation adjustment coefficient (k_Rs) yielded accurate results. ET_o estimates derived using raw (uncorrected) reanalysis data exhibit considerable bias and high RMSE (1.07–1.57 mm d⁻¹), indicating the need for bias correction. Various correction methods were tested, with the simple bias correction delivering the best overall performance, reducing RMSE to 0.99 mm d⁻¹ and 1.05 mm d⁻¹ for AgERA5 and MERRA-2, respectively, and achieving a normalized RMSE of about 22%. After implementing bias correction, the AgERA5 was found to be superior to the MERRA-2 for all the studied sites. Furthermore, the PMT outperformed the bias-corrected reanalysis in estimating ET_o. It was concluded that PMT-ET_o can be recommended for further application in countries with limited access to ground-truth meteorological data, as it requires only basic technical skills. It can also be used alongside reanalysis data, which demands more advanced expertise, particularly for data retrieval and processing. Full article

The growing demand for electric vehicles (EV) has increased the need for reliable and sustainable charging infrastructure. To address this challenge, autonomous charging stations powered by renewable energy sources (RES) are a promising solution. This paper presents a simulation-based study that determines the optimal contribution of wind farms, photovoltaic systems, and energy storage to power an autonomous EV charging station. The simulation takes into account historical weather data, EV charging patterns, and renewable energy storage capacity. The results show that by combining RES and batteries, the charging station can operate autonomously minimizing the dependence on the power grid. Battery energy storage plays a key role in balancing intermittent RES generation and variable demand from the charging station. The simulation highlights the importance of adjusting parameters to optimize the energy utilization of the charging station and minimize the dependence on the grid. Further research is warranted to optimize the design, operation, and integration with advanced energy management systems to increase the efficiency and effectiveness of these charging stations. The development of a widespread autonomous charging infrastructure powered by renewable energy sources can accelerate the transition to clean transportation and support the energy system. Full article

The article describes an Arduino-based, portable, remotely usable weather monitoring station capable of measuring temperature, relative humidity, pressure, and carbon monoxide (CO) concentration and transmitting the collected data to the Cloud through the mobile telephony network. The main modules of the station are as follows: a DHT11 sensor for temperature and relative humidity sensing, a BMP180 sensor for pressure monitoring (with temperature compensation), a MQ7 sensor for the monitoring of the CO concentration, an Arduino Uno board, a GSM SIM900 module, and a buzzer, which is activated when the temperature exceeds 35 °C. The station operates as follows: the Arduino Uno board gathers the data collected by the sensors and, by means of the GSM SIM900 module, it transmits the data to the Cloud by using the mobile telephony network as well as the ThingSpeak software which is an open-code IoT application that, among others, enables saving and recovering of sensing and monitoring data. The main novelty of this work is the combined use of the GSM network and the Cloud which enhances the portability and usability of the proposed system and enables remote collection of data in a straightforward way. Additional merits of the system are the easiness and the low cost of its development (owing to the easily available, low-cost hardware combined with an open-code software) as well as its modularity and scalability which allows its customization depending on specific application it is intended for. The system could be used for real-time, remote monitoring of essential environmental parameters in spaces such as farms, warehouses, rooms etc. Full article

This study integrated the strengths of ensemble learning and soft computing to develop a future regional rainfall model for evaluating the complex characteristics of island precipitation. Soft computing uses the well-developed adaptive neuro-fuzzy inference system, which has been successfully applied in atmospheric hydrology and combines the features of neural networks and fuzzy logic. This combination enables artificial intelligence (AI) to effectively represent reasoning derived from complex data and expert experience. Due to the multiple atmospheric and hydrological factors that influence rainfall, the nonlinear interrelations among them are highly intricate. Nonlinear principal component analysis can extract nonlinear features from the data, reduce dimensionality, and minimize the adverse effects of data noise and excessive input factors on soft computing, which may otherwise result in poor model performance. Ultimately, ensemble learning enhances prediction accuracy and reduces uncertainty. This study used Tamsui and Kaohsiung in Taiwan as case study locations. Historical monthly rainfall data (January 1950 to December 2005) from Tamsui Station and Kaohsiung Station of the Central Weather Administration, along with historical and varied emission scenario data (RCP 4.5 and RCP 8.5) from three AR5 GCM models (ACCESS 1.0, CSIRO-MK3.6.0, MRI-CGCM3), were used to evaluate future regional rainfall trends and uncertainties through the method proposed in this study. The research findings indicate the following: (1) Ensemble learning results demonstrate that all examined general circulation models effectively simulate historical rainfall trends. (2) The average rainfall trends under the RCP 4.5 emission scenario are generally consistent with historical rainfall trends. (3) The exceedance probabilities of future rainfall during the mid-term (2061–2080) and long-term (2081–2100) suggest that Kaohsiung may experience precipitation events with higher rainfall than historical data during dry seasons (October to April of next year), while Tamsui Station may exhibit greater variability in terms of exceedance probabilities. (4) Under both the RCP 4.5 and RCP 8.5 emission scenarios, the percentage changes in future rainfall variability at Kaohsiung Station during dry seasons are higher than those during wet seasons (May to September), indicating an increased risk of extreme precipitation events during dry seasons. Full article

Global warming and the increasing frequency and intensity of heat waves are resulting in more frequent unfavourable weather conditions for outdoor activities, especially during the summer. The building environment can alter weather conditions, resulting in higher temperatures (urban heat island). Human towers are cultural activities that typically take place outdoors and were declared a UNESCO Intangible Cultural Heritage in 2010. The objectives of this study are (i) to analyse the weather conditions (i.e., temperature and relative humidity) during the human tower exhibitions, (ii) to determine discomfort during the exhibitions based on the heat index (HI) resulting from the combination of temperature and humidity, and (iii) to compare records from the square with those recorded in the nearest automatic meteorological station (AMS) belonging to the Catalan Meteorological Service network. To determine the weather conditions in the squares during the human tower exhibitions, a pair of sensors recorded temperature and relative humidity data in six exhibitions performed in the summer of 2024. The temperature exceeded 30 °C in five of the six human tower exhibitions analysed. In the cases of the Santa Anna exhibition (El Vendrell) and the Sant Fèlix exhibition (Vilafranca del Penedès), one of the sensors recorded temperatures above 30 °C throughout the entire duration of the exhibition. There was a predominance of HI values falling within the caution threshold in the two sensors of three exhibitions and within the extreme caution threshold in the two sensors of the other three exhibitions. The temperature is higher in urban squares than in the surrounding rural areas. The key factor is the urban heat island phenomenon, which poses health risks to both human tower builders and attendees. Adaptation measures are therefore necessary to guarantee the safety of the participants. Full article

The increasing frequency and intensity of heat waves, combined with urban heat islands (UHIs), pose significant public health challenges. Implementing low-cost, real-time monitoring networks with distributed stations within the smart city framework faces obstacles in transforming urban spaces. Accurate data are essential for assessing these effects. This paper compares different network types in a medium-sized city in western Spain and their implications for UHI identification quality. The study first presents a purpose-built monitoring network using Open-Source platforms, IoT technology, and LoRaWAN communications, adhering to World Meteorological Organization guidelines. Additionally, it evaluates two citizen weather observer networks (CWONs): one from a commercial smart device company and another from a global community connecting environmental sensor data. The findings highlight several advantages of bespoke monitoring networks over CWON, including enhanced data accessibility and greater flexibility to meet specific requirements, facilitating adaptability and scalability for future upgrades. However, specialization is crucial for effective deployment and maintenance. Conversely, CWONs face limitations in network uniformity, data shadow zones, and insufficient knowledge of real sensor situations or component characteristics. Furthermore, CWONs exhibit some data inconsistencies in probability distribution and scatter plots during extreme heat periods, as well as improbable UHI temperature values. Full article

Weather forecasting is essential for agriculture, yet current methods often lack the localized accuracy required to manage extreme weather events and optimize irrigation. The MAGDA Horizon Europe/EUSPA project addresses this gap by developing a modular system that integrates novel European space-based, airborne, and ground-based technologies. Unlike conventional forecasting systems, MAGDA enables precise, field-level predictions through the integration of cutting-edge technologies: Meteodrones provide vertical atmospheric profiles where traditional data are sparse; GNSS-reflectometry offers real-time soil moisture insights; and all observations feed into convection-permitting models for accurate nowcasting of extreme events. By combining satellite data, GNSS, Meteodrones, and high-resolution meteorological models, MAGDA enhances agricultural and water management with precise, tailored forecasts. Climate change is intensifying extreme weather events such as heavy rainfall, hail, and droughts, threatening both crop yields and water resources. Improving forecast reliability requires better observational data to refine initial atmospheric conditions. Recent advancements in assimilating reflectivity and in situ observations into high-resolution NWMs show promise, particularly for convective weather. Experiments using Sentinel and GNSS-derived data have further improved severe weather prediction. MAGDA employs a high-resolution cloud-resolving model and integrates GNSS, radar, weather stations, and Meteodrones to provide comprehensive atmospheric insights. These enhanced forecasts support both irrigation management and extreme weather warnings, delivered through a Farm Management System to assist farmers. As climate change increases the frequency of floods and droughts, MAGDA’s integration of high-resolution, multi-source observational technologies, including GNSS-reflectometry and drone-based atmospheric profiling, is crucial for ensuring sustainable agriculture and efficient water resource management. Full article

Temperature is a core element of the regional climate system and plays a key role in energy exchange and weather evolution. The current reanalysis of temperature data faces difficulties in providing more accurate geographical temperature data due to insufficient spatial resolution (0.25° × 0.25°). In this study, a lightweight downscaling method incorporating a convolutional neural network is proposed to construct a high-resolution temperature prediction model for the Macau region based on ERA5 reanalysis data. Aiming at the existing data due to insufficient resolution, a two-stage convolutional feature extraction module is introduced to optimize the model parameters by combining them with the observation data of Macau meteorological stations. The experimental results show that the accuracy of this method is 21.4% higher than that of the traditional interpolation method in the instantaneous prediction, and the prediction effect in the next 3 h is also very good. The model is expected to be extended to other regions in the future, providing an effective solution for obtaining long-term high-resolution temperature data in other regions, which can support the refinement of meteorological services and climate research. Full article

Accurate prediction of Global Navigation Satellite System-derived precipitable water vapor (GNSS-PWV), which is a crucial indicator for climate change monitoring, holds significant scientific value for climate disaster prevention and mitigation. In the study of GNSS-PWV prediction, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm within a decomposition–integration framework effectively addresses the non-stationarity and complexity of PWV sequences, enhancing prediction accuracy. However, residual noise and pseudo-modes from decomposition can distort signals, reducing the predictor system’s reliability. Additionally, independent modeling of all decomposed components decreases computational efficiency. To address these challenges, this paper proposes a hybrid model combining the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), autoregressive integrated moving average (ARIMA), and long short-term memory (LSTM) networks. Enhanced by local mean optimization and adaptive noise regulation, the ICEEMDAN algorithm effectively suppresses pseudo-modes and minimizes residual noise, enabling its decomposed intrinsic mode functions (IMFs) to more accurately capture the multi-scale features of GNSS-PWV. Sample entropy (SE) is used to quantify the complexity of IMFs, and components with similar entropy values are reconstructed into the following three sub-sequences: high-frequency, low-frequency, and trend. This process significantly reduces modeling complexity and improves computational efficiency. We propose different modeling strategies tailored to the dynamics of various subsequences. For the nonlinear and non-stationary high-frequency components, the LSTM network is used to effectively capture their complex patterns. The LSTM’s gating mechanism and memory cell design proficiently address the long-term dependency issue. For the stationary and weakly nonlinear low-frequency and trend components, linear patterns are extracted using ARIMA. Differencing eliminates trends and moving average operations capture random fluctuations, effectively addressing periodicity and trends in the time series. Finally, the prediction results of the three components are linearly combined to obtain the final prediction value. To validate the model performance, experiments were conducted using measured GNSS-PWV data from several stations in Hong Kong. The results demonstrate that the proposed model reduces the root mean square error by 56.81%, 37.91%, and 13.58% at the 1 h scale compared to the LSTM, EMD-LSTM, and ICEEMDAN-SE-LSTM benchmark models, respectively. Furthermore, it exhibits strong robustness in cross-month forecasts (accounting for seasonal influences) and multi-step predictions over the 1–6 h period. By improving the accuracy and efficiency of PWV predictions, this model provides reliable technical support for the real-time monitoring and early warning of extreme weather events in Hong Kong while offering a universal methodological reference for multi-scale modeling of geophysical parameters. Full article

Short-term control measures are often implemented during major events to improve air quality and protect public health. In preparation for the 11th National Traditional Games of Ethnic Minorities of China (denoted as “NMG”), held from 8 to 16 September 2019 in Zhengzhou, China, the authorities introduced several air pollution control measures, including traffic restrictions and dust control. In the study presented herein, we applied automated machine learning-based weather normalisation combined with an augmented synthetic control method (ASCM) to evaluate the effectiveness of these interventions. Our results show that the impacts of the NMG control measures were not uniform, varying significantly across pollutants and monitoring stations. On average, nitrogen dioxide (NO₂) concentrations decreased by 8.6% and those of coarse particles (PM₁₀) decreased by 3.0%. However, the interventions had little overall effect on fine particles (PM_2.5), despite clear reductions observed at the traffic site, where NO₂ and PM_2.5 levels decreased by 7.2 and 5.2 μg m⁻³, respectively. These reductions accounted for 56.3% of the NMG policy’s effect on NO₂ concentration and 73.2% of its effect on PM_2.5 concentration at the traffic site. Notably, the control measures led to an increase in ozone (O₃) concentrations. Our results demonstrate the moderate effect of the short-term NMG intervention, emphasising the necessity for holistic strategies that address pollutant interactions, such as nitrogen oxides (NO_X) and volatile organic compounds (VOCs), as well as location-specific variability to achieve sustained air quality improvements. Full article

This study examines the thermal impacts of green and blue infrastructure in Hilalkent Neighborhood, Elazığ City, in Turkey, using ENVI-met 5.6.1 software. Six design scenarios were proposed and their impact on air temperature, relative humidity, mean radiant temperature (Tmrt), physiological equivalent temperature (PET), and wind speed during August and January was analyzed. The simulation results were verified via field measurements using the Lutron AM-4247SD Weather Forecast Station at a height of 2.0 m above the ground. Data were collected in August 2023 and January 2024. The findings of this study indicate that existing vegetation in the study area provides a cooling effect of 0.8 °C during August. The addition of 10% grass coverage further reduced air temperature by 0.3 °C, while a 20% increase in tree density led to a 0.6 °C temperature reduction. The inclusion of a 10% water surface resulted in a 0.4 °C decrease in air temperature, and the implementation of extensive roof gardens contributed to an additional 0.2 °C reduction during the August period. The combined implementation of blue–green infrastructure in the study area achieved a total cooling effect of 1.5 °C during August. During January, the proposed scenarios led to a reduction in average temperatures by 0.1 °C to 0.4 °C compared to the base scenario, which may not be favorable for thermal comfort in colder conditions. Relative humidity values decreased during the August and Tmrt values were directly proportional to air temperature changes in both August and January. The results of this study provide valuable insights for urban planners and policymakers, demonstrating the effectiveness of blue–green infrastructure in mitigating the urban heat island (UHI) effect. These findings highlight the importance of integrating climate-responsive design strategies into urban planning to enhance thermal comfort and environmental sustainability in cities. Full article

Traditional station passenger flow prediction can no longer meet the application needs of urban rail transit vehicle scheduling. Station passenger flow can only predict station distribution, and the passenger flow distribution in general sections is unknown. Accurate short-term travel origin and destination (OD) passenger flow prediction is the main basis for formulating urban rail transit operation organization plans. To simultaneously consider the spatiotemporal characteristics of passenger flow distribution and achieve high precision estimation of origin and destination (OD) passenger flow quickly, a predictive model based on a temporal convolutional network and a long short-term memory network (TCN–LSTM) combined with an attention mechanism was established to process passenger flow data in urban rail transit. Firstly, according to the passenger flow data of the urban rail transit section, the existing data characteristics were summarized, and the impact of external factors on section passenger flow was studied. Then, a temporal convolutional network and long short-term memory (TCN–LSTM) deep learning model based on an attention mechanism was constructed to predict interval passenger flow. The model combines some external factors such as time, date attributes, weather conditions, and air quality that affect passenger flow in the interval to improve the shortcomings of the original model in predicting origin and destination (OD) passenger flow. Taking Chongqing Rail Transit as an example, the model was validated, and the results showed that the deep learning model had significantly better prediction results than other baseline models. The applicability analysis in scenarios such as high/medium/low passenger flow could achieve stable prediction results. Full article