Summary

Given Latin hypercube samples of 18 climate model input parameter values, predict climate model simulation crashes and determine the parameter value combinations that cause the failures.

License

unknown (from UCI repository)

Dependencies

Tags

Classification climate-model ensemble-learning libsvm sensitivity-analysis simulation-crashes UQ

Attribute Types

Download

# Instances: 540 / # Attributes: 21
HDF5 (181.8 KB) XML CSV ARFF LibSVM Matlab Octave

Files are converted on demand and the process can take up to a minute. Please wait until download begins.

Original Data Format

csv

Name

climate-model-simulation-crashes

Version mldata

0

Comment

CSV

Names

Study, Run, vconst_corr, vconst_2, vconst_3, vconst_4, vconst_5, vconst_7, ah_corr, ah_bolus,

Data (first 10 data points)

Study	...	...	...	...	...	...	...	...	...	...
1	1	0.85...	0.92...	0.25...	0.29...	0.17...	0.73...	0.42...	0.56...	...
1	2	0.60...	0.45...	0.35...	0.30...	0.84...	0.93...	0.44...	0.82...	...
1	3	0.99...	0.37...	0.51...	0.50...	0.61...	0.60...	0.74...	0.19...	...
1	4	0.78...	0.10...	0.19...	0.42...	0.74...	0.49...	0.00...	0.39...	...
1	5	0.40...	0.51...	0.06...	0.63...	0.84...	0.44...	0.19...	0.48...	...
1	6	0.04...	0.62...	0.30...	0.81...	0.22...	0.97...	0.60...	0.64...	...
1	7	0.16...	0.54...	0.15...	0.65...	0.14...	0.79...	0.40...	0.66...	...
1	8	0.41...	0.89...	0.93...	0.91...	0.39...	0.00...	0.84...	0.68...	...
1	9	0.16...	0.35...	0.98...	0.28...	0.56...	0.40...	0.38...	0.47...	...
1	10	0.65...	0.41...	0.80...	0.16...	0.86...	0.94...	0.54...	0.42...	...
...	...	...	...	...	...	...	...	...	...	...

Description

This dataset contains records of simulation crashes encountered during climate model uncertainty quantification (UQ) ensembles.

Ensemble members were constructed using a Latin hypercube method in LLNL's UQ Pipeline software system to sample the uncertainties of 18 model parameters within the Parallel Ocean Program (POP2) component of the Community Climate System Model (CCSM4).

Three separate Latin hypercube ensembles were conducted, each containing 180 ensemble members. 46 out of the 540 simulations failed for numerical reasons at combinations of parameter values.

The goal is to use classification to predict simulation outcomes (fail or succeed) from input parameter values, and to use sensitivity analysis and feature selection to determine the causes of simulation crashes.

Further details about the data and methods are given in the publication "Failure Analysis of Parameter-Induced Simulation Crashes in Climate Models," Geoscientific Model Development (doi:10.5194/gmdd-6-585-2013).

URLs

http://www.geosci-model-dev-discuss.net/6/585/2013/gmdd-6-585-2013.html

Publications

Data Source

This data was constructed using LLNL's UQ Pipeline, was created under the auspices of the US Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344, was funded by LLNL's Uncertainty Quantification Strategic Initiative Laboratory Directed Research and Development Project under tracking code 10-SI-013, and is released under UCRL number LLNL-MISC-633994.

Measurement Details

Column 1: Latin hypercube study ID (study 1 to study 3)

Column 2: simulation ID (run 1 to run 180)

Columns 3-20: values of 18 climate model parameters scaled in the interval [0, 1]

Column 21: simulation outcome (0 = failure, 1 = success)

Usage Scenario

Given scaled values of climate model input parameters (columns 3-20), the goal is to predict climate model simulation outcomes (column 21, fail or succeed) and determine the causes of the crashes (sensitivity analysis and feature selection).

Please cite our paper "Failure analysis of parameter-induced simulation crashes in climate models" if you use the data in publications.

revision 1

by ddlucas on 2013-06-18 21:22

revision 2

by ddlucas on 2013-06-18 22:07

revision 3

by ddlucas on 2013-06-18 22:08

revision 4

by ddlucas on 2013-06-18 22:13

No one has posted any comments yet. Perhaps you would like to be the first?

Leave a comment

This item was downloaded 9189 times and viewed 3410 times.

No Tasks yet on dataset Climate Model Simulation Crashes