- {
- “cells”: [
- {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“# Basic Usagen”, “n”, “n”, “## Climate indicator computationsn”, “n”, “xclim is a library of climate indicators that operate on [xarray](https://xarray.pydata.org/) DataArray objects. n”, “n”, “xclim provides two layers of computations, one responsible for computations and units handling (the computation layer, the indices), and the other responsible for input health checks and metadata formatting (the CF layer, refering to the Climate and Forecast convention, the indicators). Functions from the computation layer are found in xclim.indices, while indicator objects from the CF layer are found in realm modules (xclim.atmos, xclim.land and xclim.seaIce). n”, “n”, “To use xclim in a project, import both xclim and xarray. “
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“import xclimn”, “from xclim.testing import open_datasetn”, “import xarray as xr”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“Indice calculations are performed by opening a netCDF-like file, accessing the variable of interest, and calling the indice function, which returns a new DataArray. n”, “n”, “For this example, we’ll first open a demonstration dataset storing surface air temperature and compute the number of growing degree days (the sum of degrees above a certain threshold) at the monthly frequency. “
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“# ds = xr.open_dataset("your_file.nc")n”, “ds = open_dataset(‘ERA5/daily_surface_cancities_1990-1993.nc’)n”, “ds.tas”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“gdd = xclim.atmos.growing_degree_days(tas=ds.tas, thresh=’10.0 degC’, freq=’YS’)n”, “gdd”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“This computation was made using the growing_degree_days indicator. The same computation can be made through the indice:”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“gdd = xclim.indices.growing_degree_days(tas=ds.tas, thresh=’10.0 degC’, freq=’YS’)n”, “gdd”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“The call to xclim.indices.growing_degree_days first checked that the input variable units were units of temperature, ran the computation, then set the output’s units to the appropriate unit (here K d or kelvin days). As you can see, the indicator returned the same output, but with more metadata, it also performed more checks as explained below.n”, “n”, “The growing_degree_days indice makes most sense with daily input, but could accept other source frequencies. The computational layer assumes that users have checked that the input data has the expected temporal frequency and has no missing values. However, no checks are performed, so the output data could be wrong. If you’re unsure about all those things, a safer bet is to use `Indicator` objects from the CF layer, as done in the following section.n”, “n”, “<div class="alert alert-warning">n”, ” n”, “New unit handling paradigm in xclim 0.24 for indicesn”, “n”, “As of xclim 0.24, the paradigm in unit handling has changed slightly. Now, indices are written in order to be more flexible as to the sampling frequency and units of the data. You _can_ use growing_degree_days on, for example, the 6-hourly data. The ouput will then be in degree-hour units (K h). Moreover, all units, even when untouched by the calculation, will be reformatted to a CF-compliant symbol format. This was made to ensure consistency between all indices.n”, ” n”, “Very few indices will convert their output to a specific units, rather it is the dimensionality that will be consistent. The [Unit handling](units.ipynb) page goes in more details on how unit conversion can easily be done.n”, ” n”, “This doesn’t apply to Indicators. Those will always output data in a specific unit, the one listed in the Indicators.cf_attrs metadata dictionnary.n”, ” n”, “</div>n”, “n”, “n”, “Finally, as almost all indices, the function takes a freq argument to specify over what time period it is computed. These are called "Offset Aliases" and are the same as the resampling string arguments. Valid arguments are detailed in [panda’s doc](https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#offset-aliases) (note that aliases involving "business" notions are not supported by xarray and thus could raises issues in xclim.n”, “n”, “## Health checks and metadata attributesn”, “n”, “Indicator instances from the CF layer are found in modules bearing the name of the computational realm in which its input variables are found: xclim.atmos, xclim.land and xclim.seaIce. These objects from the CF layer run sanity checks on the input variables and set output’s metadata according to CF-convention when they apply. Some of the checks involve:n”, “n”, “* Identifying periods where missing data significantly impacts the calculation and omits calculations for those periods. Those are called "missing methods" and are detailed in section [Health checks](../checks.rst).n”, “* Appending process history and maintaining the historical provenance of file metadata.n”, “* Writing [Climate and Forecast Convention](http://cfconventions.org/) compliant metadata based on the variables n”, “and indices calculated.n”, “n”, “Those modules are best used for producing NetCDF that will be shared with users. See [Climate Indicators](../indicators.rst) for a list of available indicators.n”, “n”, “If we run the growing_degree_days indicator over a non daily dataset, we’ll be warned that the input data is not daily. That is, running
`xclim.atmos.growing_degree_days(ds.air, thresh='10.0 degC', freq='MS')`will fail with a ValidationError:”]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {
- “tags”: [
“raises-exception”
]
}, “outputs”: [], “source”: [
“ds6h = xr.tutorial.open_dataset(‘air_temperature’)n”, “gdd = xclim.atmos.growing_degree_days(ds6h.air, thresh=’10.0 degC’, freq=’MS’)”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“Resampling to a daily frequency and running the same indicator succeeds, but we still get warnings from the CF metadata checks.”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“daily_ds = ds6h.resample(time=’D’).mean(keep_attrs=True)n”, “gdd = xclim.atmos.growing_degree_days(daily_ds.air, thresh=’10.0 degC’, freq=’YS’)”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“To suppress the CF validation warnings in the following, we will set xclim to send them to the log, instead of raising a warning or an error.n”, “n”, “The missing method which determines if a period should be considered missing or not can be controlled through the check_missing option, globally or contextually. The main missing methods also have options that can be modified.”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“with xclim.set_options(check_missing=’pct’, missing_options={‘pct’: dict(tolerance=0.1)}, cf_compliance=’log’):n”, ” # Change the missing method to "percent", instead of the default "any"n”, ” # Set the tolerance to 10%, periods with more than 10% of missing datan”, ” # in the input will be masked in the ouput.n”, ” gdd = xclim.atmos.growing_degree_days(daily_ds.air, thresh=’10.0 degC’, freq=’MS’)”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“Finally, xclim also allows to call indicators using datasets and variable names.”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“with xclim.set_options(cf_compliance=’log’):n”, ” gdd = xclim.atmos.growing_degree_days(tas=’air’, thresh=’10.0 degC’, freq=’MS’, ds=daily_ds)n”, “n”, ” # variable names default to xclim names, so we can even do this:n”, ” renamed_daily_ds = daily_ds.rename(air=’tas’)n”, ” gdd = xclim.atmos.growing_degree_days(thresh=’10.0 degC’, freq=’MS’, ds=renamed_daily_ds)”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“## Graphics”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“import matplotlib.pyplotn”, “%matplotlib inlinen”, “n”, “# Summary statistics histogramn”, “gdd.plot()”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“# Show time series at a given geographical coordinaten”, “gdd.isel(lon=20, lat=10).plot()”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“# Show spatial pattern at a specific time periodn”, “gdd.sel(time=’2013-07’).plot()”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“For more examples, see the directions suggested by [xarray’s plotting documentation](https://xarray.pydata.org/en/stable/plotting.html>)n”, “n”, “To save the data as a new NetCDF, use to_netcdf.”
]
}, {
“cell_type”: “code”, “execution_count”: null, “metadata”: {}, “outputs”: [], “source”: [
“gdd.to_netcdf(‘monthly_growing_degree_days_data.nc’)”
]
}, {
“cell_type”: “markdown”, “metadata”: {}, “source”: [
“<div class="alert alert-info">n”, ” n”, “It’s possible to save Dataset objects to other file formats. For more information see: [xarray’s documentation](https://xarray.pydata.org/en/stable/generated/xarray.Dataset.html)n”, “n”, “</div>”
]
}
], “metadata”: {
“celltoolbar”: “Edit Metadata”, “kernelspec”: {
“display_name”: “Python 3”, “language”: “python”, “name”: “python3”
}, “language_info”: {
- “codemirror_mode”: {
“name”: “ipython”, “version”: 3
}, “file_extension”: “.py”, “mimetype”: “text/x-python”, “name”: “python”, “nbconvert_exporter”: “python”, “pygments_lexer”: “ipython3”, “version”: “3.8.8”
}
}, “nbformat”: 4, “nbformat_minor”: 2
}