xIndices: A xarray based module for computing SST trends, SST variability modes and other atmospheric variability modes

Install

Install the xIndices library using pip or conda. If you want to updrade to v1.37, please install fresh.

using Conda

conda create -n x_indices python
conda activate x_indices (If creating the x_indices environment)
conda install -c jiveshdixit -c conda-forge xindices

using pip

conda create -n x_indices python xesmf (MANDATORY)
conda activate x_indices
pip install xIndices

Overview

xIndices is a xarray based Python library for calculating climate variability indices and patterns. For now, from preprocessing the data to final calculation of indices and patterns, it allows user to directly calculate without diving into the steps for calculations. Key Features:

Load, pre-process and Regrid Multiple data preprocessing tools are already supported including using ESMF defined methods.
Rotated EOF analysis This also allows user to examine EOF modes (Rotated: Varimax and Promax & Unrotated) in user defined regions. Returns the desired number of modes (PCs and EOFs along with their variance explained in the data). This comes in handy when one wants to play around EOF patterns in user selected regions and variables.
Various Climate variability modes and warming trend Right now, we support SST Warming mode, ENSO mode using global SST (One can calculate other modes using generaic EOF tool of this package), PDO, AMO, NAO etc. We intend to add more variability modes.
Low-pass, high-pass and band-pass We have recently added Lanczos filter as it is very useful in climate variability analysis to filter-out the signal from noise.

Update

I have added Lanczos filter, correction for error due to latitude ascending, added some more support functions such as standardize variable, stack variables, Projectdata onto eofs etc. Some small bugs has also been addressed, however, they didn't affect the accuracy of analysis eralier

Community & Support

For now we have a Slack community page for comments, suggestions and error reporting.