Grapevine (Vitis vinifera L.) is an economically important fruit crop, but its production is severely affected by anthracnose caused by Colletotrichum gloeosporioides, commonly known as bird’s eye spot. The present investigation was conducted at the Horticultural Farm, University of Agricultural Sciences, Raichur, during the Kharif seasons of 2024 and 2025 to study disease progression and to develop a prediction model for grape anthracnose in the susceptible cultivar Thompson Seedless. Disease severity was recorded at weekly intervals using a 0-4 rating scale and expressed as Per cent Disease Index (PDI). Simultaneously, weekly weather data were collected from the Main Agricultural Research Station, Raichur. Anthracnose appeared during the 27th and 24th standard meteorological weeks in 2024 and 2025, respectively, and gradually increased to 100 per cent severity by the end of the season. Observed PDI ranged from 8.50 to 100.00 per cent in 2024 and from 7.13 to 100.00 per cent in 2025. A first-order autoregressive model was developed to predict disease progression, which showed close agreement between observed and predicted PDI values, particularly during the mid-season period. The developed models exhibited high autocorrelation coefficients (R = 0.953 in 2024 and R = 0.891 in 2025), indicating a strong temporal relationship in disease development. The study demonstrates that an autoregressive approach can effectively describe the progression pattern of grape anthracnose under field conditions.

Little is known about the association between direct pollen exposure and cognitive performance. The aim of our study is to investigate the effect of pollen exposure on performance in the Finnish matriculation examination. The study was conducted among students who participated in the national high school matriculation examinations in the metropolitan area of Helsinki and Turku in southern Finland between 2006 and 2020. Daily regional pollen counts of alder and hazel were monitored throughout the study period as part of the Finnish pollen monitoring network. Extensive data on matriculation examination results were retrieved from Statistics Finland, and air pollution and weather data from the Finnish Meteorological Institute. A fixed effect regression analysis was used to identify the effect of pollen exposure (as independent variables) on matriculation examination results (as dependent variable) controlling for student-semester fixed effects, pollutants and precipitation. The regression coefficients indicated that on average an increase of 10 pollen grains in alder and hazel reduced the matriculation examination score by 0.0034 (p<0.01) and 0.0144 (p<0.05) standard deviations (SDs), respectively. Increasing pollen exposure per additional unit (an increase of 10 pollen grains) especially dropped examination scores in mathematical subjects among males (alder -0.0118 (p<0.001) and hazel -0.0328 (p<0.05) SDs). The association between alder pollen exposure (low, moderate and abundant) and examination scores was inversely U-shaped. Exposure to pollen can hinder a student's performance in the matriculation exam, which strongly determines the future opportunities and emphasises early initiation of medication.

Rainfall and temperature influence effectiveness of on-site sanitation intervention against E. coli contamination in Bangladeshi households.

The prediction of crop yields is influenced by various factors such as weather conditions, agronomic practices, and management strategies. Accurately predicting oil palm yield is crucial for sustainable production, as it plays a significant role in global food security. Challenges such as climate change and nutrient deficiencies have adversely affected yields, highlighting the necessity for a specialized web application tailored to the oil palm industry. This study presents a machine-learning-based web application that utilizes a deep learning model to estimate oil palm yields by integrating key parameters, including weather, agronomy, and satellite data. The application features a user-friendly interface and a dashboard for comparing predicted and actual yields, enhancing user engagement and facilitating collaboration among stakeholders. By deploying this tool on the cloud, plantation managers can make informed decisions early in the yield prediction process, ultimately improving plantation management and profitability. This web application is designed to provide valuable insights to stakeholders, contributing to effective decision-making in the oil palm sector.

To improve sustainability in power generation, this study presents a thorough data-driven method for maximizing renewable energy sources. It employs measures like capacity utilization factor (CUF) and efficiency to evaluate the performance of solar and wind energy using historical weather and energy-generating data. The study offers practical suggestions for improving renewable energy systems, such as weather-energy correlation analysis and machine learning-based forecasting models. In addition, a comparative analysis is carried out to ascertain which energy source is better, and useful real-world data is provided, including a summary of all India’s total renewable energy generation (excluding large hydro) for June 2023 and a performance comparison year over year. A useful, data-driven approach for enhancing renewable energy is provided by this work, which advances the topic of sustainable energy.

The substantial role of weather data in consumer spending prediction: A robust machine learning assessment

Increasing shares of wind and solar generation, together with rising electricity demand, introduce growing uncertainty into power system operations. Accurate day-ahead forecasts of electricity demand and renewable generation are essential for system operators to coordinate electricity markets and maintain reliability at low cost. Here, we show that forecasting based on joint probability distributions of demand and renewable supply can substantially improve system-level forecasting performance using publicly available weather data. We develop multiple day-ahead forecasting models that combine machine learning methods to identify relevant weather variables with probabilistic approaches to quantify forecast uncertainty, and we evaluate these models using proper scoring rules. Applied to the three zones of the California Independent System Operator, the best-performing model improves forecast skill by 25% relative to current benchmarks. We further show that forecasts based on joint probability distributions enable a more effective allocation of operating reserves than conventional deterministic approaches, highlighting the potential of probabilistic machine learning to enhance market efficiency and grid stability in increasingly decarbonized power systems.

Modified Lightweight Multimodal Transformer for Rapid Disaster Detection Using Social Media and Weather Data

Agriculture is a very crucial sector in the Nigerian economy, it provides food and income to millions of residents. However, the increasing unpredictability of climate regimes which is a direct effect of anthropogenic climate change has a highly negative impact on the total agricultural output in terms of crop yield. This paper aims at creating a predictive model based on weather, which maximizes the productivity of cassava and cocoa in Ondo State, Nigeria. Kaggle repository data on temporal crop yield were collected from 1991 to 2020 using meteorological measures provided by NASA POWER, this weather data was used to gain a calibration of forecasting models. XGBoost (Gradient Boosting) machine learning model was used to develop salient climatic factors like precipitation, temperature, and soil moisture. The findings show that the cassava model had moderate predictability (R² = 0.61) and the cocoa model had less predictability (R² = 0.55). The system seeks to provide the agricultural sector stakeholders with actionable knowledge, which will empower the stakeholders to make evidence-based decisions on how to advance agricultural production and strengthen agric-food systems during climatic disturbances. It follows that the developed predictive system have a significant contribution to sustainable agricultural development and food security enhancement across the region.

CIBSE weather files are currently used by the building industry as the standard input data for building performance assessment for the purpose of regulatory compliance in the UK. In this study, the state-of-the-art CIBSE weather files are created with four major improvements incorporated, namely, (1) the enhanced representation of the UK climate through the creation of discriminative climate zones; (2) the latest climate change signals from the UK Climate Projection 2018 (UKCP18); (3) the satellite based solar radiation data from CAMS (Copernicus Atmosphere Monitoring Service) data repository; (4) the up-to-date observation record from 1994 to 2023. The methodology for creating the latest CIBSE weather files is elaborated in detail to enhance the transparency of the new weather data. Evaluated using a simulation case study, the new weather files demonstrate spatial and temporal coherency. The new future weather files enable robust building performance assessment against future climate conditions under different scenarios and will play an important role in designing climate-resilient buildings and delivering a net zero built environment. Practical applications As per the principle of “garbage in, garbage out”, weather data plays an instrumental role in streamlining building design to achieve both energy efficiency and thermal comfort. In this study, we present the methodology for the creation of the state-of-the-art CIBSE weather files. The new CIBSE weather files not only employ the update-to-date observation and projection data, but are also grounded on a total of 28 granular climate zones to account for diverse climate characteristics and eliminate the ambiguity with weather data selection. The new files will lay a solid data foundation for future-proofing building design in the UK.

Predicting accurate electricity consumption on electrical distribution network is essential for efficient energy management, particularly in institutional settings where demand fluctuates due to population growth and weather variations. Traditional prediction models often lack real-time accessibility. This study presents a browser-based simulation model for monthly electricity consumption using historical data and dynamic weather inputs, implemented with HTML, CSS, and JavaScript. The model generates forecasts by applying statistically plausible variations (±5%) to historical consumption patterns, integrated with simulated weather data for real-time scenario testing. Compared to complex machine learning approaches, this lightweight solution offers enhanced scalability, remote accessibility, and instant updates without server dependencies which makes it more applicable for smart grid systems and utility management. Results demonstrate its utility as a practical tool for preliminary energy trend analysis, supporting integration with cloud-based data sources. This research will contribute to accessible energy forecasting tools and provides a practical tool for optimizing electricity consumption on electrical distribution network in institutional environments and for institutional planning.

Disaggregating end-use electricity consumption from aggregate meter data remains a fundamental challenge in non-intrusive load monitoring, particularly in smart buildings where heating, ventilation, and air-conditioning systems dominate demand and direct sub-metering is often unavailable. Contextual variables such as weather and calendar information provide valuable explanatory signals, but in low-frequency settings, these drivers are typically insufficient to fully characterise building operation. As a result, attribution strategies that implicitly assume complete explainability can lead to unstable driver contributions and reduced physical interpretability when building behaviour is non-stationary or partially unobserved. This paper introduces MD-ADD, a multi-driver automatic dependency disaggregation framework designed for low-frequency smart meter data in commercial and public buildings. The framework supports joint attribution of multiple contextual drivers. It explicitly represents unexplained energy as a meaningful component of the decomposition. It combines robust baseline estimation, leakage-resistant out-of-fold contextual modelling, conservative driver attribution without hard mass-balance constraints, and uncertainty quantification using block bootstrap resampling. A consistency mechanism is included to restrict driver attributions to temporal scales compatible with their expected physical influence. The framework is evaluated on the ADRENALIN Load Disaggregation Challenge dataset, which contains multi-resolution electricity and weather data from commercial and public buildings, using normalized mean absolute error alongside stability and residual-structure diagnostics. Rather than optimising solely for pointwise accuracy, the proposed formulation emphasises robustness, interpretability, and diagnostic transparency, making it suitable for decision-support and analytical workflows under realistic low-frequency monitoring conditions.

Electrochemical CO 2 reduction reaction (eCO 2 RR) offers a promising pathway toward a circular economy by converting CO 2 into value-added products such as formate, carbon monoxide, and ethanol. Among the products, formate has gained particular attention as a versatile C 1 building block. Pilot-scale demonstrations have reached kilogram-scale CO 2 conversion per day using large-area electrodes. Thus, the sustainability of this highly electricity-intensive process critically depends on its integration into the renewable energy landscape. Foremost battery storage is required to ensure continuous operation of eCO 2 RR while utilizing solar power and addressing its variability. In this study, we developed and applied a tailored photovoltaic (PV) system with battery storage to evaluate long-term renewable energy supply for eCO 2 RR at different scales (10 cm 2 –300 cm 2 Sn–GDE setups), using the UFZ location in Leipzig, Germany, as a reference site. For developing such a stand-alone power supply at the reference site, PV power generation data obtained using the Renewable Spatial–Temporal Electricity Production (ReSTEP) simulation model, which is based on real weather data, were combined with experimentally derived energy demands of the eCO 2 RR setups at different scales. As a result, the required number of PV panels and batteries for reliable year-round operation was determined. The results show that the number of solar modules scales proportionally with the electrode size, while sufficient battery storage is essential to buffer up to three consecutive days without sunlight and maintain safe discharge limits. For the 100 cm 2 setup, additional off-grid simulations demonstrate that increasing the battery capacity improves both system reliability and battery lifespan. Overall, this study demonstrates that tailored PV systems with battery buffering can enable sustainable operation of eCO 2 RR from laboratory to pilot scales, highlighting a practical route for integrating this technology into future electrobiorefineries and advancing its readiness toward industrial deployment.

Heatwaves are among India’s most critical climate-related hazards, significantly affecting public health, agriculture, energy systems, and urban infrastructure. Accurate and early prediction of heatwave events is therefore essential for reducing mortality, supporting disaster preparedness, and improving community resilience. This study investigates the effectiveness of advanced machine learning and deep learning models for forecasting heatwave occurrences using a comprehensive multicity daily weather dataset spanning 25 years (2000–2024). Data from ten major Indian metropolitan regions were used, incorporating key meteorological variables such as maximum and minimum temperature, apparent temperature, rainfall, humidity, wind speed, wind gusts, and wind direction. To capture temporal dependencies in weather patterns, a 28-day sliding window approach was applied. Data pre-processing included MinMax normalization, categorical feature embeddings, and SMOTE-based oversampling to address class imbalance. Two deep learning architectures were evaluated: a Transformer Encoder combined with BiLSTM, and a CNN-BiLSTM model. Their performance was compared with traditional machine learning methods, including XGBoost, LightGBM, SVM, Random Forest, and feed-forward neural networks. Results show that the Transformer-BiLSTM model achieved the highest accuracy (91.26%), outperforming the CNN-BiLSTM model (87.97%) and all classical approaches. The findings confirm that deep temporal sequence models provide superior performance for heatwave prediction. This study highlights the potential of deep learning-driven forecasting systems for strengthening early warning mechanisms and climate adaptation strategies in India.

Farmers are dealing with a lot of trouble from this unpredictable weather that just keeps coming at them hard these days. It leads to huge crop losses for millions of them. This affects people the most who do not get timely climate info that is practical and easy to grasp. The thing is, this project introduces a Smart Weather Alert System. It aims to deliver real time weather details along with early warnings through calls and Messages. It also offers AI-based advice for farming in a straightforward way that anyone can access. The setup pulls in live weather data straight from the OpenWeather API. It combines that with a trigger system you can adjust to spot key issues like heat stress or heavy rain and strong winds. Once those limits get hit, the whole thing kicks in automatically. It sends out advisories in multiple languages that fit the local context. A big language model creates those through an AI helper and a chat interface that talks back. This really helps farmers with low reading skills make the correct moves just when they need to. The map feature lets you pick your spot without any hassle. Meanwhile, a news section updates itself to cover the latest in farming tech and research. Early tests point out how this boosts what people know about weather. It speeds up choices they have to make. And it builds better readiness for those risks from bad conditions. In the end, this setup provides something that scales well and costs little. It stays friendly for farmers too. All in all, it helps build stronger resistance to climate changes in farming areas.

Agricultural research increasingly relies on data-driven approaches for crop yield prediction that complement more established crop growth models, including machine learning techniques. However, these approaches rely on large training datasets. Here, we present the Crop Yields, Climate, Soils, and Satellites (CYCleSS) dataset, a large-scale crop yield dataset derived from precision yield data for 934 fields across England on which a variety of crops are grown. In addition, the data also contains satellite-derived remote sensing data, weather data, and data on soil type, all aligned at a grid resolution of 10 km. Weather data is available at a daily temporal resolution, satellite data at 5-day resolution, while crop yield data is available at yearly resolution. This effort has been made possible through careful anonymisation of the yield data while preserving the alignment with remote sensing, weather, and soil data. This data will be useful both to train machine learning models of yield prediction as well as to parameterize mechanistic crop growth models. Furthermore, the anonymisation procedure itself will be of interest to the research community, as it represents a solution to a common problem on the interface of agricultural research and farming practice.

Globally agricultural emissions contribute 10-12% of all anthropogenic greenhouse gas emissions. Ruminant livestock account for a considerable proportion of these emissions, with cattle grazing in tropical and sub-tropical regions representing a major source. Enteric methane production has climatic implications and represents a loss of energy from the animal that could be directed towards growth. This study focuses on cattle grazing tropical and subtropical rangelands in Australia, aiming to identify the factors that influence recording of methane emissions in these environments. Short term methane production was recorded from 453 mixed breed cattle across 4 trial sites utilizing GreenFeed units (GFU) (C-Lock, USA). Data was generated and combined from several sources including GFU, weather data, forage samples and estimated forage biomass. Using a linear mixed model the significance and variance of each factor on methane production was evaluated. Factors identified as explaining most of the variation for methane production included GFU (P < 0.001), total visitation (P < 0.001), daily visitation (P < 0.001), trial (P < 0.001), hour of visitation (P = 0.03), air temperature (P < 0.001), initial weight (P < 0.001), visit duration (P < 0.001) and forage biomass (P < 0.001). Four factors were tested and found to be not significant in explaining variation in methane production, including Bos taurus indicus (BI) content (P = 0.99), sex (P = 0.06), rain (P = 0.07) and temperature heat index (P = 0.2). A moderate coefficient of variation (CV) of 23.2% was observed in the raw observations of methane production. This study highlights important factors to consider when utilising GFU for methane measurement of grazing beef cattle. This study provides a foundation for future studies aiming to quantifying individual animal methane emissions for use in genetic evaluation programs, to reduce industry emissions and improve the sustainability of beef production in tropical and sub-tropical regions.

Soybean growth is determined by the interaction of genetic, environmental, and management factors. In the context of future climate and climate extremes, understanding genotype by environment interaction (GxE) will be crucial for selecting resilient breeding lines and optimizing management practices to minimize stress. This requires an in depth elucidation of stressful weather conditions and differing temporal responses of genotypes to those conditions. In field studies, however, the environment is often treated as a static factor, and the specific effects of weather variability on crop growth remain poorly understood. Here, we present a longitudinal dataset comprising 17,247 high-resolution RGB images of soybean breeding lines collected throughout eight years in Eschikon, Switzerland. Top-of-canopy images were acquired throughout the entire growing seasons and complemented by hourly weather data, enabling a comprehensive analysis of soybean growth dynamics under varying field conditions. High spatio-temporal image resolution allows detailed analysis of growth dynamics and GxE, supporting identification of stress-tolerant genotypes to improve yield prediction and yield stability.

Abstract Background&amp;#xD;Climate change causes an increased number of hot days, especially in urban areas. This can cause heat stress in the human body. Despite being a vulnerable group, data on thermoregulation and sweat loss in heat for young children are missing. Therefore, model calculations can serve as an alternative. &amp;#xD;Objective&amp;#xD;This small-scale case study compares climate data from the study site Bochum, eligible as an urban heat island in Central European temperate climate zone, to available climate data of physiological model calculations of young children’s thermoregulation to assess transferability. &amp;#xD;Methods&amp;#xD;The climate station in Bochum provides 13 weather variables. For the comparison, data of the summers 2020-‘23 were used. Literature was screened for physiological model calculations for children aged 3-6 years in real-life scenarios. &amp;#xD;Results&amp;#xD;The annual temperature in Bochum has risen within the past decades, nevertheless ambient temperature of 30 °C has rarely been exceeded in 2020-‘23. Only one study using real weather data from Tokyo, Japan was suitable for displaying a model of a young child under heat stress in a real-life scenario. The maximum ambient temperature of the model calculation (36.5 °C) was only surpassed once in Bochum during the study period. &amp;#xD;Conclusion&amp;#xD;Thus, the model calculations of Tokyo’s climate scenario are not fully transferable to Bochum as a model city in Central European temperate climate zone. Consequently, further research is needed. This analysis can nevertheless serve as an impetus to initiate action on this growing public health problem in young children in Central Europe.

Lack of secure shelter is a contributing factor for healthcare utilization. Weather extremes, specifically extreme cold, is common in Canada. Because of the safety-net function of emergency departments (EDs), their use as shelters or warming centers is common in cold weather. Timely diversion resources may help mitigate ED crowding. We conducted a cohort study using linked administrative data between April 2018 and March 2024 of all adult (> 17years old) patients presenting to any EDs in the greater Edmonton area with an International Classification of Diseases, 10th revision diagnostic code with houselessness, unspecified. We obtained weather data from environment Canada for each day. The primary outcome was ED visits associated with houselessness based on mean daily temperature. Between April 1st, 2018, and March 31st, 2024, there were 2,433,282 total ED visits recorded in the study area. There were 13,421 unique patients with ED visits associated with houselessness; more were male, and the median age was 39 (31, 49). The median visits per year were 5.2 (2.1, 12.4) for patients with a visit associated with houselessness compared to 1.2 (1.0, 3.3) for patients with stable housing. Proportion of ED visits associated with houselessness as a function of mean daily temperature revealed higher ED utilization with colder temperatures (< 0.0001). There was a 21% increase in expected ED visits associated with houselessness during periods of extreme cold. Patients experiencing houselessness access EDs much more frequently than patients with stable housing. There was a much higher number of unique patients experiencing houselessness than expected. Utilization increases during times of extreme cold weather; and interventions to address these capacity issues are urgently needed.