Weather Forecasting

The nuScenes dataset is a large-scale autonomous driving dataset. The dataset has 3D bounding boxes for 1000 scenes collected in Boston and Singapore. Each scene is 20 seconds long and annotated at 2Hz. This results in a total of 28130 samples for training, 6019 samples for validation and 6008 samples for testing. The dataset has the full autonomous vehicle data suite: 32-beam LiDAR, 6 cameras and radars with complete 360° coverage. The 3D object detection challenge evaluates the performance on 10 classes: cars, trucks, buses, trailers, construction vehicles, pedestrians, motorcycles, bicycles, traffic cones and barriers. Source: PointPainting: Sequential Fusion for 3D Object Detection

Audi Autonomous Driving Dataset (A2D2) consists of simultaneously recorded images and 3D point clouds, together with 3D bounding boxes, semantic segmentation, instance segmentation, and data extracted from the automotive bus. Source: A2D2: Audi Autonomous Driving Dataset Image Source: https://www.a2d2.audi/a2d2/en/dataset.html

The Shifts Dataset is a dataset for evaluation of uncertainty estimates and robustness to distributional shift. The dataset, which has been collected from industrial sources and services, is composed of three tasks, with each corresponding to a particular data modality: tabular weather prediction, machine translation, and self-driving car (SDC) vehicle motion prediction. All of these data modalities and tasks are affected by real, `in-the-wild' distributional shifts and pose interesting challenges with respect to uncertainty estimation.

SEVIR is an annotated, curated and spatio-temporally aligned dataset containing over 10,000 weather events that each consist of 384 km x 384 km image sequences spanning 4 hours of time. Images in SEVIR were sampled and aligned across five different data types: three channels (C02, C09, C13) from the GOES-16 advanced baseline imager, NEXRAD vertically integrated liquid mosaics, and GOES-16 Geostationary Lightning Mapper (GLM) flashes. Many events in SEVIR were selected and matched to the NOAA Storm Events database so that additional descriptive information such as storm impacts and storm descriptions can be linked to the rich imagery provided by the sensors. Source: https://proceedings.neurips.cc//paper/2020/file/fa78a16157fed00d7a80515818432169-Paper.pdf

The Oxford Radar RobotCar Dataset is a radar extension to The Oxford RobotCar Dataset. It has been extended with data from a Navtech CTS350-X Millimetre-Wave FMCW radar and Dual Velodyne HDL-32E LIDARs with optimised ground truth radar odometry for 280 km of driving around Oxford, UK (in addition to all sensors in the original Oxford RobotCar Dataset). Source: The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset

Encourages machine learning research in this area and to help facilitate further work in understanding and mitigating the effects of climate change. Source: ExtremeWeather: A large-scale climate dataset for semi-supervised detection, localization, and understanding of extreme weather events

A benchmark dataset for data-driven medium-range weather forecasting, a topic of high scientific interest for atmospheric and computer scientists alike. Source: WeatherBench: A benchmark dataset for data-driven weather forecasting

This dataset contains meteorological observations (temperature) at the land-based weather stations located in the United States, collected from the Online Climate Data Directory of the National Oceanic and Atmospheric Administration (NOAA). The weather stations are sampled from the Western and Southeastern states that have actively measured meteorological observations during 2015. The 1-year sequential data of hourly temperature records are divided into small sequences of 24 hours. For training, validation, and test a sequential 8-2-2 (months) split is used.

A multivariate spatio-temporal benchmark dataset for meteorological forecasting based on real-time observation data from ground weather stations.

Floods are among the most common and devastating natural hazards, imposing immense costs on our society and economy due to their disastrous consequences. Recent progress in weather prediction and spaceborne flood mapping demonstrated the feasibility of anticipating extreme events and reliably detecting their catas- trophic effects afterwards. However, these efforts are rarely linked to one another and there is a critical lack of datasets and benchmarks to enable the direct forecast- ing of flood extent. To resolve this issue, we curate a novel dataset enabling a timely prediction of flood extent. Furthermore, we provide a representative evaluation of state-of-the-art methods, structured into two benchmark tracks for forecasting flood inundation maps i) in general and ii) focused on coastal regions. Altogether, our dataset and benchmark provide a comprehensive platform for evaluating flood forecasts, enabling future solutions for this critical challenge. Data, code & models are shared at https://github.com/Multihuntr/GFF under a CC0 license.

IowaRain is a dataset of rainfall events for the state of Iowa (2016-2019) acquired from the National Weather Service Next Generation Weather Radar (NEXRAD) system and processed by a quantitative precipitation estimation system. The dataset presented in this study could be used for better disaster monitoring, response and recovery by paving the way for both predictive and prescriptive modeling

A new spatio-temporal benchmark dataset (Hurricane), is suited for forecasting during extreme events and anomalies. The dataset is provided through the Florida Department of Revenue which provides the monthly sales revenue (2003-2020) for the tourism industry for all 67 counties of Florida which are prone to annual hurricanes. Furthermore, we aligned and joined the raw time series with the history of hurricane categories (i.e., event intensities) based on time for each county. Note that the hurricane category indicates the maximum sustained wind speed which can result in catastrophic damages as this number goes up (Category 1-6).

The Automated Surface Observing Systems (ASOS) program is a joint effort of the National Weather Service (NWS), the Federal Aviation Administration (FAA), and the Department of Defense (DOD). These automated systems collect observations on a continual basis, 24 hours a day. Automated Weather Observing System (AWOS) units are operated and controlled by the Federal Aviation Administration. These systems are among the oldest automated weather stations and predate ASOS. They generally report at 20-minute intervals and, unlike ASOS, do not report special observations for rapidly changing weather conditions. ASOS observations are operationally generated each hour, and special observations are provided whenever the weather changes. These special reports are generated when conditions exceed preselected weather element thresholds, e.g., the visibility decreases to less than 3 miles.

A satellite-based dataset called "CloudCast". It consists of 70080 images with 10 different cloud types for multiple layers of the atmosphere annotated on a pixel level. The spatial resolution of the dataset is 928 × 1530 pixels (3 × 3 km per pixel) with 15-min intervals between frames for the period January 1, 2017, to December 31, 2018. All frames are centered and projected over Europe. Source: CloudCast: A Satellite-Based Dataset and Baseline for Forecasting Clouds

DIT4BEARs Internship Project (at UiT-The Arctic University of Norway) Dataset The dataset contains data of 5 months including weather conditions, friction coefficient, distance traveled, wind speed, surface temperature, air temperature, etc. This dataset was provided by DIT4BEARs for the Smart Road Internship Project at UiT-The Arctic University of Norway It can be used for weather forecasting, road friction forecasting, smart road proposition, minimization of accident rate in Barren Euro-Arctic regions.

The model forecasts for the sub-seasonal forecasting application considered in the Online Learning under Optimism and Delay paper experiments. This dataset consists of a single ZIP archive (919MB) that contains 1) a "models" folder that contains, for each model the forecasts for the Precip. 3-4w, Precip. 5-6w, Temp. 3-4w, Temp. 5-6w tasks on the western United States geography, and 2) a "data" folder that contains supporting geographic data. The data should be used to reproduce the PoolD experiments in https://github.com/geflaspohler/poold as described in the README. (2021-06-10)

The Tornado Network (TorNet) dataset is a large, high-resolution benchmark dataset developed to support machine learning research in tornado detection and prediction. It comprises over 200,000 radar samples derived from 9 years of full-resolution, polarimetric WSR-88D (NEXRAD) level-II and level-III radar data. Each sample, called a "chip," includes multiple radar variables—such as reflectivity, radial velocity, spectrum width, differential reflectivity, correlation coefficient, and specific differential phase—captured across two elevation angles and four time steps spaced five minutes apart. Rather than converting radar data to Cartesian coordinates, TorNet retains its native polar format, preserving spatial fidelity near the radar site. This level of detail enables the dataset to support a wide range of machine learning techniques, including deep learning models that can learn directly from raw radar imagery.