README.md

In dblodgett-usgs/derivatives_runner: CMIP5 BCCA Derivatives Runner

BCCA CMIP5 Derivative Preparation

These scripts implement the climates R package. climates They also make use of pyGDP. pyGDP

The purpose of this package is to provide a record of how BCCA Climate Indices were calculated. It is not meant to be directly reusable bu may have reusable components for later climate index calculations. There are some system-specific paths and localhost urls in this repository. These have been left as is for the sake of clarity but would need to be changed if this code were to be reused.

The best way to review this code is to start with the 'main.R' script in the root of the package repository and descend into functions that are called by it. Many constants are set in the /R/derivatives_runner.R file and it is the primary function that runs the climate derivatives.

Package Description

This package was used to orchestrate processing of derived climate indices based on the BCCA CMIP5 downscaled climate data. The main.R script is where environment specific paths go and calls to python scripts are documented. The R package encoded here is used to go from daily gridded climate projections to annual climate indices. The annual climate indices are then ensembled together, climatalogical summaries are calculated, and historical-future difference summaries are generated. These calculations are orchestrated by the code in derivatives_runner.R.

With the derivative content created, a pair of python scripts are used against a localhost instance of the USGS Geo Data Portal and a THREDDS data server. Prior to calling those scripts, a script that generates .ncml aggregations for the data is run for the derivatives.

Service Configuration

Once all the climate indices are calculated, service configuration can be completed. There are scripts included in this repository that were used to do these one-time service configurations.

The python directory contains a script that will generate ncml that aggregates all the derivative data into large collections of variables rather than lots of individual files. There is also a script that is run in parallel against a localhost installation of the Geo Data Portal processing services. These scripts are called from main.R but could be called by hand just as easily. These scripts are included here for documentation purposes only.

The thredds directory contains an R script which is used to configure the THREDDS ncWMS, setting the scale range for coloring the mapped derivatives.

Ensemble Summary

Will be ensembling r1i1p1. Not all ensembles will include all GCMs/Scenarios.

| r1i1p1 | RCP26 | RCP45 | RCP60 | RCP85 | sum | hist | | ----- | ----- | ----- | ----- | ----- | ----- | ----- | | ACCESS1-0 | | 1 | | 1 | 2 | 1 | | bcc-csm1-1 | 1 | 1 | 1 | 1 | 4 | 1 | | BNU-ESM | | 1 | | 1 | 2 | 1 | | CanESM2 | 1 | 1 | | 1 | 3 | 1 | | CCSM4 | 1 | 1 | 1 | 1 | 4 | 1 | | CESM1-BGC | | 1 | | 1 | 2 | 1 | | CSIRO-Mk3-6-0 | 1 | 1 | | 1 | 3 | 1 | | GFDL-CM3 | 1 | | 1 | 1 | 3 | 1 | | GFDL-ESM2G | 1 | 1 | 1 | 1 | 4 | 1 | | GFDL-ESM2M | 1 | 1 | 1 | 1 | 4 | 1 | | inmcm4 | | 1 | | 1 | 2 | 1 | | IPSL-CM5A-LR | 1 | 1 | 1 | 1 | 4 | 1 | | IPSL-CM5A-MR | 1 | 1 | 1 | 1 | 4 | 1 | | MIROC-ESM | 1 | 1 | 1 | 1 | 4 | 1 | | MIROC-ESM-CHEM | 1 | 1 | 1 | 1 | 4 | 1 | | MIROC5 | 1 | 1 | 1 | 1 | 4 | 1 | | MPI-ESM-LR | 1 | 1 | | 1 | 3 | 1 | | MPI-ESM-MR | 1 | 1 | | 1 | 3 | 1 | | MRI-CGCM3 | 1 | 1 | 1 | 1 | 4 | 1 | | NorESM1-M | 1 | 1 | 1 | 1 | 4 | 1 | | sum | 16 | 19 | 12 | 20 | 13 |

dblodgett-usgs/derivatives_runner documentation built on May 15, 2019, 1:22 a.m.