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.
10
datasets available to search
ShareScore release 0.9.0
Dataset results
10 results for “perturbed parameter ensembles”
Example Perturbed Parameter Ensemble (Black Carbon)
<p>This dataset contains the parameter design and example ECHAM-HAM output from the AeroCom Black Carbon (BC) multi-model Perturbed Parameter Ensemble (PPE) experiment described here: https://wiki.met.no/aerocom/phase3-experiments#multi-model_ppe_bc_experiment</p>
PPMLES – Perturbed-Parameter ensemble of MUST Large-Eddy Simulations
<h2>Dataset description</h2> <p>This repository contains the PPMLES (Perturbed-Parameter ensemble of MUST Large-Eddy Simulations) dataset, which corresponds to the main outputs of 200 large-eddy simulations (LES) of microscale pollutant dispersion that replicate the MUST field experiment [Biltoft. 2001, Yee and Biltoft. 2004] for varying meteorological forcing parameters.</p> <p>The goal of the PPMLES dataset is to provide a comprehensive dataset to better understand the complex interactions between the atmospheric boundary layer (ABL), the urban environment, and pollutant dispersion. It was originally used to assess the impact of the meteorological uncertainty on microscale pollutant prediction and to build a surrogate model that can replace the costly LES model [Lumet et al. 2025]. The total computational cost of the PPMLES dataset is estimated to be about 6 million core hours.</p> <p>For each sample of meteorological forcing parameters (inlet wind direction and friction velocity), the <a href="https://www.cerfacs.fr/avbp7x/">AVBP</a> solver code [Schonfeld and Rudgyard. 1999, Gicquel et al. 2011] was used to perform LES at very high spatio-temporal resolution (1e-3s time step, 30cm discretization length) to provide a fine representation of the pollutant concentration and wind velocity statistics within the urban-like canopy. The total computational cost of the PPMLES dataset is estimated to be about 6 million core hours.</p> <h2>File list</h2> <p>The data is stored in <a href="https://www.hdfgroup.org/solutions/hdf5/">HDF5</a> files, which can be efficiently processed in Python using the <a href="https://docs.h5py.org/en/stable/index.html">h5py</a> module. </p> <ul> <li><em>input_parameters.h5:</em> list of the 200 input parameter samples<em> (alpha_inlet, ustar) </em>obtained using the Halton sequence that defines the PPMLES ensemble.</li> <li><em>ave_fields.h5</em>: lists of the main field statistics predicted by each of the 200 LES samples over the 200-s reference window [Yee and Biltoft. 2004], including: <ul> <li><em>c:</em> the time-averaged pollutant concentration in ppmv <em>(dim = (n_samples, n_nodes) = (200, 1878585))</em>, </li> <li><em>(u, v, w): </em>the time-averaged wind velocity components in m/s,</li> <li><em>crms: </em>the root mean square concentration fluctuations in ppmv, </li> <li><em>tke:</em> the turbulent kinetic energy in m^2/s^2,</li> <li><em>(uprim_cprim, vprim_cprim, wprim_cprim)</em>: the pollutant turbulent transport components</li> </ul> </li> <li><em>uncertainty.h5</em>: lists of the estimated aleatory uncertainty induced by the internal variability of the LES <em>(variability_#)</em> [Lumet et al. 2024] for each of the fields in <em>ave_fields.h5</em>. Also includes the stationary bootstrap [Politis and Romano. 1994] parameters <em>(n_replicates, block_length) </em>used to estimate the uncertainty for each field and each sample.</li> <li><em>mesh.h5</em>: the tetrahedral mesh on which the fields are discretized, composed of about 1.8 millions of nodes.</li> <li><em>time_series.h5</em>: HDF5 file consisting of 200 groups (<em>Sample_NNN</em>) each containing the time series of the pollutant concentration (c) and wind velocity components (u, v, w) predicted by the LES sample #NNN at 93 locations. </li> <li><em>probe_network.dat</em>: provides the location of each of the 93 probes corresponding to the positions of the experimental campaign sensors [Biltoft. 2001].</li> </ul> <p><strong>Warning:</strong> the propylene concentration are expressed in ppmv, except in time_series.h5 in which they are given as mass fractions. To convert them in ppmv, the formula is: <code>c = c * (rho/rho_propylene) * 10**6</code> with (rho/rho_propylene) = 0.66 the density ratio between air and propylene.</p> <h2>Code examples</h2> <p>In the following, examples of how to use the PPMLES dataset in Python are provided. These examples have the following dependencies: </p> <pre><code>requires-python = ">=3.9" dependencies = [ "h5py==3.8.0", "numpy==1.26.4", "scipy", ]</code></pre> <h3>A) Dataset reading</h3> <div> <div> <pre><code>### Imports import h5py import numpy as np ### Load the input parameters list into a numpy array (shape = (200, 2)) inputf = h5py.File('PPMLES/input_parameters.h5', 'r') input_parameters = np.array((inputf['alpha_inlet'], inputf['friction_velocity'])).T<br>### Load the domain mesh node coordinates<br>meshf = h5py.File('../PPMLES/mesh.h5', 'r')<br>mesh_nodes = np.array((meshf['Nodes']['x'], meshf['Nodes']['y'], meshf['Nodes']['z'])).T ### Load the set of time-averaged LES fields and their associated uncertainty var = 'c' # Can be: 'c', 'u', 'v', 'w', 'crms', 'tke', 'uprim_cprim', 'vprim_cprim', or 'wprim_cprim' fieldsf = h5py.File('PPMLES/ave_fields.h5', 'r') fields_list = fieldsf[var] uncertaintyf = h5py.File('PPMLES/uncertainty_ave_fields.h5', 'r') uncertainty_list = uncertaintyf[var] ### Time series reading example timeseriesf = h5py.File('PPMLES/time_series.h5', 'r') var = 'c' # Can be: 'c', 'u', 'v', or 'w' probe = 32 # Integer between 0 and 92, see probe_network.csv time_list = [] time_series_list = [] for i in range(200): time_list.append(np.array(timeseriesf[f'Sample_{i+1:03}']['time'])) time_series_list.append(np.array(timeseriesf[f'Sample_{i+1:03}'][var][probe]))</code></pre> </div> </div> <h3>B) Interpolation of one-field from the unstructured grid to a new structured grid</h3> <pre><code>### Imports import h5py import numpy as np from scipy.interpolate import griddata ### Load the mean concentration field sample #028 fieldsf = h5py.File('PPMLES/ave_fields.h5', 'r') c = fieldsf['c'][27] ### Load the unstructured grid meshf = h5py.File('PPMLES/mesh.h5', 'r') unstructured_nodes = np.array((meshf['Nodes']['x'], meshf['Nodes']['y'], meshf['Nodes']['z'])).T ### Structured grid definition x0, y0, z0 = -16.9, -115.7, 0. lx, ly, lz = 205.5, 232.1, 20. resolution = 0.75 x_grid, y_grid, z_grid = np.meshgrid(np.linspace(x0, x0 + lx, int(lx/resolution)), np.linspace(y0, y0 + ly, int(ly/resolution)), np.linspace(z0, z0 + lz, int(lz/resolution)), indexing='ij') ### Interpolation of the field on the new grid c_interpolated = griddata(unstructured_nodes, c, (x_grid.flatten(), y_grid.flatten(), z_grid.flatten()), method='nearest')</code></pre> <h3>C) Expression of all time series over the same time window with the same time discretization</h3> <pre><code>### Imports import h5py import numpy as np from scipy.interpolate import griddata ### Define a common time discretization over the 200-s analysis period common_time = np.arange(0., 200., 0.05) u_series_list = np.zeros((200, np.shape(common_time)[0])) ### Interpolate the u-compnent velocity time series at probe DPID10 over this time discretization timeseriesf = h5py.File('PPMLES/time_series.h5', 'r') for i in range(200): sample_time = np.array(timeseriesf[f'Sample_{i+1:03}']['time']) - \ np.array(timeseriesf[f'Sample_{i+1:03}']['Parameters']['t_spinup']) # Offset the spinup time u_series_list[i] = griddata(sample_time, timeseriesf[f'Sample_{i+1:03}']['u'][9], common_time, method='linear')</code></pre> <h3>D) Surrogate model construction example</h3> <p>The training and validation of a POD-GPR surrogate model [Marrel et al. 2015] learning from the PPMLES dataset is given in the following <a href="https://github.com/eliott-lumet/pod_gpr_ppmles">GitHub repository</a>. This surrogate model was successfully used by Lumet et al. 2025 to emulate the LES mean concentration prediction for varying meteorological forcing parameters.</p> <h2>Acknowledgments</h2> <p>This work was granted access to the HPC resources from GENCI-TGCC/CINES (A0062A10822, project 2020-2022). The authors would like to thank Olivier Vermorel for the preliminary development of the LES model, and Simon Lacroix for his proofreading.</p>
Perturbed Parameters for ICEPACK-DART Study Titled "Exploring Bounded Non-parametric Ensemble Filter Impacts on Sea Ice Data Assimilation"
<p>The file contains the values of the two perturbed CICE parameters that were used in the study titled "Exploring Bounded Non-parametric Ensemble Filter Impacts on Sea Ice Data Assimilation." The tw perturbed parameters are the standard deviation of the dry snow grain radius (Rsnow), and the thermal conductivity of snow (Ksnow). There are 80 values since the ensemble used in the study had 80 members.</p>
Outputs from Isca perturbed parameter ensemble (PPE) simulations (Part II)
<p>Outputs from Isca perturbed paramter ensemble (PPE) simulations under 1xCO2 and 4xCO2, in which the simulations are prescribed with Q-flux. The output is the extracted data for the first 20-year simulations for the surface temperature and radiative fluxes at the top of the atmosphere (TOA).</p> <p>This is Part II, and Part I can be found at: <a href="https://doi.org/10.5281/zenodo.5150241">10.5281/zenodo.5150241</a></p>
Simulation results from HadGEM-UKCA perturbed parameter ensembles for Yoshioka et al. 2019 JAMES
<p>This dataset was created from perturbed parameter ensembles (PPEs) using HadGEM-UKCA atmospheric composition climate model and used in Yoshioka et al. (Ensembles of Global Climate Model Variants Designed for the Quantification and Constraint of Uncertainty in Aerosols and their Radiative Forcing) submitted to The Journal of Advances in Modeling Earth Systems (JAMES). It contains the following data;</p> <p>N50_sfc_data.tar.gz contains simulated number concentrations of particles larger than 50 nm from one-at-a-time screening experiments used in Figure 1 of the paper. teaca-teacl are job IDs of different experiments where values of parameters RAIN_FRAC and CLOUD_ICE_THRESH were varied;<br> teaca: RAIN_FRAC=0.6, CLOUD_ICE_THRESH=0.7<br> teacb: RAIN_FRAC=0.6, CLOUD_ICE_THRESH=0.5<br> teacc: RAIN_FRAC=0.6, CLOUD_ICE_THRESH=0.3<br> teacd: RAIN_FRAC=0.6, CLOUD_ICE_THRESH=0.1<br> teace: RAIN_FRAC=0.4, CLOUD_ICE_THRESH=0.7<br> teacf: RAIN_FRAC=0.4, CLOUD_ICE_THRESH=0.5<br> teacg: RAIN_FRAC=0.4, CLOUD_ICE_THRESH=0.3<br> teach: RAIN_FRAC=0.4, CLOUD_ICE_THRESH=0.1<br> teaci: RAIN_FRAC=0.2, CLOUD_ICE_THRESH=0.7<br> teacj: RAIN_FRAC=0.2, CLOUD_ICE_THRESH=0.5<br> teack: RAIN_FRAC=0.2, CLOUD_ICE_THRESH=0.3<br> teacl: RAIN_FRAC=0.2, CLOUD_ICE_THRESH=0.1</p> <p>CCN0p2_Stations_for_figure6.xlsx contains cloud condensation nuclei concentration at 0.2% supersaturation calculated for the AER ensemble of simulations at station locations used to create Figure 6 of the paper.</p> <p>Mean_Variance_CCN0p2_L9_AER_2008ANN.nc and Mean_Variance_CCN0p2_L9_AER-ATM_2006ANN.nc are emulator means and variances of annual average cloud condensation nuclei concentration at 0.2% supersaturation at model level 9 (~650m) in AER and AER-ATM ensembles and used in Figures 7 and 8.</p> <p>Mean_Variance_AOD550_AER_2008ANN.nc and Mean_Variance_AOD550_AER-ATM_2006ANN.nc are emulator means and variances of annual average aerosol optical depth at 550 nm in AER and AER-ATM ensembles and used in Figures 7 and 8.</p> <p>Mean_Variance_RFnet_AER_2008ANN.nc and Mean_Variance_ERF_AER-ATM_2006ANN.nc are emulator means and variances of annual average aerosol radiative forcing in AER ensemble and aerosol effective radiative forcing in AER_ATM ensemble and used in Figures 7, 8 and 9.</p> <p>Mean_Variance_SeaSalt_Load_AER_2008ANN.nc and Mean_Variance_SeaSalt_Load_AER-ATM_2006ANN.nc are emulator means and variances of annual average column mass loading of sea salt aerosol in AER and AER-ATM ensembles and used in Figure 9.</p>
Outputs from Isca perturbed parameter ensemble (PPE) simulations (Part I)
<p>Outputs from Isca perturbed paramter ensemble (PPE) simulations under 1xCO2 (control run) and 4xCO2 (perturbed run), in which the simulations are prescribed with Q-flux.</p> <ul> <li>cld_fbk_cmp.zip: The outputs used for the comparison of cloud feedback computation methods.</li> <li>qflux_clisccp_data.zip: The outputs contain the variable clisccp from control and perturbed runs, used for the cloud feedabck calculation</li> <li>qflux_extracted_data_toa_flux_30yr.zip: The last 30-year top of the atmosphere (TOA) flux data used for deriving the equilibrium climate sensitivity, total climate feedback and effective radiative forcing through the Gregory plot</li> <li>qflux_extracted_data.zip: The last 5-year simulation outputs for control and perturbed runs.</li> <li>qflux_extracted_data_10yr.zip: The last 10-year simulation outputs for control and perturbed runs.</li> </ul> <p>This is Part I, and Part II can be found at: <a href="https://doi.org/10.5281/zenodo.5188175">10.5281/zenodo.5188175</a></p>
An LES perturbed parameter ensemble of free-tropospheric cloud-controlling factors on stratocumulus
<p>This dataset contains a perturbed parameter ensemble of large-eddy simulations to assess the effect of two free-tropospheric cloud-controlling factors on stratocumulus clouds properties. The simulations were run on the UK Met Office/NERC cloud model (MONC) for the DYCOMS-II RF01 nocturnal stratocumulus case (Stevens et. al., 2005). Each simulation had the same initial conditions except for the two perturbed parameters, which were the jumps in moisture and temperature at the temperature inversion at cloud top. The corresponding analysis code can be found at this <a href="https://github.com/eers1/dycoms_analysis">dycoms_analysis Github page</a>. </p><p> </p><p>Stevens, B., Moeng, C. H., Ackerman, A. S., Bretherton, C. S., Chlond, A., de Roode, S., . . . Zhu, P. (2005). Evaluation of large-eddy simulations via observations of nocturnal marine stratocumulus. Monthly Weather Review , 133 (6), 1443–1462. doi: 10.1175/MWR2930.1</p>
Atmospheric and sea ice model fields from the perturbed parameter ensemble E3SMv0-HILAT used to examine emergent relationships among climate variables in the Arctic
<p>These files contain time series of several sea ice ad atmospheric fields produced in an ensemble of perturbed parameter simulations using the E3SMv0-HiLAT model. The time series are used to produced seasonal means, which are used to examine emerging relationships in the ensemble discussed in our manuscript, as of September 2019, in review at JGR</p>
CLM5 Perturbed Parameter Ensembles
<p>These data are the results of parameter sensitivity simulations and multiple perturbed parameter ensembles (PPE) with the Community Land Model, version 5 (CLM5). These simulations form the basis of a publication documenting a machine learning approach to studying parameter uncertainty in CLM5.</p> <p>The parameter sensitivity simulations include thirty-four CLM5 biophysical parameters perturbed one at a time to their minimum and maximum values, as well as a control simulation with default parameter values. The output files are labeled by the parameter name with "_min" or "_max" to denote simulation type, where each file includes 5 years of monthly output. The simulation with default parameter values includes 30 years of monthly output, labeled "CLM_control".</p> <p>The PPE simulations include six of the aforementioned parameters perturbed simultaneously. We use Latin Hypercube (LHC) sampling to generate 100 unique parameter sets for the top six parameters identified by the sensitivity simulations. There are two 100-member PPEs with present-day climate forcing conditions, but each uses a different LHC sampling. The output files are labeled "PPE_x" and "PPE_v2_x" with the ensemble member x from 1 to 100. There is also a simulation with default parameter values labeled "PPE_control". Again each file includes 5 years of monthly output.</p> <p>We test optimized parameter values with CLM, the results of which are labeled "CLM_test_calibration", and this file also includes 5 years of monthly output.</p>
Data for the bachelor thesis "Investigating Tuning Parameters in ECHAM-HAM with a Perturbed Parameter Ensemble"
<p>This repository contains the data for the bachelor thesis:<br> <br> Author: Melina Abeling<br> Title: Investigating Tuning Parameters in ECHAM-HAM with a Perturbed Parameter Ensemble<br> Date: July 2022</p> <p><br> The scripts are to be found in the accompanying package (https://doi.org/10.5281/zenodo.6866140).</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.