Nowcasting Model Research Articles

LLMDiff: Diffusion Model Using Frozen LLM Transformers for Precipitation Nowcasting.

Precipitation nowcasting, which involves the short-term, high-resolution prediction of rainfall, plays a crucial role in various real-world applications. In recent years, researchers have increasingly utilized deep learning-based methods in precipitation nowcasting. The exponential growth of spatiotemporal observation data has heightened interest in recent advancements such as denoising diffusion models, which offer appealing prospects due to their inherent probabilistic nature that aligns well with the complexities of weather forecasting. Successful application of diffusion models in rainfall prediction tasks requires relevant conditions and effective utilization to direct the forecasting process of the diffusion model. In this paper, we propose a probabilistic spatiotemporal model for precipitation nowcasting, named LLMDiff. The architecture of LLMDiff includes two networks: a conditional encoder-decoder network and a denoising network. The conditional network provides conditional information to guide the denoising network for high-quality predictions related to real-world earth systems. Additionally, we utilize a frozen transformer block from pre-trained large language models (LLMs) in the denoising network as a universal visual encoder layer, which enables the accurate estimation of motion trend by considering long-term temporal context information and capturing temporal dependencies within the frame sequence. Our experimental results demonstrate that LLMDiff outperforms state-of-the-art models on the SEVIR dataset.

Enhanced rainfall nowcasting of tropical cyclone by an interpretable deep learning model and its application in real-time flood forecasting

Reliable Tropical Cyclone (TC) rainfall and flood forecasts play an important role in disaster prevention and mitigation. Numerous studies have demonstrated the promising performance of deep learning in hydrometeorological forecasts. However, few studies have investigated the potential enhancement of advanced TC track forecasts in predicting rainfall and induced flood. In this study, a novel rainfall nowcasting model (TCRainNet) is developed by fusing TC track characteristics with antecedent rainfall in a Convolution LSTM to predict hourly rainfall with a lead time of 6 h. The nowcasts are subsequently used to drive an event-based Xin’anjiang hydrological model for real-time flood forecasting. The model performance is interpretated by the occlusion sensitivity approach, and the propagation of errors from TC track forecasts to flood forecasts is quantified. The results underscore the superiority of TC track characteristics as input features for rainfall nowcasts, as indicated by a Mutual Information value of up to 0.51. The generated nowcasts are found to have averaged Probability of Detection (POD) and Critical Success Index (CSI) greater than 0.27 and 0.2 respectively. The Mean Absolute Error (MAE) of the nowcasts falls below 2.6 mm, which is only 46 % of the ECMWF operational high-resolution forecasts. The rainfall-driven flood forecasts have NSE greater than 0.7 and PBIAS smaller than 20 % with lead time up to + 4 h. It is shown that the position error of 0.45° and intensity error of 10 hPa&7.8 m/s in TC track forecasts generally result in 0.9 mm degradation in rainfall forecasts and 10% decline in the accuracy of rainfall-driven flood forecasts. The effectiveness of our method presents favorable applicability in advancing disaster mitigation efforts.

A nowcasting model of industrial production using alternative data and machine learning approaches

Recent years have seen a growing trend to utilize "alternative data" in addition to traditional statistical data in order to understand and assess economic conditions in real time. In this paper, we construct a nowcasting model for the Indices of Industrial Production (IIP), which measure production activity in the manufacturing sector in Japan. The model has the following characteristics: First, it uses alternative data (mobility data and electricity demand data) that is available in real-time and can nowcast the IIP one to two months before their official release. Second, the model employs machine learning techniques to improve the nowcasting accuracy by endogenously changing the mixing ratio of nowcast values based on traditional economic statistics (the Indices of Industrial Production Forecast) and nowcast values based on alternative data, depending on the economic situation. The estimation results show that by applying machine learning techniques to alternative data, production activity can be nowcasted with high accuracy, including when it went through large fluctuations during the spread of the COVID-19 pandemic.

Food price inflation nowcasting and monitoring

Rising food prices may rapidly push vulnerable populations into food insecurity, especially in developing economies and in low-income countries, where a substantial share of the financial resources available to the poorest households is spent on food. To capture soaring food prices and help in designing mitigating measures, we developed two complementary products: a nowcasting model that estimates official food consumer price inflation up to the current month and a daily food price monitor that checks whether the growth rate of a few basic food commodities exceeds a statistical threshold. Both products were designed with the consideration that the rapid acquisition of data and the automated extraction of insights are indispensable tools for policymakers, particularly in times of crisis. Our framework is characterized by three key aspects. Firstly, we leverage two non-traditional data sources to emphasize the importance of real-time information: a crowdsourced repository of daily food prices and textual insights obtained from newspapers articles. Secondly, our framework offers a global perspective, encompassing 225 countries and territories, which enables the monitoring of food prices dynamics on a global scale. Thirdly, results are made accessible daily via an intuitive and user-friendly interactive dashboard.

Relationship Between NM Data and Radiation Dose at Aviation Altitudes During GLE Events

AbstractGround‐level enhancements (GLEs) are sporadic events that signal the arrival of high fluxes of solar energetic particles (SEPs) that have been produced by solar eruptions. Ground‐level enhancement events are characterized by a significant increase in the count rate of ground‐based neutron monitors (NMs). The arrival of high‐energy SEPs in the atmosphere leads to an enhancement of the radiation environment, with the enhancement at aviation altitudes being particularly hazardous to human health as pilots, crew, and airline passengers can be subjected to dangerous levels of radiation during a GLE. Through the use of a currently expanding library of analyzed GLEs and the application of a newly developed atmospheric radiation model, both of which have been created in‐house, we found a strong statistically significant relationship between real‐time NM data during GLE events and the radiation doses at aviation altitudes. This result provides a strong scientific basis for the use of real‐time NM data as a proxy for radiation dose estimates during GLE events and aids in the development of future nowcasting models to help mitigate the dangerous impacts of future GLEs.

Nowcasting methods to improve the performance of respiratory sentinel surveillance: lessons from the COVID-19 pandemic

Respiratory diseases, including influenza and coronaviruses, pose recurrent global threats. This study delves into the respiratory surveillance systems, focusing on the effectiveness of SARI sentinel surveillance for total and severe cases incidence estimation. Leveraging data from the COVID-19 pandemic in Chile, we examined 2020–2023 data (a 159-week period) comparing census surveillance results of confirmed cases and hospitalizations, with sentinel surveillance. Our analyses revealed a consistent underestimation of total cases and an overestimation of severe cases of sentinel surveillance. To address these limitations, we introduce a nowcasting model, improving the precision and accuracy of incidence estimates. Furthermore, the integration of genomic surveillance data significantly enhances model predictions. While our findings are primarily focused on COVID-19, they have implications for respiratory virus surveillance and early detection of respiratory epidemics. The nowcasting model offers real-time insights into an outbreak for public health decision-making, using the same surveillance data that is routinely collected. This approach enhances preparedness for emerging respiratory diseases by the development of practical solutions with applications in public health.

Nowcasting of High-Intensity Rainfall for Urban Applications in the Netherlands

Abstract Radar rainfall nowcasting has mostly been applied to relatively large (often rural) domains (e.g., river basins), although rainfall nowcasting in small urban areas is expected to be more challenging. Here, we selected 80 events with high rainfall intensities (at least one 1-km2 grid cell experiences precipitation >15 mm h−1 for 1-h events or 30 mm day−1 for 24-h events) in five urban areas (Maastricht, Eindhoven, The Hague, Amsterdam, and Groningen) in the Netherlands. We evaluated the performance of 9060 probabilistic nowcasts with 20 ensemble members by applying the short-term ensemble prediction system (STEPS) from Pysteps to every 10-min issue time for the selected events. We found that nowcast errors increased with decreasing (urban) areas especially when below 100 km2. In addition, at 30-min lead time, the underestimation of nowcasts was 38% larger and the discrimination ability was 11% lower for 1-h events than for 24-h events. A set of gridded correction factors for the Netherlands, CARROTS (Climatology-based Adjustments for Radar Rainfall in an Operational Setting) could adjust the bias in real-time QPE and nowcasts by 70%. Yet, nowcasts were still found to underestimate rainfall more than 50% above 40-min lead time relative to the reference, which indicates that this error originates from the nowcasting model itself. Also, CARROTS did not adjust the rainfall spatial distribution in urban areas much. In summary, radar-based nowcasting for urban areas (between 67 and 213 km2) in the Netherlands exhibits a short skillful lead time of about 20 min, which can only be used for last-minute warning and preparation.

Ionospheric Nowcasting Over Italy Through Data Assimilation: A Synergy Between IRI UP and IONORING

AbstractAn accurate modeling of the ionosphere electron density is pivotal to guarantee the effective operation of communication and navigation systems, particularly during Space Weather events. Despite the crucial contribution of empirical models like the International Reference Ionosphere (IRI), their limitations in predicting ionospheric variability, especially under geomagnetically disturbed conditions, are acknowledged. The solution proposed in this work involves integrating real‐time, spatially distributed ionospheric measurements into climatological models through data assimilation. To enhance our predictive capabilities, we present an upgrade of the IRI UP data‐assimilation method, incorporating real‐time vertical total electron content (vTEC) maps from the IONORING algorithm for nowcasting ionospheric conditions over Italy. This approach involves updating the IRI F2‐layer peak electron density description through ionospheric indices, to finally produce real‐time maps over Italy of the ordinary critical frequency of the F2‐layer, foF2, which is crucial for radio‐propagation applications. The IRI UP–IONORING method performance has been evaluated against different climatological and nowcasting models, and under different Space Weather conditions, by showing promising outcomes which encourages its inclusion in the portfolio of ionospheric real‐time products available over Italy. The validation analysis highlighted also what are the current limitations of the IRI UP–IONORING method, particularly during nighttime for severely disturbed conditions, suggesting avenues for future enhancements.

Nowcasting Vietnam's Export Growth with Mixed Frequency Data

Purpose: The primary objective of this study is to investigate and employ a practical and meaningful nowcasting model to predict Vietnam's export growth based on factors of export supply and demand alongside relevant financial indicators. Theoretical Framework: This study employs the concepts and theories of nowcasting model with mixed frequency data to create the conceptual framework. Methodology: This study employs four commonly-used models in nowcasting: the bridge equation model (BEQ), Bayesian VAR model (BVAR), mixed frequency vector autoregressive model (MFVAR), and mixed data sampling regression (MIDAS). Findings: According to the experimental findings, the mixed frequency data models outperformed the models utilizing the same frequency data in nowcasting Vietnam's export growth. Additionally, this model demonstrated effectiveness in instantaneous and short-term forecasting. MIDAS emerged as the most suitable choice for nowcasting Vietnam's export growth among the models examined. Implication of Research: using data with mixed frequency along with corrresponding methods is the good way for nowcasting. Originality/Value: This study used macroeconomics factors to nowcast the export growth in Vietnam. It applied four different models including BEQ, BVAR, MFVAR, and MIDAS. The study reveals the roles of data and the potential capability in nowcasting of MIDAS model.

Are national or regional surveys useful for nowcasting regional jobseekers? The case of the French region of Pays‐de‐la‐Loire

AbstactIn this paper we develop nowcasting models for the Pays‐de‐la‐Loire's jobseekers, a dynamic French regional economy. We ask whether these regional nowcasts are more accurate by only using the regional data or by combining the national and regional data. For this purpose, we use penalized regressions, random forest, and dynamic factor models as well as dimension reduction approaches. The best nowcasting performance is provided by the DFM estimated on the regional and regional‐national databases as well as the Elastic‐Net model with a prior screening step for which the national data are the most frequently selected data. For the latter, it appears that the Change in foreign orders in the industry sector, the OECD Composite leading indicator, and the BdF Business sentiment indicator are among the major predictors.

Exploring the use of 3D radar measurements in predicting the evolution of single-core convective cells

Object-based radar rainfall nowcasting is a widely used technique for convective storm prediction. Currently, most existing object-based nowcasting methods primarily focus on predicting cell movements, neglecting the temporal evolution of cell properties such as size, shape, and intensity. Incorporating this evolution is critical for improving predictability in convective storms. While previous studies have used three-dimensional (3D) radar observations to capture vertical changes during convective cell formation, these efforts often analyse or reconstruct specific convective events. Integrating 3D radar information into operational object-based radar rainfall nowcasting remains an open challenge. This research addresses this challenge using deep learning (DL) techniques. More specifically, a DL-based prediction model is developed, which uses 2D and 3D cells' properties retrieved from 3D radar reflectivity data at the current time and across the past 15 min to predict the evolution of these properties over the next 15 min. This model could eventually be integrated into existing object-based nowcasting models. A total of 4708 cell lifecycles, extracted from high-resolution (5-min, 1-km, 24 levels at 0.5 km intervals) 3D radar data across the UK, are used to train the model, and a total of 1177 lifecycles are used for testing. The proposed model is shown to predict the evolution of single-core convective cells effectively, including changes in 2D projected geometry and mean 2D and 3D reflectivity. In particular, by incorporating information on the vertical evolution of convective cores, the prediction errors of mean reflectivity (in both 2D and 3D) can be reduced by approximately 50% at 15-min forecast lead time, as compared to a persistence forecast. Keywords: radar, tracking, convective cell, nowcasting, 3D, deep learning, lstm.

Deep learning model for heavy rainfall nowcasting in South Korea

Accurate nowcasting is critical for preemptive action in response to heavy rainfall events (HREs). However, operational numerical weather prediction models have difficulty predicting HREs in the short term, especially for rapidly and sporadically developing cases. Here, we present multi-year evaluation statistics showing that deep-learning-based HRE nowcasting, trained with radar images and ground measurements, outperforms short-term numerical weather prediction at lead times of up to 6 h. The deep learning nowcasting shows an improved accuracy of 162%–31% over numerical prediction, at the 1-h to 6-h lead times, for predicting HREs in South Korea during the Asian summer monsoon. The spatial distribution and diurnal cycle of HREs are also well predicted. Isolated HRE predictions in the late afternoon to early evening which mostly result from convective processes associated with surface heating are particularly useful. This result suggests that the deep learning algorithm may be available for HRE nowcasting, potentially serving as an alternative to the operational numerical weather prediction model.

Assessing Nowcast Models in the Central Mexico Region Using Radar and GOES-16 Satellite Data

In this study, the nowcast models provided by the Python pySTEPS library were evaluated using radar derived rain rate data and the satellite product Split-Window Difference (SWD) based on GOES-16 data, focusing on central Mexico. Initially, we obtained a characterization of the rainfall that occurred in the region using the radar rain rate and the SWD. Subsequently the nowcasts were evaluated using both variables. Two nowcast models were employed from pySTEPS: Extrapolation and S-PROG. The results indicate that average SWD is below 2.5 K, 90 min before the onset of rainfall events, and, on average, the SWD is 2 K during rainfall events. The results from both nowcast models were accurate and produced similar results. The nowcasts performed better when SWD data were used as input, having an average Probability of Detection (PoD) above 70% and a False Alarm Rate (FAR) reaching 40% for the 15-min prediction. The nowcasts were less accurate using the radar rain rate as input for the 15-min forecast, where the PoD was maximum 70% and FAR reaching 40%. However, these nowcasts were more reliable during well-organized precipitation events. In this work, it was determined that the nowcast models provided by pySTEPS can provide valuable rain forecasts using GOES-16 satellite and radar data for the central Mexico region.

A nowcast model to predict outdoor flea activity in real time for the contiguous United States.

The cat flea (Ctenocephalides felis), a parasite commonly found on both dogs and cats, is a competent vector for several zoonotic pathogens, including Dipylidium caninum (tapeworms), Bartonella henselae (responsible for cat scratch disease) and Rickettsia felis (responsible for flea-borne spotted fever). Veterinarians recommend that both cats and dogs be routinely treated with medications to prevent flea infestation. Nevertheless, surveys suggest that nearly one third of pet owners do not routinely administer appropriate preventatives. A mathematical model based on weighted averaging over time is developed to predict outdoor flea activity from weather conditions for the contiguous United States. This 'nowcast' model can be updated in real time as weather conditions change and serves as an important tool for educating pet owners about the risks of flea-borne disease. We validate our model using Google Trends data for searches for the term 'fleas.' This Google Trends data serve as a proxy for true flea activity, as validating the model by collecting fleas over the entire USA is prohibitively costly and time-consuming. The average correlation (r) between the nowcast outdoor flea activity predictions and the Google Trends data was moderate: 0.65, 0.70, 0.66, 0.71 and 0.63 for 2016, 2017, 2018, 2019 and 2020, respectively. However, there was substantial regional variation in performance, with the average correlation in the East South Atlantic states being 0.81 while the average correlation in the Mountain states was only 0.45. The nowcast predictions displayed strong seasonal and geographic patterns, with predicted activity generally being highest in the summer months. The nowcast model is a valuable tool by which to educate pet owners regarding the risk of fleas and flea-borne disease and the need to routinely administer flea preventatives. While it is ideal for domestic cats and dogs to on flea preventatives year-round, many pets remain vulnerable to flea infestation. Alerting pet owners to the local increased risk of flea activity during certain times of the year may motivate them to administer appropriate routine preventives.

Enhancing ozone nowcasting over East Asia using a data-to-data translation approach with observations from a geostationary environment monitoring spectrometer

The degradation of air quality caused by excessive anthropogenic ozone (O3) concentrations negatively affects ecosystems and the atmosphere. To monitor this issue, the Geostationary Environment Monitoring Spectrometer (GEMS) aboard the Geostationary Korea Multi-Purpose Satellite (GK)-2B provides ozone products over Asia-centered circular regions based on the ozone algorithm derived from representative polar-orbit atmospheric environmental satellites. In this study, we developed a few-hour GEMS O3 nowcasting model using data-to-data (D2D) translation with a conditional generative adversarial network. This model is based on hourly GEMS O3 time-series products and can be used to predict ozone concentrations. The D2D model underwent training and testing employing paired input and output datasets of GEMS O3, with data collected from March 22, 2020, to June 21, 2020, and from March 22, 2021, to June 18, 2021, respectively. The resulting D2D ozone nowcasting model was used to determine ozone concentrations in time zones where GEMS O3 products were unavailable. Test results of the D2D model demonstrated excellent statistical scores, including a bias of 2.162 Dobson Units (DU, where 1 DU corresponds to 2.687 × 1016 molecules/cm2), root-mean-square error (RMSE) of 5.606 DU, a correlation coefficient (CC) of 0.994 for 1-h prediction, a bias of 1.421 DU, RMSE of 5.903 DU, and CC of 0.992 for 2-h prediction, and a bias of 1.169 DU, RMSE of 6.797 DU, and CC of 0.988 for 3-h prediction. Despite the dataset pairing and number limitations, the D2D prediction model accurately forecasted GEMS O3 within 3 h in East Asia.

Blending of a novel all sky imager model with persistence and a satellite based model for high-resolution irradiance nowcasting

High shares of variable energy sources such as photovoltaics (PV) make balancing network load and generation increasingly challenging. Electricity grids with high PV-penetration benefit from the consideration of intra-minute and intra-hour variabilities via nowcasts (shortest-term forecasts).In this study we present a novel all-sky imager (ASI) nowcasting system which is benchmarked against an established ASI method, a satellite nowcasting system and persistence. Nowcasts of our novel ASI method, satellite method and persistence are subsequently combined to a hybrid model with improved accuracy.ASI systems exploit sky images from fisheye cameras to analyze sky conditions, predict upcoming cloud situations and derive irradiance nowcasts from these. Our ASI method uses a novel machine-learning (ML) based approach that makes direct use of pixel values and other image features. Measurements from a spatially distributed irradiance measuring network of eight stations within a radius of 10km around the camera position are used to train our ASI and our hybrid model and validate our nowcasting results. In the evaluation we put a special focus on irradiance variability conditions. The high share of stable cloud situations within our dataset (∼68 %) results in a relatively good performance of persistence nowcasts. In higher variability situations persistence is quickly outperformed by all other nowcasting methods.Our ASI method clearly outperforms the reference ASI method at all lead times (LTs). It moreover exhibits a significantly lower root mean square error (RMSE) than the satellite-based method up to LT≤11min ahead and a lower mean absolute error (MAE) throughout the entire interval. The hybrid model brings about an RMSE improvement score of 5-13% over the respective optimal individual method.

Advancing very short-term rainfall prediction with blended U-Net and partial differential approaches

Accurate and timely prediction of short-term rainfall is crucial for reducing the damages caused by heavy rainfall events. Therefore, various precipitation nowcasting models have been proposed. However, the performance of these models still remains limited. In particular, the current operational precipitation nowcasting method, which is based on radar echo tracking, such as the McGill Algorithm for Precipitation Nowcasting by Lagrangian Extrapolation (MAPLE), has a critical drawback when predicting newly developed or decayed precipitation fields. Recently proposed deep learning models, such as the U-Net convolutional neural network outperform the models based on radar echo tracking. However, these models are unsuitable for operational precipitation nowcasting due to their blurry predictions over longer lead times. To address these blurry effects and enhance the performance of U-Net-based rainfall prediction, we propose a blended model that combines a partial differential equation (PDE) model based on fluid dynamics with the U-Net model. The evaluation of the forecast skill, based on both qualitative and quantitative methods for 0–3-h lead times, demonstrates that the blended model provides less blurry and more accurate rainfall predictions compared with the U-Net and partial differential equation models. This indicates the potential to enhance the field of very short-term rainfall prediction. Additionally, we also evaluated the monthly-averaged forecast skills for different seasons, and confirmed the operational feasibility of the blended model, which contributes to the performance enhancement of operational nowcasting.

SnotWatch: data collaboration informing disease impact

The spectrum of health conditions associated with the circulation of respiratory viruses remains poorly quantified. The SnotWatch platform has established a databank containing test results for respiratory polymerase chain reaction tests conducted by eight major Victorian pathology laboratories and has obtained access to major healthcare presentation databases, such as the Victorian Admitted Episodes Dataset, Victorian Emergency Minimum Dataset and POLAR GP. By establishing a statistical approach to describe the associations between healthcare presentations and circulating viruses in space and time at a population level, SnotWatch is uncovering the health conditions associated with virus circulation. SnotWatch’s methods have been validated by demonstrating the capacity to describe well-known associations, but have also uncovered novel associations. Methods used have also been able to incorporate environmental exposures. By elucidating the associations between viruses and health conditions, SnotWatch allows clinicians to make better informed diagnoses, anticipate healthcare presentations and outcomes, optimises resource management by healthcare providers and allows the benefits of vaccination to be more fully appreciated. SnotWatch’s future includes establishing methods for attributable burden, developing models for nowcasting and deepening our understanding of the economic impact of viruses.

A Lightning Nowcasting Model using GNSS PWV and Multi-source Data

A Lightning Nowcasting Model using GNSS PWV and Multi-source Data

A Deep Learning Model for Precipitation Nowcasting Using Multiple Optical Flow Algorithms

Abstract The optical flow technique has advantages in motion tracking and has long been employed in precipitation nowcasting to track the motion of precipitation fields using ground radar datasets. However, the performance and forecast time scale of models based on optical flow are limited. Here, we present the results of the application of the deep learning method to optical flow estimation to extend its forecast time scale and enhance the performance of nowcasting. It is shown that a deep learning model can better capture both multispatial and multitemporal motions of precipitation events compared with traditional optical flow estimation methods. The model comprises two components: 1) a regression process based on multiple optical flow algorithms, which more accurately captures multispatial features compared with a single optical flow algorithm; and 2) a U-Net-based network that trains multitemporal features of precipitation movement. We evaluated the model performance with cases of precipitation in South Korea. In particular, the regression process minimizes errors by combining multiple optical flow algorithms with a gradient descent method and outperforms other models using only a single optical flow algorithm up to a 3-h lead time. Additionally, the U-Net plays a crucial role in capturing nonlinear motion that cannot be captured by a simple advection model through traditional optical flow estimation. Consequently, we suggest that the proposed optical flow estimation method with deep learning could play a significant role in improving the performance of current operational nowcasting models, which are based on traditional optical flow methods. Significance Statement The purpose of this study is to improve the accuracy of short-term rainfall prediction based on optical flow methods that have been employed for operational precipitation nowcasting. By utilizing open-source libraries, such as OpenCV, and commonly applied machine learning techniques, such as multiple linear regression and U-Net networks, we propose an accessible model for enhancing prediction accuracy. We expect that the improvement in prediction accuracy will significantly improve the practical application of operational precipitation nowcasting.