justinalsing / basic

BAyeSian Integrated and Consolidated (BASIC) composite merging for atmospheric/climate time-series data.

BASIC

BAyeSian Integrated and Consilidated (BASIC) composite super-merging for atmospheric and climate time-series data.

This package implements the BASIC algorithm descibed in Ball et al 2017 for constructing a super-merged composite product from a set of potentially artefact biased composite time-series records.

Installation

The code is python3 and has the following dependencies (which can be installed using eg., pip install):

numpy
pystan
scipy

The tutorial BASIC_tutorial.ipynb also requires:

matplotlib
netCDF4

Usage

Functionality

A detailed annotated tutorial walk-through of how to use the code is given in the jupyter notebook BASIC_tutorial.ipynb -- this tutorial builds a BASIC super-merged composite for stratospheric ozone time-series data as an example. The notebook takes you step-by-step through the complete functionality of the code: loading in your own data, running the BASIC merging algorithm, and plotting the results.

Pre-built BASIC composite products

The directory basic_composites/stratospheric_ozone/ contains a number of pre-built BASIC composite products for stratospheric ozone, that can be downloaded and used out-of-the-box.

The netCDF files include variables for time, pressure and latitude giving the time (Julian date) and pressure (hPa) and latitude (degrees) grids respectively. The BASIC composite time-series data are then given in the variable o3[time, pressure, latitude] and associated (time-varying) 1-sigma uncertainties are given in sigma_o3[time, pressure, latitude].

BASIC_V1_swooshV2.6_gozcardsV1.0_sbuvmodV8.6_sbuvmer.nc is built from SWOOSH v2.6, GOZCARDS v1.0, SBUV-MOD v8.6 and SBUV-MER (as described in Tummon et al 2015). This corresponds to the BASIC composite presented in Ball et al 2017; the data runs up until Dec 2012.

BASIC_V1_swooshV2.6_gozcardsV2.20.nc is built from SWOOSH v2.6 and GOZCARDS v2.20; the updated data runs up until Dec 2016. This data was used in Ball et al 2018 (referred to as merged-swoosh/gozcards in that paper).

Citing this code

There is a JOSS paper in preparation to accompany the code (appearing soon), and the method is introduced in detail in Ball et al 2017. Please cite these two papers if you use this code or the resulting BASIC data products.

Contributions, reporting issues and feature requests

If you want to contribute (eg., extended models) to this package, request a feature, or report a bug/issue, please use the issues channel associated with this Git repository.