Abstract: Method, apparatus, and computer program product are provided for generating synthetic bathymetry using one or more fractal techniques. In some embodiments, real terrain data generated from sensor-based measurement of terrain associated with but outside a water body are received. This data may include, for example, real terrain data generated from sensor-based measurement (e.g., LiDAR) of at least a portion of an area proximate the water body. One or more fractal dimensions is/are extracted from the real terrain data, and synthetic bathymetry of the water body is generated based on the one or more fractal dimensions using one or more fractal terrain generation techniques. In some embodiments, one or more simulated geological processes is/are applied to the synthetic bathymetry. For example, a simulated sedimentation process and/or a simulated erosion process may be applied to soften the synthetic bathymetry.

Abstract: A method, computer program product and computer system to generate safety alerts is provided. A processor retrieves a plurality of measurements associated with a location. A processor determines a set of features based on the plurality of measurements. A processor identifies a set of membership functions for the set of features. A processor determines a safety index for the body of water based on the set of membership functions and one or more input value ranges for the set of features. In response to the safety index being above a threshold value, a processor sends an alert to one or more users regarding the location.

Abstract: Train a machine learning model, using an image-based knowledge graph of tropical cyclone data, for implementing a surface field modeling architecture that produces images of at least surface wind fields and surface rainfall fields from images of at least tropical cyclone tracks and pressure intensities. Generate model images of a modeled surface wind field and a modeled surface rainfall field by providing images of at least a user-generated tropical cyclone track and pressure intensity to the trained machine learning model.

Abstract: A method, computer system, and a computer program for weather prediction are provided. The method may include receiving a first weather event associated with a first location. The present invention may further include inputting the first weather event into a machine learning model generated via mapping historical weather data into a latent space and via identifying, in the latent space, climate teleconnections amongst historical weather events at various locations. The method may further include in response to the inputting, receiving by the computing device a weather prediction for a second location, the weather prediction being based on a predicted climate teleconnection between the first location and the second location with respect to the first weather event, wherein the teleconnections machine learning model maps the first weather event into latent code for the latent space in order to generate the weather prediction for the second location.

Abstract: A method, computer system, and a computer program for generating weather simulations based on significant events is provided. The present invention may include receiving a plurality of environmental data associated with a geographic area. The present invention may then include retrieving one or more data feeds including a plurality of events. The present invention may further include associating the plurality of events with a subset of the plurality of environmental data. The present invention may further include generating a knowledge graph for the geographic area based on the association.

Abstract: A processor may receive weather event data. The processor may determine, utilizing an artificial intelligence model mapping weather events having weather impacts with a higher likelihood of occurrence proximate to each other in the latent space, a weather impact associated with a weather event. In some embodiments, the artificial intelligence model may be trained using historical weather event data and historical weather impact data associated with the historical weather event data. The processor may output the weather impact associated with the weather event to a user.

Abstract: Initial and boundary conditions, and parameters associated with geophysical modeling can be received. Based on the received initial and boundary conditions and parameters, a multiscale model can be trained for data generation to produce first resolution simulation data and second resolution simulation data for a surrogate machine learning model training, where the second resolution simulation data has higher resolution than the first resolution simulation data. A surrogate model can be created using neural operators, where the surrogate model is trained using the first resolution simulation data and second resolution simulation data. An operational forecasting model can be generated using the surrogate model.

Abstract: A method and system for outputting a state of a physical system using a calibrated model of the physical system, where the calibrated model is used to generate a model prediction. The system includes a plurality of sensors connected to a routing node are used to monitor measured data of the physical system. A first sensor of the plurality of sensors includes a logic module configured to determine an uncertainty quantification, and to combine the uncertainty quantification with the model prediction to output the state of the physical system.

Abstract: Classifying land use by receiving geographic data and land use data for a geographic area, receiving surface temperature data for the geographic area, mapping the geographic data and temperature data to a set of map grid cells, determining temperature statistics for each map grid cell, training a machine learning model according to the land use data and temperature statistics, and classifying land use for map grid cells of a different geographic area according to the machine learning model.

Abstract: A computer implemented method of predictive weather occurrences includes generating, by a computer processor, a training model through artificial intelligence. The training model is based on climate data processed by a variational autoencoder. A geographic location is selected for climate study. Historical weather measurements associated with the selected geographic location are retrieved from a knowledge climate database. The retrieved historical weather measurements are processed using the training model. The training model receives threshold parameters defining extremeness of weather. Extremeness is based on a weather intensity data point being farther from a norm than closer to the norm. Synthetic weather data is generated for the selected location, wherein the synthetic weather data predicts weather events satisfying the extremeness threshold parameters.

Abstract: Train a machine learning model, using an image-based knowledge graph of tropical cyclone data, for implementing a surface field modeling architecture that produces images of at least surface wind fields and surface rainfall fields from images of at least tropical cyclone tracks and pressure intensities. Generate model images of a modeled surface wind field and a modeled surface rainfall field by providing images of at least a user-generated tropical cyclone track and pressure intensity to the trained machine learning model.

Abstract: A method, a computer system, and a computer program product for regionalized climate models is provided. Embodiments of the present invention may include selecting a class of a reduced order model. Embodiments of the present invention may include building a neural network in a reduced order space. Embodiments of the present invention may include recovering full state dynamics. Embodiments of the present invention may include training a model. Embodiments of the present invention may include providing an output.

Abstract: Method, apparatus, and computer program product are provided for generating synthetic bathymetry using one or more fractal techniques. In some embodiments, real terrain data generated from sensor-based measurement of terrain associated with but outside a water body are received. This data may include, for example, real terrain data generated from sensor-based measurement (e.g., LiDAR) of at least a portion of an area proximate the water body. One or more fractal dimensions is/are extracted from the real terrain data, and synthetic bathymetry of the water body is generated based on the one or more fractal dimensions using one or more fractal terrain generation techniques. In some embodiments, one or more simulated geological processes is/are applied to the synthetic bathymetry. For example, a simulated sedimentation process and/or a simulated erosion process may be applied to soften the synthetic bathymetry.

Abstract: A computer implemented method of predicting ice coverage on a body of water includes generating a first ice cover prediction with a thermodynamics module and generating a second ice cover prediction with a machine learning module. The first ice cover prediction is combined with the second ice cover prediction to generate a combined ice cover prediction. Error statistics are computed based on a comparison of the combined ice cover prediction with an ice coverage observation and the combined ice cover prediction is updated based on the error statistics.

Abstract: Method, apparatus, and computer program product are provided for assessing road surface condition. In some embodiments, candidate locations each forecast to have a dangerous road surface condition are determined, an optimized flight path is determined comprising a sequence of sites corresponding to the candidate locations, dispatch is made to a first site within the sequence, and a road surface condition at the first site is assessed using an onboard sensor (e.g., spectroradiometer). In some embodiments, a check for new information is performed before dispatch is made to a second site. In some embodiments, the candidate locations are determined using both a model forecast and data-mined locations considered hazardous. In some embodiments, the optimized flight path is determined using TSP optimization constrained by available flight time and prioritized by frequency of historical incident and severity of forecast road surface condition.

Abstract: Classifying land use by receiving geographic data and land use data for a geographic area, receiving surface temperature data for the geographic area, mapping the geographic data and temperature data to a set of map grid cells, determining temperature statistics for each map grid cell, training a machine learning model according to the land use data and temperature statistics, and classifying land use for map grid cells of a different geographic area according to the machine learning model.

Abstract: A method, a computer program product, and a computer system notify passengers of an imminent brake event of a vehicle. The method includes determining that the imminent brake event is about to occur. The method includes determining a delay value to be applied to a brake system of the vehicle. The method includes determining a safety threshold associated with the imminent brake event based on at least one of conditions of the vehicle and surroundings of the vehicle. As a result of the delay value being less than the safety threshold, the method includes postponing use of the brake system for a duration corresponding to the delay value. The method includes broadcasting a notification to the passengers of the imminent brake event.

Abstract: Road condition prediction for potentially hazardous road segments is described. For roadways that may contain accumulated frozen precipitation, a road segment for road condition prediction is selected based on weather conditions. Various models including a solar radiation budget model, a permanent structures model, a dynamic structures model, and a road condition model are generated for the selected road segment and account for shading effects on the road segment caused by objects near the road segment. A road condition prediction for hazardous conditions on the road segment is determined based on the road condition model and provided to a driver to alert the driver of any potentially hazardous conditions.

Abstract: A simulation model recovery method, system, and computer program product include initiating a simulation model, during an operation of a model, periodically writing a solution space of the model to a checkpoint restart file, during an operation of the model, periodically writing diagnostic information on model progression to a log file, detecting a failure of the model, based on the log of the model, determining a time of the failure, based on the model outputs and restart files, determining a period of a numerical instability preceding the failure, selecting a checkpoint of the model preceding the period of the numerical instability, based on the numerical instability and diagnostic information in log files, modifying a configuration of the model, and restarting the model based on the selected checkpoint and the modified configuration.

Abstract: A system and method to augment model-to-model coupling include obtaining an output signal fs from an upstream model. The method also includes obtaining an observation signal gob from a region of interest, extracting a high-frequency signal g from the observation signal gob using a linear operator, and providing to a downstream model the high-frequency signal g in addition to information based on the output signal fs from the upstream model. The downstream model is implemented to obtain a prediction or estimation of one or more parameters of interest or drive a mechanical process.