With the rise of smart cities and digital tourism, understanding tourist movement is essential for enhancing personalized experiences and informing business decisions. Traditional methods struggle to integrate multi-source data and interpret complex behaviors, while large language models (LLMs) offer untapped potential. This study proposes an LLM-driven “Perception-Modeling-Generation” framework, validated by a Kulangsu smart tourism case. First, a five-stage spatiotemporal-semantic alignment pipeline integrates GPS, social media, POI, and weather data. Second, to address LLM limitations, we introduce a knowledge-enhanced framework combining geospatially-aware LoRA fine-tuning and a hierarchical Retrieval-Augmented Generation (RAG) mechanism. This approach reduces hallucinations by 15% and boosts geographic reasoning accuracy, achieving state-of-the-art intent recognition (15% gain) and landmark identification (72% accuracy). Third, we design a reprogramming strategy with geo-semantic constraints for generating spatially plausible, personalized trajectories in data-scarce settings, showing 89% similarity to real-world data. Supported by the Kulangsu Smart Tourism System, our method provides a scalable paradigm for spatiotemporal modeling and synthetic trajectory generation, benefiting personalized services and decision-making.

Forecasting regional COVID-19 hospitalisation in England using ordinal machine learning method.

An IoT-enabled AI system for real-time crop prediction using soil and weather data in precision agriculture

Improving soybean growth prediction: Harnessing vision transformer and long short-term memory models with UAS imagery and weather data

Temporal structure-preserving transformer for industrial load forecasting.

Nowadays, proper determination of the thermal efficiency of new building envelope solutions focusing on energy efficiency is vital for effective energy management. Determining the thermal efficiency of thermal storage (Trombe) wall modified with phase change material (TWPCM) is challenging, and its inaccurate estimation may lead to unnecessary waste of resources, failures, and financial losses. The aim of this work is to develop a reliable deep learning prediction model to determine the thermal efficiency of the TWPCM. The performance of the proposed Convolutional Neural Network combined with Long Short-Term Memory (CNN + LSTM) was compared with seven other developed machine learning models. Eight input variables were used: outdoor air-dry bulb temperature, relative humidity, wind speed, wind direction, total solar radiation intensity on the horizontal surface, direct solar radiation intensity on the horizontal surface, and time of day and year. Input variables from the last 240 h were input data for the models. A model consisting of 4 LSTM layers, 5 CNN layers joined together with fully connected layers was used. The models were trained, tested, and validated in the data set from real-world energy performance data. The CNN + LSTM model was found to outperform the other models with the highest determination coefficient (0.99891) and the lowest mean absolute error (0.19188 W/m2) and root mean square error (0.26324 W/m2). The results show that the proposed deep learning model (1) effectively predicts the thermal behavior of TWPCMs by taking into account heat storage capacity of phase change materials, (2) has very good generalization ability verified on a new data set, (3) enables comparison of results with other building envelopes under typical conditions, e.g. in relation to a Typical Meteorological Year (TMY), by forecasting using weather data from a TMY, and (4) enables a reduction in the time required for direct testing, thus reducing the cost of the analysis.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-26886-1.

Vitamin D status is influenced by genetic and environmental factors—primarily sun exposure. Using satellite weather data, we estimated an ambient UVB dose for each participant based on residential address and date of sampling. We conducted genome-wide tests in 338,977 UK Biobank White British participants, adjusted for age, sex, supplements, UVB dose, and 10 principal components to account for population structure. We applied three models to test for genetic effects: marginal only, main and interaction, and joint effects. We identified 307 variants associated with standardised log-transformed 25-hydroxyvitamin D (25OHD) concentration, 162 of which were not previously identified in GWAS. We identify an increase in SNP-heritability by increasing ambient UVB exposure quintiles (h2Q1 = 8.48% vs. h2Q5 = 15.56%). Downstream annotation implicated genes in the 25OHD pathway, including the circadian regulator, BMAL1. This and further findings suggest that vitamin D status and circadian rhythm may be entangled and that vitamin D metabolites may have a role as mediators of seasonal physiological fluctuations, including metabolism, and in turn explain the established associations with lipid metabolism pathways.

Overhead transmission lines (TLs) are typically rated using static methods that assume conservative environmental conditions, such as high ambient temperatures and low wind speeds. These conservative ratings often result in underutilization of the line’s full ampacity, as extreme conditions occur only intermittently. Dynamic line rating (DLR) addresses this issue by utilizing real-time weather and grid load data to dynamically calculate the available ampacity, considering favorable environmental conditions. The DLR continuously updates ampacity calculations, allowing operators to access unused capacity when conditions are more favorable. This approach improves transmission efficiency and enhances grid reliability, ensuring a more adaptive and resilient power system. By integrating real-time weather data, DLR provides a more accurate representation of a TL’s actual capacity than static ratings. Both IEEE and CIGRÉ standards employ the heat balance equation to calculate ampacity, accounting for heat absorption and dissipation. This paper compares the methodologies outlined in IEEE Std 738 and CIGRÉ TB 601, focusing on their approaches to calculating conductor temperature and ampacity. The study examines the impact of different modeling approaches on ampacity calculations and the performance of TLs. The analysis shows significant differences between the two methods. In summer, the IEEE method increases ampacity by 47.2% compared to static line rating (SLR), while the CIGRÉ method increases it by 46.5%. In winter, the IEEE method shows a 36.9% increase, and CIGRÉ shows a 38.6% increase. These results demonstrate the potential of DLR to optimize transmission capacity and improve grid performance by adapting to real-time environmental conditions. Seasonal variations further highlight how factors like temperature and wind speed impact ampacity, reinforcing the value of DLR systems for maximizing TL efficiency year-round. Cite this article as: Z. Zia and C. F. Kumru, "Assessing ampacity performance through dynamic line rating: A comparison of IEEE std 738 and CIGRÉ TB 601," Electrica, 2025, 25, 0109, doi: 10.5152/electrica.2025.25109.

Traditional data-driven approaches emphasize input–output correlations and neglect dependencies among inputs, risking missed insights into key drivers of energy performance. Consequently, approaches that transcend correlation-centric analysis are warranted. Within this context, causal inference, which accounts for both statistical associations and temporal cause–effect relations, constitutes a promising direction. However, researchers cannot feasibly specify all causal relations relying solely on domain knowledge. Causal discovery is a data-driven methodology for analyzing causal relationships among variables, providing not only measures of association but also information on causal directionality. The authors employ two causal discovery algorithms—PC (Peter-Clark) and FCI (Fast Causal Inference)—on weather data. The discovered causal structures are compared, and two validation approaches are introduced to evaluate their statistical reliability; the authors also build on the identified causal structure to analyze the resulting causal pathways. The results show that both algorithms provide insights into causal relationships among variables, and the proposed validation approaches help establish the statistical reliability of the discovered structures. Moreover, the analysis of causal pathways indicates that causal effects can be identified and estimated with reliability.

Abstract The Interstellar Mapping and Acceleration Probe (IMAP) is a NASA heliophysics science mission that provides new coordinated and comprehensive observations of the inner and outer heliosphere. The IMAP observatory orbits at the Sun-Earth L1 Lagrange point, which is an ideal location for observing the space weather conditions upstream of Earth. Thus, in addition to providing new and groundbreaking heliophysics science observations, five in-situ instruments on IMAP make measurements that are critical for advancing space weather research and operational forecasting. These measurements are continuously telemetered in near real-time as part of the IMAP Active Link for Real-Time (I-ALiRT) space weather data system. I-ALiRT is based on the Real-Time Solar Wind (RTSW) data system from the NASA Advanced Composition Explorer (ACE) mission and provides similar space weather data products at enhanced cadences as well as additional new data products. This paper describes the I-ALiRT instruments and measurements, real-time data flow architecture, and publicly available space weather data products.

Context: Extreme heat events are a serious concern for the conservation management of wildlife. In flying-foxes (Pteropus spp.), exposure to air temperatures (Ta) > 42° C can result in mass mortality, sometimes at catastrophic scales. To mitigate the worst of the impacts on flying-foxes, sprinklers are increasingly being deployed in roosts to cool flying-fox roosting habitat and/or the flying-foxes directly. However, while anecdotal reports suggest positive outcomes from these interventions, the effects of sprinklers on microclimatic conditions and flying-fox thermoregulatory responses have not been studied empirically. Aims: We aimed to test experimentally the impacts of sprinklers on microclimatic conditions of a flying-fox roost and so provide a much-needed evidence base for flying-fox heat stress mitigation. Methods: We used an automated split-system sprinkler setup in the understory of a permanently occupied grey-headed flying-fox (P. poliocephalus) roost site in Campbelltown, NSW. High-resolution weather data were systematically collected at sprinkler and control areas throughout the roost across a range of daily meteorological conditions between the austral summers of 2020-21 and 2023-24, including during an extreme heat event that resulted in mortality of flying-foxes across the region. Key results: Our results showed that on days ≥ 35° C, sprinklers cooled Ta by 1.5° C and increased dewpoint (Tdew) by 1.2° C, on average, and during an extreme heat event, the sprinklers kept the local microclimatic conditions within known thermoregulatory tolerances of flying-foxes, in contrast to the lethal conditions that were observed elsewhere in the roost. The effects of the sprinklers were highly localised to the treatment area both horizontally and vertically; and the timing, and duration of effect differed for Ta and Tdew. Conclusions: This study has made important progress in identifying the impacts of understory-based sprinklers on roost microclimatic conditions, and the results are promising for the utility of sprinklers for flying-fox heat stress mitigation. Implications: To understand where, when, and how sprinklers can be used to achieve net-positive outcomes for heat stressed flying-foxes, it remains necessary to determine directly how this management intervention influences flying-fox thermoregulatory responses.

Accurate estimation of reference evapotranspiration (ETo) is essential for effective irrigation scheduling, water-resource management, and crop-climate modelling in tropical regions where weather data are often limited. This study presents the development and validation of a Flutter-based Android application capable of computing near-real-time ETo using the FAO-56 Penman–Monteith (FAO-56 PM) model. The app integrates meteorological data from the OpenWeatherMap API and utilises geolocation for site-specific retrieval of temperature, humidity, wind speed, and solar radiation. Validation was performed against the NASA POWER database for 11 consecutive days (4–14 April 2024) using 24-hour averaged datasets to remove diurnal bias. Results showed close agreement between app- and NASA-derived parameters: temperature (mean bias = +0.35 °C), relative humidity (-0.9 %) wind speed (+0.01 m s -1) and ETo (+0.03 mm day-1) Statistical evaluation indicated a strong correlation (r = 0.93; R² = 0.86) and low root mean square error (RMSE = 0.05 mm day-1). These results confirm the app’s reliability and computational fidelity. Comparative analysis with existing tools such as EVAPO, EvapoCalc, and pyfao56 revealed comparable accuracy while offering the advantage of real-time mobile accessibility. The application bridges a technological gap by providing a cost-effective, portable, and adaptive system for field-level ETo estimation, particularly valuable for smallholder farmers and irrigation managers in data-scarce environments.

Smart Agriculture (SA) combines cutting edge technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and real-time sensing systems with traditional farming practices to enhance productivity, optimize resource use, and support environmental sustainability. A key aspect of SA is the continuous monitoring of field conditions, particularly Soil Moisture (SM), which plays a crucial role in crop growth and water management. Accurate forecasting of SM allows farmers to make timely irrigation decisions, improve field management, and conserve water. To support this, recent studies have increasingly adopted soil sensors, local weather data, and AI-based data-driven models for SM forecasting. In the literature, most existing review articles lack a structured framework and often overlook recent advancements, including privacy-preserving Federated Learning (FL), Transfer Learning (TL), and the integration of Large Language Models (LLMs). To address this gap, this paper proposes a novel taxonomy for SM forecasting and presents a comprehensive review of existing approaches, including traditional machine learning, deep learning, and hybrid models. Using the PRISMA methodology, we reviewed over 189 papers and selected 68 peer-reviewed studies published between 2017 and 2025. These studies are analyzed based on sensor types, input features, AI techniques, data durations, and evaluation metrics. Six guiding research questions were developed to shape the review and inform the taxonomy. Finally, this work identifies promising research directions, such as the application of TinyML for edge deployment, explainable AI for improved transparency, and privacy-aware model training. This review aims to provide researchers and practitioners with valuable insights for building accurate, scalable, and trustworthy SM forecasting systems to advance SA.

Abstract Late blight caused by Phytophthora infestans , poses a significant threat to potato cultivation globally. In A Limia region located in the North-West of Spain, climatic conditions are favorable for late blight development, but disease-forecasting systems are absent, resulting in excessive use of fungicides each growing season. The present study aimed to predict the onset of late blight infections in the potato crop by analyzing weather factors, airborne P. infestans , and the resistance of potato cultivars to late blight over three growing seasons. The study assessed the accuracy of existing weather-based models (Indice Potenziale Infettivo [IPI], Infection Pressure [IP], and hours of risk of sporulation [HOSPO90]) and developed a new model, combining aerobiological and weather data for predicting late blight occurrence. Notably, the 2022 growing season displayed lower sporangia concentrations and late blight severity below 5% across all cultivars. Cluster analysis based on different epidemiological parameters highlighted that the growing season significantly influenced the epidemic response. The weather-based models predicted an Aerobiological Risk Period [ARP], which defines the period with significant sporangial levels to initiate infection. Some risk periods according to ARP were identified by the IPI and HOSPO90 models. Furthermore, increased sporangia concentrations were observed before the onset of first symptoms in all potato cultivars. A logarithmic aerobiological-based model was subsequently developed to predict the timing of 5% severity with 92% accuracy. This approach enhances late blight outbreak predictions by integrating aerobiological metrics into disease management strategies, reducing unnecessary fungicide application and fostering more sustainable agricultural practices.

Abstract This study reports development of ambient meteorological year (AMY) weather data for example locations in fifteen US climate zones for simulating buildings that are heated and cooled entirely by ambient energy. Indoor comfort in such buildings is more directly tied to weather, specifically to solar radiation and outdoor temperature, than in conventional buildings, which simply use more auxiliary energy to maintain comfort during extreme weather. Therefore, typical meteorological year (TMY) weather data, which is standard for conventional buildings, produces inconsistent results for ambient-conditioned buildings. More importantly, TMY fails to predict significantly uncomfortable indoor temperature that occurs during extreme seasons in some climates. Extreme seasons were identified by simulating ambient-conditioned buildings designed to remain comfortable during TMY2023 weather from all years (1998-2023) of real weather from which TMY2023 was derived. AMY files were assembled from the summer and winter seasons that produced the most extreme indoor temperatures. Indoor temperature for AMY weather matched within 0.2°C those predicted from the extreme years. Simulation with these new AMY files reduces the effort and time required to design ambient-conditioned buildings, while providing reasonably accurate indoor temperature predictions.

Climate change poses significant risks to the insurance industry, particularly through increased frequency and severity of weather events. In this paper, we examine the extent to which temperature, wind speed, and rainfall influence the stock prices of companies listed on the UK stock market. By using data of 1. Andrews Sykes Group plc. (ANSY) 2. Aviva plc (AVIVA) 3. Legal & General Group plc (LGEN) 4. Prudential plc (PRU) 5. Direct Line Insurance Group plc (DLGD) and conducting the multiple linear regression model and random forest model, the findings, illustrated through scatterplots, indicate a weak relationship and correlation between the dependent (stock prices) and independent variables (weather data), as shown by the P-values and R-values.

Soil fertility in sub-Saharan Africa is increasingly affected by climatic stressors that accelerate nutrient loss and threaten farming sustainability. In Adamawa State, Nigeria, where many people depend on agriculture, rising temperatures, inconsistent rainfall, and frequent floods make soil degradation worse. This study assesses the vulnerability of soil nutrients to climate change impacts by using soil sampling, lab analysis, and long-term weather data. Composite soil samples were taken at two depths (0–30 cm and 30–60 cm). Rainfall, temperature, and relative humidity records from 2001 to 2021 were analyzed along with rainfall erosivity data and recorded flood events. The study adopted Pearson correlation and paired t-tests for its statistical analyses, the findings revealed significant negative correlations between rainfall and phosphorus (r = –0.68), potassium (r = –0.54), and organic matter (r = –0.63), particularly in surface soils. Flood events significantly reduced phosphorus (p < 0.001), potassium (p = 0.003), and microbial biomass carbon (p = 0.004). Rainfall erosivity exhibited marked interannual variability, peaking in 2008, 2012, 2014, 2018, 2019, and 2021. The study concludes that surface soils in Adamawa are highly vulnerable to climatic stressors and recommends the adoption of climate-smart soil management and erosion control strategies to enhance resilience and sustain agricultural productivity.

In the context of a lack of educational tools for learning space technologies and satellite development, CanSats were created as an educational tool. This article proposes the mechanical, electrical and software design of a CanSat with an autogyro descent system where the novelty is the implementation of AWS IoT services and Node-RED to store, manipulate and display in real-time the collected weather data. This picosatellite design is capable of safeguarding the integrity of the CanSat's payload where a chicken egg will be placed during the flight and landing phases. Often, other designs of CanSats use local servers implemented on the computer or laptop of the team for storage and display of the data. This makes it more difficult to share the information to people without access to the computer where the server was specifically deployed. The use of AWS services for the Internet of Things is very useful in sharing and displaying the collected information to the public interested in the collected weather data. One of the AWS services implemented allows data subscription through Gmail. The findings made in this paper hold implications for applications involving the transportation and safe landing of delicate payloads in space exploration missions. As a result of the implementation of this design, the separation between the secondary and primary load was successfully achieved and the weather data was transmitted.

Wildfires present a significant threat to ecosystems, property, and human life in Kazakhstan. Understanding fire hazards is essential for effective management and mitigation of these risks. This study develops a comprehensive fire hazard index for Kazakhstan by integrating static, long-term landscape factors with dynamic, real-time weather and vegetation conditions. The static component employs a machine learning approach, specifically the Random Forest algorithm, trained on a dataset that includes topographic variables derived from the SRTM DEM, land cover classifications from MODIS Terra/Aqua LULC products, and historical fire occurrence data from NASA FIRMS. This model quantifies the inherent fire susceptibility of various landscapes based on these enduring characteristics. The dynamic component captures short-term fluctuations in fire risk by incorporating satellite-derived vegetation information and meteorological observations. The MODIS-derived Normalized Difference Vegetation Index (NDVI) serves as a proxy for fuel availability and moisture content. Spatially interpolated weather data such as temperature, humidity, wind speed, and precipitation provide the necessary meteorological context. The dynamic index is calculated using a modified Canadian Fire Weather Index (FWI) system, specifically adapted to account for the influence of live fuel moisture, as indicated by NDVI, on fire ignition and spread dynamics. The final fire risk index is created by additively combining the static and dynamic components, offering a spatiotemporal perspective on fire risk. This integrated approach allows for the assessment of both the underlying susceptibility of a landscape to fire and the immediate effects of weather and vegetation conditions. The resulting high-resolution fire hazard maps are intended to inform fire management decisions, optimize resource allocation for fire prevention and suppression efforts, and support targeted interventions in high-risk areas. This research underscores the value of combining machine learning techniques with remotely sensed data for enhanced fire risk assessment in Kazakhstan, facilitating more proactive and effective fire management strategies.

Wind is a critical meteorological factor for crop production, yet the effect of its long-term changes on crop yield has not been adequately discussed. Here, we assess the effect of wind on maize yield and production in China with statistical modeling of weather and crop data for 1980−2017. The analysis, applied in different regions and scales, consistently reveals that maize yield increases by ~9.4% for every 1 m·s−1 decrease in growing-season wind speed anomaly after controlling for temperature and precipitation. Three decades of decreasing wind speed between 1980 and 2010 have enhanced China’s maize production by 101.2 Mt, fully offsetting the loss due to warming temperatures. However, this compensatory effect has rapidly diminished in the last decade due to the reversal of wind trends, with expanding areas experiencing both increasing wind speeds and warming temperatures. The wind effect is substantially underestimated by crop models, exposing gaps in our current evaluation of climate risks to food security.