Data and code for the publication "Day-to-day temperature variability reduces economic growth"

This repository contains code and data for the reproduction of analysis and figures from:

Kotz, Wenz, Stechemesser, Kalkuhl and Levermann (2020).

~

Data included:

GADM (https://gadm.org/data.html) dataset of regional shapefiles: gadm36_levels.gpkg 

Calculated regional climate variables: T.5_???_measure.npy and P.5_???_measure.npy, contained in ZIPPED folders T5_measures.zip & P5_measures.zip (??? denotes three letter country code).

World bank GDP and population data: WB_GDP.csv

Economic and regional climate data: T_econ.dta. Economic data is provided by Matthias Kalkuhl and is documented at https://doi.org/10.1016/j.jeem.2020.102360.

Economic and regoinal climate data with lagged variables: T_econ_5_lags.dta

Economic and regional climate data restricted to regions with certain seasonal temp. differences: T_econ_seas_g*.dta

Daily climate data used to plot Fig_1: corrientes_ARG_T_days.npy & ocampo_MEX_T_days.npy

~

Data not included:

Raw ERA-5 daily temperature and precipitation data, 1979-2018, interpolated to 0.5x0.5 grid by ISIMIP. Available from ISIMIP (https://www.isimip.org/) or from authors upon request.

~

Code included:

T_scatter.py - code to calculate and aggregate the main temperature variables (annual average and day-to-day variability) from ERA-5 grid to regional level.

P_scatter.py - code to calculate and aggregate precipitation variables (total annual) from ERA-5 grid to regional level.

Regression_Table_1 - Stata do file including code to run regressions for Table 1 of the manuscript.

Table_SX - Stata do files including code to run regressions for Tables S1-11 of the SI.

partitions.py - code to partition data based on national and regional income, to export sub-data sets for analysis in R, and for plotting Fig S4.

partition_regressions.R - code to calculate marginal effects used for plotting in Figs 4, S1, and S5.

partitions_plot.py - code to plot Figs 4, S1 and S5.

plot_Fig_1.py

plot_Fig_2.py

plot_Fig_3.py

~

Order:

1. T_scatter.py and P_scatter.py to calculate regional climate variables using raw ERA-5 data and GADM regional shapefile data. (Note: set correct directory for raw ERA-5 data on l101 of T_scatter.py and l102 of P_scatter.py).

2. Regression_Table_1 and Table_SX to run main and supplementary regressions in Stata using the economic and climate data set T_econ.dta (and T_econ_5_lags.dta for Table_S_10 and T_econ_seas_g*.dta for Table_S_2).

3. plot_Fig_1.py using corrientes_ARG_T_days.npy & ocampo_MEX_T_days.npy.

4. plot_Fig_2.py using T_econ.dta and gadm36_levels.gpkg.

5. plot_Fig_3.py using T_econ.dta, T.5_???_measure.npy and WB_GDP.csv.

6. partitions.py to partition data based on regional and national income and to plot Fig. S4 using T_econ.dta and gadm36_levels.gpkg.

7. partition_regressions.R to estimate marginal effects of partitioned data, using the output of partitions.py.

8. partitions_plot.py, plot Figs. 4, S1 and S5 using the output of partition_regressions.R and T_econ.dta.