AN ARTIFICIAL INTELLIGENCE APPROACH TO PREDICTION OF EXTREME EVENTS: THE CASE OF STORMS IN WESTERN FRANCE

Storms represent an increased source of risk that affects human life, property, and the environment. Prediction of these events, however, is challenging due to their low frequency of occurrence. This paper proposed an artificial intelligence approach to address this challenge and predict storm characteristics and occurrence using a gated recurrent unit (GRU) neural network and a support vector machine (SVM). Historical weather and marine measurements collected from buoy data, as well as a database of storms containing all the extreme events that occurred in Brittany and Pays de la Loire regions, Western France, since 1996, were used. Firstly, GRU was used to predict the characteristics of storms (wind speed, pressure, humidity, temperature, and wave height). Then, SVM was introduced to identify storm-specific patterns and predict storm occurrence. The approach adopted leads to the prediction of storms and their characteristics, which could be used widely to reduce the awful consequences of these natural disasters by taking preventive measures.

Studying auroral activity using the SME index at the magnetic storm main phase during CIR and ICME events

In this paper, we examine the relationship of the SME index with magnetic storm characteristics and interplanetary medium parameters during the main phase of magnetic storms caused by CIR and ICME events. Over the period 1990–2017, 107 magnetic storms driven by (64) CIR and (43) ICME events have been selected. In contrast to AE and Kp, a stronger correlation is shown to exist between the average SME index (SMEaver) and interplanetary medium parameters during the magnetic storm main phase. Close correlation coefficients between SMEaver and the SW electric field (southward IMF Bz) have been obtained for CIR and ICME events. SMEaver has been found to increase with the rate of magnetic storm development and |Dstmin|. For CIR and ICME events, no difference has been revealed between SMEaver and |Dstmin| in linear regression equations.

Studying auroral activity using the SME index at the magnetic storm main phase during CIR and ICME events

. In this paper, we examine the relationship of the SME index with magnetic storm characteristics and interplanetary medium parameters during the main phase of magnetic storms caused by CIR and ICME events. Over the period 1990–2017, 107 magnetic storms driven by (64) CIR and (43) ICME events have been selected. In contrast to AE and Kp, a stronger correlation is shown to exist between the average SME index (SMEaver) and interplanetary medium parameters during the magnetic storm main phase. Close correlation coefficients between SMEaver and the SW electric field (southward IMF Bz) have been obtained for CIR and ICME events. SMEaver has been found to increase with the rate of magnetic storm development and |Dstmin|. For CIR and ICME events, no difference has been revealed between SMEaver and |Dstmin| in linear regression equations.

An Object-based Climatology of Precipitation Systems in Sydney, Australia

The climate is warming and this is changing some aspects of storms, but we have relatively little knowledge of storm characteristics beyond intensity, which limits our understanding of storms overall. In this study, we apply a cell-tracking algorithm to 20 years of radar data at a mid-latitude coastal-site (Sydney, Australia), to establish a regional precipitation system climatology. The results show that extreme storms in terms of translation-speed, size and rainfall intensity usually occur in the warm season, and are slower and more intense over land between ~10am and ~8pm (AEST), peaking in the afternoon. Precipitation systems are more frequent in the cold season and often initiate over the ocean and move northward, leading to precipitation mostly over the ocean. Using clustering algorithms, we have found five precipitation system types with distinct properties, occurring throughout the year but peaking in different seasons. While overall rainfall statistics don't show any link to climate modes, links do appear for some system types using a multivariate approach. This climatology for a variety of precipitation system characteristics will allow future study of any changes in these characteristics due to climate change.

Biases in the Prediction of Convective Storm Characteristics with a Convection Allowing Ensemble

The severity, duration, and spatial extent of thunderstorm impacts is related to convective storm mode. This study assesses the skill of the High Resolution Rapid Refresh Ensemble (HRRR-E) and its deterministic counterpart (HRRRv4) at predicting convective mode and storm macrophysical properties using 35 convective events observed during the 2020 warm season across the eastern U.S. Seven cases were selected from each of five subjectively-determined convective organization modes: tropical cyclones, mesoscale convective systems (MCSs), quasi-linear convective systems, clusters, and cellular convection. These storm events were assessed using an object-based approach to identify convective storms and determine their individual size. Averaged across all 35 cases, both the HRRR-E and HRRRv4 predicted storm areas were generally larger than observed, with this bias being a function of storm lifetime and convective mode. Both modeling systems also under-predicted the rapid increase in storm counts during the initiation period, particularly for the smaller-scale storm modes. Interestingly, performance of the HRRRv4 differed from that of the HRRR-E, with the HRRRv4 generally having a larger bias in total storm area than the HRRR-E due to HRRRv4 predicting up to 66% more storm objects than the HRRR-E. The HRRR-E accurately predicted the convective mode 65% of the time, with complete misses being very rare (<5% of the time overall). However, an evaluation of rank histograms across all 35 cases revealed that the HRRR-E tended to be under-dispersive when predicting storm size for all but the MCS mode.

Environmental Evolution of Long-Lived Supercell Thunderstorms in the Great Plains

Long-lived supercells (containing mesocyclones persisting for at least 4 hours) are relatively rare, but present significant risk for society as a result of their intensity and associated hazards over an extended time period. The persistence of a rotating updraft is tied to near-storm environmental characteristics; however, given the established prevalence of mesoscale environmental heterogeneity near severe convection, it is unknown to what extent those near-storm characteristics vary over the lifetime of a supercell, nor how quickly the storm responds to such changes. This study examines 147 long-lived, isolated supercells, focusing on the evolution of their near-storm environments using model analysis soundings generated each hour throughout the storm’s lifetime. Environmental variability is quantified via a series of common forecasting parameters, with impacts of measured changes related to production of severe weather and overall storm longevity. The diurnal and maturity-relative distributions of forecasting parameters are examined, along with comparisons among subsets of marginally vs. very long-lived supercells, as well as dissipation before vs. after sunset. The diurnal cycle is a dominant trend over the lifetime of all supercells, with attendant impacts to relevant thermodynamic and kinematic parameters, timing of storm initiation and dissipation, as well as severe weather production. Notably, changes in the near-storm environment are connected to supercell longevity and generation of severe weather reports. The long-term goal of the above analyses is to enhance short-term forecasts of supercells by better anticipating storm evolution as a result of environmental variations.

A Radar Object-Based Examination of Rain System Climatology and Including Climate Variability

We know the climate is warming and this is changing some aspects of storms, but we have little knowledge of storm characteristics beyond intensity, which limits our understanding of storms overall. In this study, we apply a cell-tracking algorithm to 20 years of radar data at a mid-latitude coastal-site (Sydney, Australia), to establish a regional storm climatology. The results show that extreme storms in terms of translation-speed, size and rainfall intensity usually occur in the warm season, and are slower and more intense over land between ~10am and ~8pm (AEST), peaking in the afternoon. Storms are more frequent in the cold season and often initiate over the ocean and move northward, leading to precipitation mostly over the ocean. Using clustering algorithms, we have found five storm types with distinct properties, occurring throughout the year but peaking in different seasons. While overall rainfall statistics don't show any link to climate modes, links do appear for some storm types using a multivariate approach. This climatology for a variety of storm characteristics will allow future study of any changes in these characteristics due to climate change.

Wave Height Distributions and Rogue Waves in the Eastern Mediterranean

There is a lack of scientific knowledge about the physical sea characteristics of the eastern part of the Mediterranean Sea. The current work offers a comprehensive view of wave fields in southern Israel waters covering a period between January 2017 and June 2018. The analyzed data were collected by a meteorological buoy providing wind and waves parameters. As expected for this area, the strongest storm events occurred throughout October–April. In this paper, we analyze the buoy data following two main objectives—identifying the most appropriate statistical distribution model and examining wave data in search of rogue wave presence. The objectives were accomplished by comparing a number of models suitable for deep seawater waves. The Tayfun—Fedele 3rd order model showed the best agreement with the tail of the empirical wave heights distribution. Examination of different statistical thresholds for the identification of rogue waves resulted in the detection of 99 unique waves, all of relatively low height, except for one wave that reached 12.2 m in height which was detected during a powerful January 2018 storm. Characteristics of the detected rogue waves were examined, revealing the majority of them presenting crest to trough symmetry. This finding calls for a reevaluation of the crest amplitude being equal to or above 1.25 the significant wave height threshold which assumes rogue waves carry most of their energy in the crest.

Application of Global Dust Detection Index (GDDI) for Sand and Dust Storm Monitoring Over Kingdom of Saudi Arabia

Sand and dust storm events are frequent natural hazards in the Kingdom of Saudi Arabia. Sand and dust storm monitoring is therefore essential to mitigate their environmental-related issues. Satellite remote sensing has been successfully used for sand and dust storm monitoring in various parts of the world. In the current endeavor, we are applying the Global Dust Detection Index (GDDI) on Moderate Resolution Imaging Spectroradiometer (MODIS) data onboard Terra satellite to monitor sand and dust storm activities over the Kingdom of Saudi Arabia. In the current study, fourteen sand and dust storm events are analyzed between the years 2000 to 2017. The GDDI based results are validated by using MODIS combined Dark Target (DT) and Deep Blue (DB) Aerosol Optical Depth (AOD) product, Meteosat satellite images, ground-based meteorological stations data, and AOD data from AERONET (Aerosol Robotic Network) stations in the study area. Also, GDDI based results are analyzed by determining algorithm accuracy, Probability Of Correct positive Detection (POCD), and Probability Of False positive Detection (POFD). Results of the study show that GDDI can successfully identify sand and dust storm events with various threshold values over the Kingdom of Saudi Arabia. It is envisaged that the outcome of this study would be very beneficial to understand sand and dust storm characteristics in the study region.

Impact of SST and Surface Waves on Hurricane Florence (2018): A Coupled Modeling Investigation

Hurricane Florence (2018) devastated the coastal communities of the Carolinas through heavy rainfall that resulted in massive flooding. Florence was characterized by an abrupt reduction in intensity (Saffir-Simpson Category 4 to Category 1) just prior to landfall and synoptic-scale interactions that stalled the storm over the Carolinas for several days. We conducted a series of numerical modeling experiments in coupled and uncoupled configurations to examine the impact of sea surface temperature (SST) and ocean waves on storm characteristics. In addition to experiments using a fully coupled atmosphere-ocean-wave model, we introduced the capability of the atmospheric model to modulate wind stress and surface fluxes by oceanwaves through data from an uncoupled wave model. We examined these experiments by comparing track, intensity, strength, SST, storm structure, wave height, surface roughness, heat fluxes, and precipitation in order to determine the impacts of resolving ocean conditions with varying degrees of coupling. We found differences in the storm’s intensity and strength, with the best correlation coefficient of intensity (r=0.89) and strength (r=0.95) coming from the fully-coupled simulations. Further analysis into surface roughness parameterizations added to the atmospheric model revealed differences in the spatial distribution and magnitude of the largest roughness lengths. Adding ocean andwave features to the model further modified the fluxes due to more realistic cooling beneath the stormwhich in turn modified the precipitation field. Our experiments highlight significant differences in how air-sea processes impact hurricane modeling. The storm characteristics of track, intensity, strength, and precipitation at landfall are crucial to predictability and forecasting of future landfalling hurricanes.