LMRt 0.6.8

A lightweight, packaged version of the Last Millennium Reanalysis (LMR) framework

LMR Turbo (LMRt)

A lightweight, packaged version of the Last Millennium Reanalysia (LMR) framework, inspired by LMR_lite.py originated by Professor Hakim (Univ. of Washington). Ultimately, it aims to provide following features:

• Greater flexibility
  • Easy installation
  • Easy importing and usage in Jupyter notebooks (or scripts)
  • No assumption of a fixed folder structure; just feed the correct files to functions
  • Easy setup for different priors, proxies, and Proxy System Models (PSMs) included in PRYSM API
• Faster speed
  • Easy parallel computing with multiprocessing and other techniques
• Leveraging the power of Machine Learning (added in v0.6.0)

Results

Mean temperature

Niño 3.4 index

Package dependencies

• cartopy: a Python package designed for geospatial data processing in order to produce maps and other geospatial data analyses (conda install -c conda-forge cartopy).
• pyspharm: an object-oriented python interface to the NCAR SPHEREPACK library (conda install -c conda-forge pyspharm).
• tqdm: A fast, extensible progress bar for Python and CLI (pip install tqdm).
• prysm-api: The API for PRoxY System Modeling (PRYSM) (pip install prysm-api).
• dotmap: Dot access dictionary with dynamic hierarchy creation and ordered iteration (pip install dotmap).
• xarray: N-D labeled arrays and datasets in Python (pip install xarray).
• netCDF4: the python interface for netCDF4 format (pip install netCDF4).
• nitime: Timeseries analysis for neuroscience data (pip install nitime).
• statsmodels: Statistical models, hypothesis tests, and data exploration (pip install statsmodels).
• pyyaml: The next generation YAML parser and emitter for Python (pip install pyyaml).
• seaborn: Statistical data visualization using matplotlib (pip install seaborn).
• scikit-learn: Machine Learning in Python (pip install -U scikit-learn).
• keras: Deep Learning for humans (pip install keras).
• tensorflow: An Open Source Machine Learning Framework for Everyone (pip install tensorflow or pip install tensorflow-gpu).

How to install

Taking a clean install as example, first let's create a new environment named LMRt via conda

conda create -n LMRt python=3.7
conda activate LMRt

Then install two dependencies that is not able to be installed via pip:

conda install -c conda-forge cartopy pyspharm

Once the above dependencies have been installed, simply

pip install LMRt

and you are ready to

import LMRt

in python.

Notebook tutorials

• a quickstart
• building Ye files
• multiprocessing

References

• Hakim, G. J., J. Emile‐Geay, E. J. Steig, D. Noone, D. M. Anderson, R. Tardif, N. Steiger, and W. A. Perkins, 2016: The last millennium climate reanalysis project: Framework and first results. Journal of Geophysical Research: Atmospheres, 121, 6745–6764, https://doi.org/10.1002/2016JD024751.
• Tardif, R., Hakim, G. J., Perkins, W. A., Horlick, K. A., Erb, M. P., Emile-Geay, J., Anderson, D. M., Steig, E. J., and Noone, D.: Last Millennium Reanalysis with an expanded proxy database and seasonal proxy modeling, Clim. Past Discuss., https://doi.org/10.5194/cp-2018-120, in review, 2018.

License

BSD License (see the details here)

How to cite

If you find this package useful, please cite it with DOI:

... and welcome to Star and Fork!