climate indices
indices for extreme climate change detection (ETCCDI)
Climate models output terabytes of raw temperature and precipitation data at a sub-daily resolution. Climate indices are often used to identify extreme conditions such as drought or intense rainfall from the climate data.
Here I calculate ETCCDI climate change detection indices for Peru using daily precipitation from the ERA5-back extension (1950-1978) and ERA5 (1979-2019) reanalysis data. Number of days in year with precipitation > 20mm indicates intense rainfall that could potentially trigger a glacier lake outburst flood (shown purple).
The code to calculate the climate change detection indices is here

ETCCDI climate indices calculated using daily reanalysis precipitation. The open source Climate Data Operators (CDO) are used to calculate the indices.