Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
815
datasets available to search
ShareScore release 0.7.1
Dataset results
815 results for “Forecasting”
Multimodel AI Prediction Network (MAPNet) applied to temperature forecasts
<p>In this repository are all the data/scripts necessary to reproduce the results of the article 'Multimodel AI Prediction Network (MAPNet) applied to temperature forecasts.'</p> <p>The files in NetCDF format and nomenclature <strong>TYPE</strong>_<strong>FORECAST</strong>_202201_202312.nc are the input data for the neural network, where <strong>TYPE</strong> (TMAX, TMIN) and <strong>FORECAST</strong> (F24, F48, F72). These files contain all the predictions from the atmospheric models and the 'observations.' This input data was divided into training, validation, and test periods.</p> <p>The files in NetCDF format and nomenclature <strong>CNN</strong>_<strong>TYPE</strong>_<strong>FORECAST</strong>.nc are the output data of the neural network, where <strong>TYPE</strong> (TMAX, TMIN) and <strong>FORECAST</strong> (F24, F48, F72). These files contain the TMAX and TMIN forecasts for the test period.</p> <p>The files in NetCDF format and nomenclature <strong>MODEL</strong>_<strong>TYPE</strong>_<strong>FORECAST</strong>.nc are the forecasts from the atmospheric models (BRAMS, ETA, WRF, SMEC) and the observations (SAMET) for the same test period. This data was used to compare the predictions of each model with the predictions of the neural network.</p> <p>The file CCN_Unet.py is a Python program containing the configurations of the convolutional neural network with a U-Net architecture used in the study.</p> <p>The files in ASCII format (.py, .gs) are scripts used for visualizing the results and generating the graphs. The other files are shapefiles and masks used for generating figures.</p>
Weather data (forecast and observation) at 48 locations in France for beginning of 2024 for Machine Learning Training
<p>The data provided data are historical weather measurement and forecast at 48 locations in France and its boundary.</p> <p>Measurements are inside files named MES_YYYY.csv with YYYY is the id code of the station.</p> <p>The file "Station_list.csv" contains the list of the 45 locations with the id code, the name and then the latitude and longitude.</p> <p><br>Forecasts are inside files named XXX_YYYY.csv with YYYY the id code corresponding of the location of the grid ouput close to the associated observation location.<br>XXX is the id of the numerical forecast:<br> "GFS0.25-Complet" for GFS file at 0.25° resolution<br> "LEXIS" for WRF produced by EVEREST project using the LEXIS chain<br> "WRF3KM-Complet" for WRF at 3km resolution produced by NUMTECH<br> "WRF12KM-Complet" for WRF at 12km resolution produced by NUMTECH</p> <p><br>Description of MES-YYYY files:<br>- One line per measurement with hourly resolution<br>- columns are: Date(TU),Temperature2m_degC,WindSpeed10m_m/s,WindDirection10m_m/s<br> Date = date of measurement in TU and format DD/MM/YYYY HH:MM<br> Temperature2m_degC = air temperature at 2m height in °Celsius<br> WindSpeed10m_m/s = wind speed at 10m height in m/s<br> WindDirection10m_deg = wind direction at 10m height in deg. (0 or 360 = wind from north to south, 45°=wind from east to east, ....)<br>If measurement is not available for a specific hour for one parameter, the value "-999" is used.</p> <p>The observation data gocfrom 28/01/2024 00HTU to 17/03/2024 23HTU</p> <p><br>Description of XXX_YYYY forecast files:<br>- One line per forecast with hourly resolution<br>- columns are: First date run (TU),Forecast date,Temperature2m_degC,WindSpeed10m_m/s,WindDirection10m_m/s<br> First date run (TU) = date of start of the forecast in TU and format DD/MM/YYYY HH:MM. HH could be 00 and 12 according to the cycle of forecast start.<br> Forecast date = date of the forecast in TU and format DD/MM/YYYY HH:MM. HH go from 00 to 23. <br> Temperature2m_degC = air temperature at 2m height in °Celsius<br> WindSpeed10m_m/s = wind speed at 10m height in m/s<br> WindDirection10m_deg = wind direction at 10m height in deg. (0 or 360 = wind from north to south, 45°=wind from east to east, ....)<br>If forecast is not available for a specific hour for one parameter, the value "-999" is used.</p> <p>The forecast data go from 28/01/2024 00HTU to 17/03/2024 23HTU</p>
Outputs of the Jupyter Notebook - Sea ice forecasting using the IceNet Library
<p>The dataset contains the outputs of the notebook "Sea ice forecasting using the IceNet library" published in The Environmental Data Science Book.</p> <p><strong>Contributions</strong></p> <p><em>Notebook</em></p> <ul> <li>James Byrne (author), British Antarctic Survey, <a href="https://github.com/JimCircadian">@JimCircadian</a></li> <li>Bryn Noel Ubald (author), British Antarctic Survey, <a href="https://github.com/tom-andersson">@tom-andersson</a></li> <li>Wei Ji (reviewer), Development Seed, <a href="https://github.com/weiji14">@weiji14</a></li> <li>William Gregory (reviewer), Princeton University, <a href="https://github.com/William-gregory">@William-gregory</a></li> <li>Anne Fouilloux (editor), Simula Research Laboratory, <a href="https://github.com/annefou">@annefou</a></li> </ul> <p><em>Modelling codebase</em></p> <ul> <li>James Byrne (Code author)</li> <li>Tom Andersson (Science author)</li> <li>Bryn Noel Ubald (Code maintainer and contributor)</li> </ul>
Seasonal footprints on ecological time series and jumps in dynamic states of protein configurations from a nonlinear forecasting method characterization (DataSet)
<p>Data from the article: <br>"Seasonal footprints on ecological time series and jumps in dynamic states of protein configurations from a nonlinear forecasting method characterization", L. Reyes, K. Campos, G. D. Avendaño, L. González-Paz, A. Vivas, Y. J. Alvarado, and S. Flores.</p> <p>Data to be used with some implementation of the forecasting method of reference:<br>Sugihara G. and May R. M., Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series, <em>Nature</em> <strong>344</strong>, 734–741 (1990).</p>
Supplementary materials for "Forecasting Soccer Matches With Betting Odds: A Tale of Two Markets"
<p>Replication package for International Journal of Forecasting article "Forecasting Soccer Matches With Betting Odds: A Tale of Two Markets".</p>
Smogseer: A Convolutional LSTM for forecasting air quality from Sentinel-5P data
<h1><strong>Training and checkpoint datasets </strong></h1> <h2><a href="#h_1501865622611722161313186" target="_blank" rel="noopener">S5PL2_5D.nc</a></h2> <p>This is the Sentinel-5P traning dataset for the Smogseer ConvLSTM model. The dataset was created using the xcube Sentinel Hub data store from the <a href="https://deepesdl.readthedocs.io/en/latest/datasets/datastores/#xcube-sentinel-hub-data-store">Deep Earth System Data Lab</a>.</p> <p>The dataset has the following characteristics:</p> <ul> <li>bbox= [68.137207,24.886436,84.836426,34.379713] #WGS84 // lon,lat,lon,lat</li> <li>res = (bbox[2]-bbox[0])/512 # ~3629m</li> <li>date_range = ['2019-01-01', '2023-12-31']</li> <li> <div>timesteps = '5D'</div> </li> </ul> <h2><a href="#h_4062378263291722161323566">X_val.npy</a></h2> <p>Validation feature data with shape: (74, 1, 291, 512, 6)</p> <ul> <li>74: dates</li> <li>1: time steps</li> <li>291: latitudes</li> <li>512: longitudes</li> <li>6: Features ['SO2', 'NO2', 'CH4', 'O3', 'CO', 'HCHO']</li> </ul> <h2><a href="#h_9861938573981722161330905">Y_val.npy</a></h2> <p>Validation target data with shape: (74, 1, 291, 512, 1)</p> <ul> <li>74: dates</li> <li>1: time steps</li> <li>291: latitudes</li> <li>512: longitudes</li> <li>6: Features ['SO2', 'NO2', 'CH4', 'O3', 'CO', 'HCHO']</li> </ul> <h2><a href="#h_321677095651722161353847">smogseer50.keras</a></h2> <p>Model weights for training the ConvLSTM with 50 epochs.</p> <h2><a href="#h_5931467847131722161401921">smogseer100.keras</a></h2> <p>Model weights for training the ConvLSTM with 10 epochs.</p>
Preprocessed EUMETSAT H-SAF h61 Satellite and ECMWF HRES 24h Forecast Precipitation Datasets for Hydrometeorological Applications over Central Europe
<p>This dataset contains preprocessed precipitation data from the ECMWF's high resolution HRES forecast (24h) and EUMETSAT's blended infrared and microwave remotely sensed data for use in hydrological and meteorological research. <br><br><em>* Preprocessing procedure and codes are accessible <a href="https://gitlab.jsc.fz-juelich.de/kiste/atmoscorrect/-/blob/master/HRES_PP.ipynb?ref_type=heads">here for HRES</a>, and <a href="https://gitlab.jsc.fz-juelich.de/kiste/atmoscorrect/-/blob/master/HSAF_PP.ipynb?ref_type=heads">here for H-SAF</a> datasets.</em></p> <p><strong>HRES Data (HRES_pr.nc):</strong></p> <ul> <li><strong>Source:</strong> <a href="https://confluence.ecmwf.int/display/FUG/Section+2.1.2.4+HRES+-+High+Resolution+Forecasts">ECMWF's high-resolution, deterministic HRES 24-hour precipitation forecast at 12UTC.</a></li> <li><strong>Resolution and domain:</strong> 0.1° × 0.1° grid in (longmin: -1.1, longmax: 18.4, latmin: 44.1, latmax: 56.5)</li> <li><strong>Preprocessing Steps:</strong> <ol> <li>Extracted the precipitation variable (tp) out of variables.</li> <li>Converted precipitation units from meters (m) to millimeters (mm).</li> <li>Changed cumulative precipitation to instantaneous.</li> <li>Selected the first 24 hours of forecast data from the available 90-hour forecasts.</li> <li>Merged all processed files into a single NetCDF file.</li> </ol> </li> </ul> <p><strong>H-SAF Data (HSAF_pr.nc):</strong></p> <ul> <li><strong>Source:</strong> <a href="https://hsaf.meteoam.it/Products/Detail?prod=H61B">EUMETSAT's H-SAF h61B</a></li> <li><strong>Resolution and domain:</strong> Original product: ~4.8 km at nadir, ~8km in Europe; preprocessed product: resampled to 0.1° × 0.1° (~10 km) grid in (longmin: -1.1, longmax: 18.4, latmin: 44.1, latmax: 56.5).</li> <li><strong>Preprocessing Steps:</strong> <ol> <li>Trimmed the MSG coverage data to cover the study domain.</li> <li>Calculated the grid correspondance using lat/lon information from MSG grid using a <a href="https://www-cdn.eumetsat.int/files/2020-04/pdf_conf_2018_s1_mueller_p.pdf">reference method</a>.</li> <li>Merged all processed files into a single NetCDF file.</li> <li>Regridded the data to the 0.1° × 0.1° resolution using bilinear function in <a href="https://code.mpimet.mpg.de/projects/cdo">CDO</a></li> </ol> </li> </ul> <p><strong>Data Period:</strong> 01/07/2020-25/04/2023</p>
Data and software from: Decomposition of skill scores for conditional verification – Impact of AMO phases on the predictability of decadal temperature forecasts
<p>Data and software from "Decomposition of skill scores for conditional verification – Impact of AMO phases on the predictability of decadal temperature forecasts"</p> <ul> <li>synthetic data and score computation routine</li> <li>data from MiKlip decadal prediction systems preop-dcpp-HR, preop-LR and observational HadCRUT4 data used for the conditional verification</li> <li>AMO calculation routine</li> <li>AMO data and phases defining the ocean state used in the conditional verification</li> </ul> <p>The computation routines from the conditional verification of decadal predcition systems is publicly available as the Freva plug-in "ProblEMS" at <a href="https://doi.org/10.5281/zenodo.10469658">https://doi.org/10.5281/zenodo.10469658</a></p> <p> </p>
Arno River flooding 1966 | WRF-ENS forecasts
<p>Contains WRF-ENS forecasts for the Arno River flooding event occurred in Italy in November 1966. Files format is binary with auxiliary description file.</p>
NAM forecasts for Folsom, CA (lat=38.642N, lon=121.148W)
<p>This repository contains 4 comma-separated values files with Numerical Weather Prediction (NWP) forecasts for the 4 closest nodes to Folsom, CA (lat=38.642N, lon=121.148W).</p> <p>The data in the files was collected from the North American Mesoscale Forecast System (NAM) for the 12Z forecast cycle.</p> <p>Each file corresponds to one of the 4 nearest nodes to lat=38.642N, lon=121.148W. The lat,lon for each file is given in the file name and also in the first line of the file.</p> <p>Each file contains 10 columns:</p> <ol> <li>reftime: the timestamp (UTC) for the forecast creation</li> <li>valtime: the timestamp (UTC) for which the values are valid</li> <li>dwsw: downward short-wave radiation flux or GHI in W/m<sup>2</sup></li> <li>cloud_cover: total cloud cover in %</li> <li>precipitation: total precipitation in kg/m<sup>2</sup></li> <li>pressure: Surface pressure in Pa</li> <li>wind-u: U-Component of Wind 10 m above ground in m/s</li> <li>wind-v: V-Component of Wind 10 m above ground in m/s</li> <li>temperature: Surface temperature in K</li> <li>rel_humidity: Relative humidity 2 m above ground in %</li> </ol> <p> </p> <p> </p>
IMPREX - TUC - Local scale seasonal hydrological and drought indices forecasts
<p>The compressed files are seasonal forecasts of discharge based on ECMWF System4 (1981-2003) and GLOSEA5 (1996-2003) for the Koutsoulidis catchment, Crete, Greece. The location of the Koutsoulidis catchment as well as the methodology of data development and validation are described in the open access publication by Grillakis et al., 2018 (<a href="https://www.mdpi.com/2073-4441/10/11/1593">https://www.mdpi.com/2073-4441/10/11/1593</a>). Seasonal forecasts of Standardized Precipitation Index (SPI) based on ECMWF System4 (1981-2009) and GLOSEA5 (1996-2009) for the Messara catchment are also included. A description of the study site and methodology is included in the poster attached in the compressed file (EGU201712072-Koutroulis).</p>
Molecular Forecaster Inc (MFI): USP5 Zf-UBD
<p>Self-docking of ligands in USP5 zinc finger ubiquitin binding domain (Zf-UBD) co-crystal structures using the Forecaster FITTED docking platform is assessed and the ranking success of FITTED rigid protein docking vs. MATCHUP flexible protein docking on a library of experimentally tested compounds is investigated. </p>
Videos for "Weather and climate forecasting with neural networks: using GCMs with different complexity as study-ground"
<p>Supplementary videos for the paper "Weather and climate forecasting with neural networks: using GCMs with different complexity as study-ground" by S. Scher and G. Messori, Geoscientific Model Development 2019</p>
Data for journal article: "Improving medium-range forecasts of rain-on-snow events in pre-alpine areas"
<p>This dataset contains scripts and data in R software environment format used for the journal article "Improving medium-range forecasts of rain-on-snow events in pre-alpine areas" submitted to Water Resources Research.</p>
Supplementary Data for Massively Parallel Implicit Equal-Weights Particle Filter for Ocean Drift Trajectory Forecasting
<p>This data repository is provided as a supplement to the paper *Massively Parallel Implicit Equal-Weights Particle Filter for Ocean Drift Trajectory Forecasting* written by Håvard Heitlo Holm, Martin Lilleeng Sætra and Peter Jan van Leeuwen. It contains the complete datasets (initial conditions and results of the ensemble simulations) obtained from the experiments presented therein.</p> <p>This data set is generated by, and can be further post-processed and visualized by, the code published as *metno/gpu-ocean: Supplementary Software for Massively Parallel Implicit Equal-Weights Particle Filter for Ocean Drift Trajectory Forecasting* by Håvard Heitlo Holm, Martin Lilleeng Sætra and André Rigland Brodtkorb (DOI 10.5281/zenodo.3458291). </p> <p> </p>
Figure 1 in The potential effects of future climate change on suitable habitat for the Taiwan partridge (Arborophila crudigularis): an ensemble-based forecasting method
Figure 1. Modeled range and presence records for Arborophila crudigularis.
Unveiling the Future Water Pulse of Central Asia: A Comprehensive 21st Century Hydrological Forecast from Stochastic Water Balance Modeling
<p>This dataset and the scripts accompany the manuscript "<strong>Unveiling the Future Water Pulse of Central Asia: A Comprehensive 21st Century Hydrological Forecast from Stochastic Water Balance Modeling</strong>". The manuscript is published in the Journal Climatic Change.</p>
Supporting data ocean model GMD submission: From Weather Data to River Runoff: Leveraging Spatiotemporal Convolutional Networks for Comprehensive Discharge Forecasting
<p>Ocean model salinity data used for the comparison of the ConvLSTM river runoff model and the original E-HYPE based model simulations.</p>
Forecasting model of seasonal dynamics of boll weevil Anthonomus grandis grandis (Coleoptera: Curculionidae) in cotton crops using artificial neural networks
Open the record for dataset details and reuse information.
Dataset - "An Adaptive Multi-Seasonal ARIMA Approach for Domestic Hot Water Load Forecasting: A Pilot Study"
<p>The description of the data is in README.txt.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.