Adding percentiles to a plot

Software and Tools Other - Software and tools
,
1

Dear CDS-Team,


the solution to my question regarding the addition of a trend line to my plot was very helpful (see "Plotting a trend line").

Continuing on a similar case the next step would be a line (or two) which represent the percentiles (e.g. 2.5% and 97.5%). There is indeed a function to calculate the values (ct.cube.quantile) but plotting a line is again tricky.

First idea was to use the function provided in the cds-forum thread "Plotting a trend line" and adjust the start- and end-values somehow to get a horizontal line. As I took notice of the possibility to import python libraries (like matplotlib and xarray), it seems still tough to replace xarray values with the required percentiles. Is this nevertheless a productive way or are there other possibilities to plot a horizontal line with percentiles?

Please see a shortened version of my code attached.

Any help is appreciated!


Thanks and regards

Tim Usedly

import cdstoolbox as ct

layout = ct.Layout(rows=1)

layout.add_widget(row=0, content=‘output-0’)


@ct.application(layout=layout)

@ct.output.livefigure()


def application():

NOAA = ct.catalogue.retrieve(

‘reanalysis-era5-land’,

{

‘variable’: ‘2m_temperature’,

‘year’: [‘2010’],

‘month’: [

‘01’, ‘02’, ‘03’,

‘04’, ‘05’, ‘06’,

‘07’, ‘08’, ‘09’,

‘10’, ‘11’, ‘12’,

],

‘day’: ‘01’,

‘time’: ‘00:00’,

}

)


NOAA_cut = ct.cube.box_select(NOAA, lat=[-60,60])
NOAA_avg = ct.cube.average(NOAA_cut, dim=['lon','lat'])

coords = ct.cdm.get_coordinates(NOAA_avg)
start_date = coords['time']['data'][0]
end_date = coords['time']['data'][-1]
    
fig  = ct.chart.line(
    NOAA_avg,
    layout_kwargs = {
        'xaxis':{
            'showline':True,
            'showgrid': True,  
        },
        'yaxis': {
            'showline':True,
            'showgrid': True,
            'tickmode' : 'auto'
        },
    },
    scatter_kwargs = {
        'mode': 'lines',
        'line.color':'black'
    }
)

# Percentil
quantile25 = ct.cube.quantile(NOAA_avg, q=0.025)
quantile975 = ct.cube.quantile(NOAA_avg, q=0.975)
print(quantile25)
print(quantile975)


# Calculate linear fit
fit = ct.stats.extrapolate(
    NOAA_avg,
    date_range_kwargs=
    {
        'start': start_date, 
        'end': end_date, 
        'periods': 2
    },
    func='linear'
)

# Plot the calculated fit
fig = ct.chart.line(
    fit,
    fig=fig,
    name='linear trend line',
    scatter_kwargs = {
        'marker':{
            'color':'orange',
        }
    }
)

return fig</pre>



2

Dear Tim,

The following app shows an example of how to add a horizontal lines to a livefigure:
https://cds.climate.copernicus.eu/toolbox-editor/168/forum_percentile

263×206 4.11 KB

I hope this helps.

Regards.

Vivien

3

Dear Vivien,

once again, thanks for your fast support! This function is helpful.

Seems like I tried it way too complicated ...

Regards,

Tim