An open API service indexing awesome lists of open source software.

https://github.com/chilipp/gridding-temp12k

Package to grid the Temperature12K database
https://github.com/chilipp/gridding-temp12k

Last synced: about 1 year ago
JSON representation

Package to grid the Temperature12K database

Awesome Lists containing this project

README

          

# gridding-temp12k
Package to grid the Temperature12K database

This repository contains the scripts to compute a gridded representation of the
Temperature12k LiPD files. The important scripts are

- [scripts/compute-annual-gmst.ipynb](scripts/compute-annual-gmst.ipynb): This
notebook computes the mean surface temperature for the 6 latitudinal bands
using grid-cell specific GAMs for
Kaufman et al., *Holocene global mean surface temperature: A multi-method
reconstruction approach*, Scientific Data, submitted. Please see the
corresponding publication and/or the notebook for a more detailed description
- [scripts/grid-temp12k.ipynb](scripts/grid-temp12k.ipynb): This notebook
(which is not yet finalized) computes the anomaly per grid cell with reference
to PI.

## Installation
To install the necessary dependencies, please

1. Clone (or [download](https://github.com/Chilipp/gridding-temp12k/archive/master.zip)) this repository
2. install [anaconda](https://conda.io/en/latest/miniconda.html) for your
platform and run
3. Open the terminal (or Anaconda Prompt on Windows) and navigate to the
downloaded folder from step 1
4. Install the dependencies via `conda env create`
5. Activate the new environment via `conda activate temperature12k`
6. install the pyleogrid dependency via
`pip install git+https://github.com/Chilipp/pyleogrid.git@master`
7. run the jupyter notebook server on your machine via `jupyter notebook` (an
internet browser will open)
8. and in your internet browser, navigate to the desired jupyter notebook in
the [scripts](scripts) directory

If you are not familiar with jupyter notebooks, you can find some tutorials
at: [jupyter.org/try](https://jupyter.org/try)