Code and data for paper: Evolution of tropical cyclone properties across the development cycle of the GISS-E3 global climate model

Jupyter notebooks, Python scripts, and data files used to make figures analyzing tropical cyclones (TCs) in the NASA GISS-E3 global climate model. Much of this code was adapted from Matlab code written by Jeffrey Strong.

This code and dataset is for a paper submitted to Journal of Advances in Modeling Earth Systems in 2021 by Rick Russotto, Jeffrey Strong, Suzana Camargo, Adam Sobel, Gregory Elsaesser, Maxwell Kelley, Anthony Del Genio, Yumin Moon, and Daehyun Kim. It is being uploaded to Zenodo in advance of submitting the revised version to JAMES. The code was earlier posted here: https://github.com/rdrussotto/GISS-E3_TC_notebooks

Versions of libraries used for paper: Python 3.6.12, NumPy 1.19.2, Matplotlib 3.2.0 (including Basemap), Pandas 1.1.5, XArray 0.16.2, SciPy 1.5.2

Notebooks used to make figures:

Figure 1: Plot_Tracks.ipynb

Figure 2: Plot_Density.ipynb

Figures 3-7: Plot_Statistics.ipynb

Figures 8-9: Plot_Storm_Tangential.ipynb

Figure 10: Plot_Radial_Profiles.ipynb

Figures 11-12: Sensitivity_Test_Plots.ipynb

Figure 13-14: Intermediate_Error_Plots_Simplified.ipynb

Figures 15-16: Intermediate_Error_Plots_Simplified_2.ipynb

Additional dependencies:

TC track preprocessing scripts:

Read_Zhao_TCs.ipynb (for modeled TCs)

Preprocess_IBTrACS.ipynb (for observed TCs)

Storm-centered diagnostics:

Storm_Centered_Plots.py

Storm_Centered_V1_Prec.py

Retrieve_Utan_Vtan.py

Radial-height profiles:

Figures 10 and S1 depend on Matlab scripts (not included here) developed by Yumin Moon for paper in Journal of Climate, Moon et al. (2020), https://doi.org/10.1175/JCLI-D-19-0172.1, and adapted by Jeffrey Strong for this study.

Sensitivity test analysis:

Figures 11-16 depend on Matlab scripts developed by Jeffrey Strong (not included here) for processing climate variables in the model results and comparing to observations.

Data files:

Radial profiles:

YuminEtAl_Figs.mat

Sensitivity test line plots (Figures 11-12):

JS_C180_ParamSensTest.mat

JS_C180v2_ParamSensTest.mat

Values plotted in error matrix plots for Figures 13-16: (use np.load in NumPy)

Fig13.npy

Fig14.npy

Fig15.npy

Fig16.npy

Modeled TC tracks:

zhao_tracks_v1.nc

zhao_tracks_v2.nc

(Observed TC data from IBTrACS project available at https://www.ncdc.noaa.gov/ibtracs/index.php?name=ib-v4-access)

Storm-centered TC quantities:

storm_centered_peak_intensity_v1.nc

storm_centered_peak_intensity_v2.nc

storm_centered_reference_winds_v1.nc

storm_centered_reference_winds_v2.nc

storm_centered_peak_intensity_prec_v1.nc