This shows the mean surface air temperature anomalies for the year 2020 as compared to the base period means from 1950-1980. The topographies shown are scaled by a factor of 20 to be more prevalent.
The data comes from a NASA GISS 5 X 5 curvelinear grid dataset with 1200km smoothing. The data was interpolated using the Python package xarray to scale the data by a factor of 64 prior to projecting it onto a polygonal mesh. Paraview was used to render this model.
https://data.giss.nasa.gov/gistemp/
Temperature anomoly colormap (degrees Celsius)
![](https://raw.githubusercontent.com/ingridarreola/E484/main/files/paraviewcolormap.png)