Data and codes related to the article: Renard et al. A Hidden Climate Indices Modeling Framework for Multi-Variable Space-Time Data. Water Resources Research.

This package contains data and codes related to the article:

B. Renard, M. Thyer, D. McInerney, D. Kavetski, M. Leonard and S. Westra.  A Hidden Climate Indices Modeling Framework for Multi-Variable Space-Time Data. Water Resources Research.

R scripts

The main computations of the paper have been performed using a computing code named STooDs, which is called using the bash script launchpad.sh.

The R scripts in this package only perform pre-processing (create configuration files) and post-processing (analyze results) steps.

• Funk.R: a set of functions called by other scripts.
• 1_defineModel.R: define the model to be inferred and create STooDs configuration files in dataset_XXX/runs.
• 2_analyzeResults.R: analyze the outputs of STooDs runs.
• 3_crossValidation.R: analyze the outputs of cross-validation experiments in dataset_XV and dataset_XV_1971-1990.

Data

Data for the 3 cases (full dataset and 2 cross-validation experiments) are located in folders dataset_XXX/data.

• dat.txt: raw dataset in text format.
• dataset.RData: dataset in RData format.
• DMI.txt, NINO.txt, SAM.txt: 3 standard climate indices.
• spaceP.txt, spaceQ.txt, spaceT.txt: properties of Precipitation (P), Streamflow (Q) and Temperature (T) stations.
• [only for cross-validation experiments] validation.RData: left-out data used for validation.