Climate Indicators

xclim.core.indicator.Indicator instances essentially perform the same computations as the functions found in the xclim.compute library, but also run a number of health checks on input data, assign attributes to the output arrays and return a xarray.Dataset. For example, if there are missing values in a time series, compute functions will ignore them, but indicators will return NaN for periods with missing values (depending on the missing values algorithm selected, see module xclim.core.missing). Indicators also check that the input data has the expected frequency (e.g. daily) and that it is indeed the expected variable (e.g. a precipitation flux). The output is assigned attributes that conform as much as possible with the CF-Convention.

Indicators are split into realms (atmos, land, seaIce), according to the variables they operate on. See Defining new indicators for instruction on how to create your own indicators. This page allows a simple free text search of all indicators. Click on the python names to get to the complete docstring of each indicator.

Include virtual submodules in results.