Weather Forecasts Research Articles

Can Ingredients-Based Forecasting Be Learned? Disentangling a Random Forest’s Severe Weather Predictions

Abstract Machine learning (ML)–based models have been rapidly integrated into forecast practices across the weather forecasting community in recent years. While ML tools introduce additional data to forecasting operations, there is a need for explainability to be available alongside the model output such that the guidance can be transparent and trustworthy for the forecaster. This work makes use of the algorithm tree interpreter (TI) to disaggregate the contributions of meteorological features used in the Colorado State University Machine Learning Probabilities (CSU-MLP) system, a random forest–based ML tool that produces real-time probabilistic forecasts for severe weather using inputs from the Global Ensemble Forecast System. TI feature contributions are analyzed in time and space for CSU-MLP day 2 and day 3 individual hazard (tornado, wind, and hail) forecasts and day 4 aggregate severe forecasts over a 2-yr period. For individual forecast periods, this work demonstrates that feature contributions derived from TI can be interpreted in an ingredients-based sense, effectively making the CSU-MLP physically interpretable. When investigated in an aggregate sense, TI illustrates that the CSU-MLP system’s predictions use meteorological inputs in ways that are consistent with the spatiotemporal patterns seen in meteorological fields that pertain to severe storm climatology. A discussion on how these insights could benefit forecast operations more broadly is also provided. Significance Statement Machine learning tools are becoming more common in weather forecasting settings, and there is a need to provide information to meteorologists on how machine learning models make their predictions in real time. Severe weather forecasts made by an operational machine learning model are deconstructed into meteorological components in a way that offers physically insightful context to the model’s predictions. The results show that the machine learning model uses the input meteorological fields to make predictions that resemble various aspects of severe storm climatology and environments. This work presents an avenue for using explainable artificial intelligence in operational weather forecasting by illustrating a method that could provide trust, transparency, and confidence in machine learning–based forecast guidance.

A semi-centralized Multi-Agent RL framework for efficient irrigation scheduling

Efficient water management in agriculture is essential for addressing the growing freshwater scarcity crisis. Multi-Agent Reinforcement Learning (MARL) has emerged as a promising method for solving daily irrigation scheduling problems in spatially variable fields, where management zones are employed to account for field variability. To enhance the application of MARL to address daily irrigation scheduling in large-scale fields with significant spatial variation, this study proposes a Semi-Centralized MARL (SCMARL) framework. The SCMARL framework adopts a hierarchical structure, decomposing the daily irrigation scheduling problem into two levels of decision-making. At the top level, a centralized coordinator agent determines irrigation timing, which is modeled as a discrete variable, based on field-wide soil moisture data, crop conditions, and weather forecasts. At the lower level, decentralized local agents use local soil moisture, crop, and weather information to determine the appropriate irrigation amounts for each management zone. To address the issue of non-stationarity in this framework, a state augmentation technique is employed, wherein the coordinator’s decision is incorporated into the decision-making process of the local agents. The SCMARL framework, which leverages the Proximal Policy Optimization algorithm for training the agents, is evaluated on a large-scale field in Lethbridge, Canada, and compared with an existing MARL irrigation scheduling approach. The results demonstrate improved performance, achieving a 4.0% reduction in water use and a 6.3% increase in irrigation water use efficiency.

Advanced Solar Energy Potential Assessment in Malawi: Utilizing High-Resolution WRF Model and GIS to Identify Optimal Sites for Photovoltaic Generation

Solar PV technology offers a promising electricity alternative in developing countries like Malawi, which face limited electricity access and increased vulnerability to climate change. This study assesses solar resources using the Weather Research and Forecasting (WRF) model's high-resolution capabilities. A two-year simulation helped compute solar PV's technical potential, focusing on annual energy generation, capacity factor, and global horizontal irradiance (GHI) variation. GIS and expert-based fuzzy analytic hierarchy process (FAHP) identified optimal solar PV sites, considering eight key constraints. Findings show an annual average daily peak GHI of 1106.45 W/m2 and a daily energy inflow of 6.76 kWh/m2/day, with a peak in October and a diffuse fraction of 10.61%, indicating clear skies ideal for solar PV. The average annual energy yield is about 14.11 TWh with a 21.48% capacity factor. Notably, the most suitable lands for solar PV, covering 10,288.2 km2 or 15.6% of acceptable land, are predominantly in the southern region. Thus, these solar PV potentials offer the possibility to significantly reduce CO2 emissions, placing Malawi in agreement with worldwide sustainability goals. Generally, this research lays a crucial foundation for Malawi's renewable energy strategies, highlighting the opportunity for improved electrification and more effective climate change mitigation endeavors.

Assessment of the Rainfed Wheat Productivity in a Changing Climate in Irbid, Jordan Using Statistical Downscaling and Random Forest Regression Prediction Under RCP4.5 & 8.5 Pathways.

Jordan faces risks associated with climate change and endeavors to sustain resources through the SDGs. This study examines the impacts of climate change on rainfed wheat productivity in Irbid Governorate during the historical period (1994-2021) and simulated for the future (2024-2100) using climate models (SDGs 1, 2, 11, 12, 13, & 15). Monthly precipitation, mean temperature, and annual wheat yield data were collected for the period (1994-2021) and analyzed using the Mann-Kendall test with Sen’s slope to investigate the historical trends and elaborate Phi-k correlation coefficient to determine their association. Monthly precipitation and mean temperature projections were collected from CSIRO-MK3.6.0 & GFDL-ESM2M CMIP5 ensemble models under two concentration pathways: RCP4.5 and RCP8.5. The projections were biased and corrected by the dynamic bilinear interpolation method and Delta, EQM, and QM statistical downscaling to enhance the projections' reliability and performance in capturing the region's climate. Taylor diagram was utilized to choose the best model to represent observed data. Cumulative Distribution Function and Probability Density Function Curves were plotted to describe the changes over decades based on climate models. Future wheat yields were predicted using the Random Forest Regression model. The results revealed a non-significant increase in total annual precipitation of 1.8% and 13.8% and a rise in annual mean temperature of 5.4% and 2.7% for Irbid and Baqura stations, respectively during the baseline timeframe on which wheat yield increased by 21.3%. The CSIRO model outperformed the GFDL model with greater fidelity in simulating historical monthly precipitation and mean temperatures. The CSIRO-MK3.6.0 model indicates precipitation and temperature shifts for near and far future periods. Temperature increases are expected to have more severe impacts in the far future (2073-2100) while projected decreases in precipitation of 0.06 mm/day and 0.10 mm/day under RCP4.5 and RCP8.5 affect yield reductions by 11.42 kg/ha and 12.21 kg/ha. Temperature increases of 2.7°C and 4.1°C under RCP 4.5 & 8.5 correspond to yield decreases of 24.825 kg/ha and 33.395 kg/ha, respectively. The study highlights the need for enhanced adaptation strategies; cultivating resilient wheat varieties, promoting crop rotation, improving meteorological monitoring and risk response approaches, and providing timely weather forecasts to mitigate the adverse effects of climate change on wheat productivity and ensure sustainable agriculture in Jordan.

Exploring the potential impacts of anthropogenic heating on urban climate during heatwaves

This study modeled the impact of anthropogenic heat (AH) on urban climates, focusing on Sydney during 2017’s heightened temperatures. The motivation behind this study stems from the increasing significance of understanding urban heat dynamics as cities globally grapple with regional climate change, necessitating targeted strategies for effective climate resilience cities. By investigating how varying levels of AH influence local urban climate conditions, this study addresses a critical gap in current urban climate study, particularly in the context of Sydney, an area that has not yet been extensively explored. Utilizing the weather research and forecasting (WRF) model coupled with building effect parameterization (BEP) and building energy model (BEM), i.e., the WRF/BEP + BEM model, four AH release scenarios were analyzed. Higher AH levels, especially at 14:00 LT, exhibited significant peaks: 266.5 W m−2 for sensible heat and 35.3 W m−2 for latent heat compared to the control scenario. This increase corresponded to a notable rise in ambient temperatures by 2.1 °C, with surface temperatures surging by 8.1 °C. Wind speeds notably increased by 4.6 m s−1 during higher AH release periods, affecting city airflow patterns. Moreover, elevated AH levels amplified the convective planetary boundary layer (PBL) height by 2013.7 m at 14:00 LT, potentially impacting pollutant dispersion and atmospheric quality. Notably, heightened AH profiles intensified sea breeze circulations, particularly impacting densely populated urban areas. These findings demonstrate a direct link between AH, exacerbated local urban warming, altered boundary layer dynamics, and intensified sea breeze circulations. This study emphasizes the urgent need to comprehend and manage AH for sustainable urban development and effective climate resilience strategies in Sydney and similar urban environments. By shedding light on these relationships, this study aims to contribute to the formulation of policies that mitigate urban overheating and enhance the livability of urban areas.

An Integrated Paradigm for Advanced Irrigation Systems Leveraging Internet of Things (IoT)

As the demand for precision in agriculture and effective sustainable resource management increases, there has been a lot of pressure to have more accurate and efficient soil moisture data-transmission systems. This paper includes the discussion on how machine learning (ML) and deep learning (DL) techniques can be improvise to further accuracy and performance in IoT-based smart irrigation systems. The system is based on the perception of soil moisture using IoT sensors collecting real-time environmental data in fields, such as soil moisture, temperature, humidity, and sunlight, to be transmitted to farmers or end-users using ThingSpeak and Thinger.io platforms for analysis, storage, and visualization. It enables real-time decisions and remote agricultural systems control using web page or mobile applications. The WSNs(Wireless sensor network) helps to automate irrigation and water management in agricultural sites. The developed SIS(Smart Irrigation Systems) is based on the sensor network that carries real-time information relating to moisture content in the soil, temperature, and humidity levels as very critical determinants of the proper irrigation schedule. These are analyzed every 15 minutes at the edge server. The system uses deep learning models to predict when the soil moisture falls below a threshold and automatically activates water pumps or sprinklers, which reduces human intervention and optimizes water usage in agriculture. An important part of research forms machine learning algorithms to improvise the performance. The recent advancements relies on several models, among them KNN (K-Nearest neighbors) and TimeGPT models [TimeGPT is a time-series forecasting model that utilizes the power of GPT (Generative Pre-trained Transformer) architecture to predict future values in a sequence.], in the prediction of soil moisture while achieving optimal irrigation schedules from previous available data and weather forecasting. The comparative analyses included an accuracy rate of 97 Parsant to 98 Parsant in KNN thus describing the high level of accuracy that the system can attain regarding the soil conditions and water requirements. This research provides the possibility of embedding IoT into machine learning that may form a new age smart agriculture system that, apart from irrigation automation, enhances decision-making in farmer practices. Adoption of such systems could contribute immensely to the sustainability of such agricultural practices mainly due to reduced water wastage, decrease operation costs, and improved crop yield. Integrating real-time environmental data and predictive analytics farmlands can be optimized and maintained even the land has water scarcity or difficult to receive rainfall or other climatic conditions.

Climatology of waterspouts in Mexico and numerical modeling of a particular case

Abstract Numerous waterspout events occur practically in all Mexican waters. The principal aim was to carry out a climatology of these events. A data compilation allowed the identification of 103 cases from 2010-2022. A Szilagyi waterspout nomogram was constructed to explain the formation mechanism using the ERA5 reanalysis data. The nomogram classified waterspout events according to their origin using thermodynamic parameters. The results revealed that the thermodynamics of some waterspouts in Mexico do not fit the current definition of the Szilagyi nomogram. Most waterspouts are thunderstorm-related and fit the nomogram. A minority that arises from land breezes and upper-low conditions also fit. A relevant case, which occurred in an area with a recurrent incidence of waterspouts, allowed observation of relevant dynamic properties, such as its strength, occurred on August 8, 2019, in Coatzacoalcos, Veracruz. The documented waterspout trajectory revealed that it moved through the sea and city. The Weather Research and Forecasting (WRF) model was used for modeling this event at a resolution of 250 m. The results showed the instability, a cyclonic disturbance (misocyclone) that caused the waterspout. The analysis showed the behavior of meteorological variables, such as temperature, humidity, wind, vorticity, and unstable atmospheric variables during the storm.

On the predictability of springtime ozone depletion events using the ECCC Global Deterministic Prediction System

Abstract Ozone depletion events are recurring phenomena in both polar regions, characterized by significant interannual variability. In this study, the Environment and Climate Change Canada (ECCC) Global Deterministic Prediction System is used to investigate the medium-range predictability of ozone and weather throughout the anomalous polar ozone depletion events of 2020. The system includes ozone assimilation and makes use of a prognostic ozone field for the computation of heating rates. The ozone scheme uses simplified photochemical modules to represent the impact of both gas-phase and heterogeneous reactions throughout polar ozone depletion events. The study shows that during the Boreal and Austral spring seasons, the predictability of the total ozone column exceeds 10 days and is comparable to the predictability of large-scale weather variables. It also demonstrates that over both polar regions, the inclusion of ozone radiative coupling has a significant impact on the temperature and wind distributions throughout the stratosphere. Over Antarctica, the ozone coupled forecasts are systematically colder at all lead times, which helps eliminate a temperature bias present in the model using climatological ozone. The strength of the polar vortex also increases significantly throughout the lower stratosphere, in better agreement with zonal wind analyses. Over the Arctic the use of an ozone-interactive model also produces significant changes in the temperature and wind forecasts, but the impact on the quality of the weather forecasts is generally neutral. The study shows the overall benefits of using ozone coupled models in the highly perturbed springtime conditions of the polar regions.

Quantifying and simulating the weather forecast uncertainty for advanced building control

Weather forecast uncertainty is unavoidable despite technological advancements. Accurately quantifying and modelling this uncertainty is essential for developing and comparing advanced building controllers. In this study, we present a structured approach using a first-order autoregressive model (AR(1)) to model uncertainty in ambient temperature and global solar irradiation (GHI) forecasts. We analyzed weather data from four cities and employed Jensen–Shannon divergence (JSD) to evaluate the similarity between synthetic and actual forecast errors. The average JSD values for temperature are 0.027 (Berkeley), 0.021 (Leuven), 0.018 (Berlin), and 0.008 (Oslo), and for GHI, the average JSD values are 0.016 (Berkeley), 0.058 (Leuven), and 0.013 (Berlin). The low JSD values indicate a high similarity between the synthetic and real forecast error distributions. Our approach successfully generates synthetic weather forecasts that mirror the statistical properties of actual forecasts. The implementation of our method for uncertain forecast generation is being added to the BOPTEST framework.

Projected changes in moisture sources and sinks affecting the US East Coast and the Caribbean Sea.

This study uses a combination of the FLEXPART Lagrangian dispersion model with the Weather Research and Forecasting (WRF) mesoscale Eulerian model (FLEXPART-WRF) to analyze the expected mid- to late-century changes in the moisture sources and sinks of the North American East Coast (ENA) and the Gulf of Mexico (GM), as well as their most relevant abrupt moisture transport events-atmospheric rivers (ARs) and low-level jets of the Great Plains (GPLLJ) and the Caribbean (CLLJ). Both the ENA and GM are expected to increase in importance as moisture source regions over the century, both overall and in their contributions to the ARs and both LLJs. A notable increase in the intensity of the GPLLJ and CLLJ moisture sources is also observed. All of these behaviors are neither spatially nor temporally homogeneous and need to be analyzed in a seasonal context. Likewise, the most relevant signs of change are practically all observed by the end of the century. Other noteworthy behaviors are also observed, including an increase in humidity associated with landfalling atmospheric river events in the winter months, or a notable latitudinal shift of the CLLJ's area of influence. These findings are best understood within the context of an observed increase in both continental precipitation and sea surfacetemperature.

Evaluation of Some Secondary Radio Meteorological Variables for Line-of-Sight Applications over Some Locations in Nigeria

Reliable data on radio propagation is required to suggest useful models for radio-climatic study. Computation of some secondary radio parameters across ten locations in Nigeria was done to deduce their effects on Line-of-Sight links. ERA-5 data obtained from the archive of European Centre for Medium-Range Weather Forecast (ECMWF) comprising of surface air and dew temperature, atmospheric pressure and relative humidity covering eight years (January 2010 – December 2017) was utilized. The results show that average radio refractivity values during the wet season (343.4 N-units) was higher than the dry season (273 N-units) and radio refractivity gradient values increase as the wet season progresses. Mean effective earth radius factor (k-factor) for the period of study were 1.38, 1.34, 1.67 and 1.72 for the rainforest, mangrove swamp, Sudan and guinea savannah regions respectively. It was also observed that a distinct relationship exists between the geo-climatic factor (K) and the seasons of the year with a range of 2.2 ×10-5 to 1.0 ×10-4.

Comparison of Identified Ice Supersaturated Regions for Contrail Avoidance Using Three Standard Weather Forecast Databases

Contrails form as a result of water vapor bonding with soot emitted from jet engines at cruise altitudes, leading to contrail formation in Ice Supersaturated Regions (ISSRs). Contrails are estimated to contribute approximately 2% to total anthropogenic global warming. Some researchers have developed simulation models to estimate the frequency, duration, and spatial distribution of contrails. Other researchers have identified issues with the accuracy of the data for predicting the timing and precise geographic positioning of ISSRs. This study presents a systematic review of 22 peer-reviewed articles that included detailed models of ISSR identification, identifying three atmospheric data sources, four parameters, and two equations for calculating the parameters derived. A further analysis revealed differences in the temperature and RHW readings across the three databases, resulting in differences in the RHI calculations and the identification of ISSRs. Over an 18-month period in Sterling, Virginia, USA, the radiosonde data and two atmospheric forecast databases identified the ISSR conditions on 44%, 47%, and 77% of days, respectively. Broken down by a flight level between 30,000 and 39,999 feet in altitude, these differences are highlighted further. The forecast databases overestimated the presence of ISSRs compared to the radiosonde data. These findings underscore the variability inherent in atmospheric datasets and the conversion methods, highlighting potential areas for refinement in ISSR prediction, notably in the development of ensemble forecasts based on several atmospheric databases. The implications of these results, the limitations of this study, and future work are discussed.

Extended drag-based model for better predicting the evolution of coronal mass ejections

Coronal mass ejections (CMEs) are one of the primary drivers of space weather disturbances, affecting both space-based and terrestrial technologies. The accurate prediction of CME trajectories and their arrival times at Earth is crucial for mitigating potential impacts. In this work, we introduce an extended drag-based model (EDBM) that incorporates an additional acceleration term to better capture the complex dynamics of CMEs as they propagate through the heliosphere. Preliminary results suggest that the EDBM can improve upon the classical drag-based model by providing more reliable estimates of CME travel times, particularly in cases where the CME experiences residual acceleration. However, further validation is required to fully assess the operational potential of the model for space weather forecasting. This study lays the groundwork for future investigations and applications, with the aim of enhancing the accuracy of CME prediction models.

Predict Weather with Machine Learning

Everything, from farming to air travel, in our contemporary world is dependent on the weather forecast. However, traditional methods often ignore the intrinsic uncertainty in weather patterns leading to inaccurate predictions. This paper presents a groundbreaking method of predicting weather using Bayesian deep learning models; specifically, Bayesian neural networks are employed to capture the probabilistic nature of weather events and produce more accurate and reliable forecasts. These models represent a significant milestone in foretelling strategies as they are evaluated against popular metrics and trained with archived atmospheric data. Additionally, gradient boosting classifiers and decision tree classifiers have been performed as queries for performing comparative analyses about different machine learning algorithms used for predicting occurrence of rain tomorrow or any other type of climate-related information. For enhanced interpretability, this paper has developed interactive visualization tools that allow users to interactively explore predicted weather patterns and analyze levels of uncertainty with an aim of advancing the field of meteorology through providing actionable insights for responsible decision making within weather sensitive domains.

Efficient magnetohydrodynamic modelling of the time-evolving corona by COCONUT

Magnetohydrodynamic (MHD) solar corona models are critical in the Sun-to-Earth modelling chain and are the most complex and computationally intensive component. Compared to quasi-steady-state corona models that are constrained by a time-invariant magnetogram over a Carrington rotation (CR) period, time-evolving corona models driven by time-varying photospheric magnetograms are more realistic and can maintain more useful information to accurately describe solar wind evolution and forecast coronal mass ejection propagation. Implicit methods have significantly improved the efficiency of quasi-steady MHD coronal modelling. However, developing efficient time-evolving corona models to improve space weather forecasting is also important. This paper aims to demonstrate that time-evolving corona simulations can be performed efficiently and accurately using an implicit method with relatively large time steps, thus reducing the overall computational cost. We also evaluate differences between coronal structures captured by time-evolving and quasi-steady simulations over a CR period during solar minimum. We extended the quasi-steady COCONUT model, a global MHD corona model that uses implicit methods to select large time steps, into a time-evolving corona model. Specifically, we used a series of hourly updated photospheric magnetograms to drive the evolution of coronal structures from the solar surface to $25; R_s$ during two CRs around the 2019 eclipse in an inertial coordinate system. At each time step, the inner-boundary magnetic field was temporal-interpolated and updated from adjacent observation-based magnetograms. We compare the time-evolving and quasi-steady simulations to demonstrate that the differences in these two types of coronal modelling can be obvious even for a solar minimum. The relative differences in radial velocity and density can be over $15 %$ and $25 %$ at 20$;R_s$ during one CR period. We also evaluated the impact of time steps on the simulation results. Using a time step of approximately 10 minutes balances efficiency and necessary numerical stability and accuracy for time-evolving corona simulations around solar minima. The chosen 10-minute time step significantly exceeds the Courant-Friedrichs-Lewy stability condition needed for explicit corona modelling, and the time-evolving COCONUT can thus simulate the coronal evolution during a full CR within only 9 hours (using 1080 CPU cores for 1.5M grid cells). The simulation results demonstrate that time-evolving MHD coronal simulations can be performed efficiently and accurately using an implicit method, offering a more realistic alternative to quasi-steady-state simulations. The fully implicit time-evolving corona model thus promises to simulate the time-evolving corona accurately in practical space weather forecasting.

Advanced Tour Guide: A Web Application for Travelers

This research paper explores the development of an "Advanced Tour Guide" web application designed to assist travelers by providing detailed location information, booking options, and relevant travel advice. The project incorporates a front-end architecture built using HTML, CSS, and JavaScript, focusing on user-friendly interfaces like the landing and location details pages. The back-end employs Node.js and SQL to manage user data and enable booking functionalities, while third-party APIs for weather forecasts and travel advisories enrich the user experience. In the literature review, existing tour guide applications like Google Travel and TripAdvisor are evaluated, identifying both their strengths and limitations. Furthermore, the critical role of APIs in travel applications is explored, emphasizing challenges such as rate limiting and API key management. The system's design includes responsive web design techniques for mobile and desktop views, with future plans to integrate machine learning for personalized recommendations and additional APIs like currency exchange and local transport details. The project aims to overcome challenges in API integration and front-end scalability, with an outlook toward expanding functionalities and providing a comprehensive solution for modern travelers.

Investigation of double geomagnetic storms on 3 and 4 February 2022 using machine learning approach

ABSTRACT The current work utilising machine learning algorithms to investigate the precursor that follows halo coronal mass ejection (CME), eventually leading to moderate geomagnetic storms on the 3rd and 4th of February 2022. The methodology involved developing and testing a machine learning model on collected data, implemented with a Gradient Boosting Regressor (GBR) technique. The GBR algorithm demonstrated strong performance, yielding high accuracy and precision over various error metrics. These findings underscore the potential of machine learning methods to effectively estimate geomagnetic storms. Specifically, they position the GBR algorithm as an optimal choice for this prediction task, outperforming other evaluated options. This study provides evidence for the suitability of the GBR regressor for reliable SYM-H index modelling. These results show that geomagnetic storms may be directly predicted from solar wind parameter data, with a lead time of several days for forecasting, which is important for improving space weather forecasts. According to the study, using the GBR regressor model improves the performance to up to 95%.

Optimizing photovoltaic power plant forecasting with dynamic neural network structure refinement

Reliable prediction of photovoltaic power generation is key to the efficient management of energy systems in response to the inherent uncertainty of renewable energy sources. Despite advances in weather forecasting, photovoltaic power prediction accuracy remains a challenge. This study presents a novel approach that combines genetic algorithms and dynamic neural network structure refinement to optimize photovoltaic prediction. This methodology dynamically adjusts the neural network parameters during training, including the number of neurons, transfer functions, weights, and biases, to minimize the root mean square error. Evaluation was performed on twelve representative days using annual, monthly, and seasonal data, and a comparison was made with multiple linear regression and nonlinear autoregressive neural network models, demonstrating the approach’s effectiveness. Evaluation metrics such as mean square error, R-value, and mean percentage error reveal promising prediction accuracy. MATLAB is used for modeling, training, and testing, and a real 4.2 kW PV plant is used for validation. The results indicate significant improvements, with mean square errors as low as 20 W on cloudy days and 175 W on sunny days. The proposed methodology achieves prediction versus target regressions consistency, with R values ranging from 0.95824 to 0.99980, highlighting its efficiency in providing reliable predictions of PV power generation.

Solar radiation estimation in West Africa: impact of dust conditions during the 2021 dry season

Abstract. The anticipated increase in solar energy production in West Africa requires high-quality solar irradiance estimates, which are affected by meteorological conditions and in particular the presence of desert dust aerosols. This study examines the impact of incorporating desert dust into solar irradiance and surface temperature estimations. The research focuses on a case study of a dust event in March 2021, which is characteristic of the dry season in West Africa. Significant desert aerosol emissions at the Bodélé Depression are associated with a Harmattan flow that transports the plume westwards. Simulations of this dust event were conducted using the meteorological Weather Research and Forecasting (WRF) Model alone, as well as coupling it with the CHIMERE chemistry transport model, using three different datasets for the dust aerosol initial and boundary conditions (CAMS, GOCART, and MERRA-2). Results show that considering desert dust reduces estimation errors in global horizontal irradiance (GHI) by about 75 %. The dust plume caused an average of 18 % reduction in surface solar irradiance during the event. Additionally, the simulations indicated a positive bias in aerosol optical depth (AOD) and PM10 surface concentrations. The choice of dataset for initial and boundary conditions minimally influenced GHI, surface temperature, and AOD estimates, whereas PM10 concentrations and aerosol size distribution were significantly affected. This study underscores the importance of incorporating dust aerosols into solar forecasting for better accuracy.

Study on the Fire Spread Characteristics of High-Rise Building Facades Under Strong Wind Conditions Based on the Combination of WRF and CFD

The spread of fires on the facades of high-rise buildings is highly influenced by atmospheric wind conditions, particularly in strong wind environments. A strong wind environment refers to the situation where the wind speed reaches level 6 or above, or the wind speed is between 10.8 m/s and 13.8 m/s. We conducted an in-depth study of the characteristics of flame spread on the facades of high-rise buildings under strong wind conditions. A nested coupling method based on WRF (Weather Research and Forecasting) and CFD (computational fluid dynamics) software (Ansys Fluent 2021) was used. The mesoscale meteorological simulation software WRF was utilized to obtain regional airflow variation data within a radius of 2 km around the high-rise building. Subsequently, these data were coupled with the CFD software (Ansys Fluent 2021) to simulate and obtain realistic wind field data within a 400 m range around the building. Finally, these realistic wind field data were used for FDS (Fire Dynamics Simulator) fire simulations and model experiments to accurately replicate building fire scenarios under strong wind conditions. The results indicate that using grid nesting for boundary condition division would help to improve the accuracy of fire spread characteristics on the facades of high-rise buildings under strong wind conditions. For a high-rise building, both headwinds and tailwinds promote vertical and horizontal flame spread, with a more significant impact on vertical flame spread speed. Side winds enhance horizontal flame spread but inhibit vertical flame spread. These findings provide a reference for the effective design of fire protection systems for the facades of high-rise buildings.