CDS Toolbox

The Climate Data Store (CDS) Toolbox is the path to a wealth of past and future climate information. Free and available to everyone, the Toolbox links raw data to online computing power through a programming interface. In your own online workspace, you can create applications in Python (a programming language) and run them on the CDS computers, allowing you to retrieve the data you are interested in, make the calculations you require and display the results in the format that suits your needs. You can download graphs, maps, and data, and also share your online creations with other users.

The Toolbox is aimed at a wide range of users, from amateur climate enthusiasts to students, researchers and software developers. To make use of the Toolbox, you will need a basic working knowledge of Python and access to the internet. You will not need a particularly powerful computer or a lot of storage, as the calculations and the data processing take place online within the CDS.

The CDS application page showcases the range of applications that is possible to build using the Toolbox.

CDS Toolbox

The CDS Toolbox service is accessible here.

Below there are two examples relative to CEMS-Flood datasets.

GloFAS discharge map

import cdstoolbox as ct

@ct.application(title='Download data')
@ct.output.download()
@ct.output.livemap()
def download_application():
    data = ct.catalogue.retrieve(
        'cems-glofas-historical',
        {
            'system_version': 'version_3_1',
            'hydrological_model': 'lisflood',
            'product_type': 'consolidated',
            'variable': 'river_discharge_in_the_last_24_hours',
            'hyear': '1995',
            'hmonth': 'january',
            'hday': '01',
        }
    )
    # update attribute to 'discharge-global' to set the GloFAS palette
    data =ct.cdm.update_attributes(data,attrs={'cds_magics_style_name':'discharge-global'})
    
    # call live plot
    fig = ct.livemap.plot(data)
    
    return data,fig

Retrieve time series

import cdstoolbox as ct
import calendar
import datetime

# === USER PARAMETERS  ===

VARIABLES = 'River discharge in the last 24 hours'
DATASETS = {
    'Consolidated reanalysis':'consolidated_reanalysis',
    'Intermediate dataset': 'intermediate_dataset'
}



# === MAIN APPLICATION ===

layout = ct.Layout(rows=4)
layout.add_widget(row=0, content='Dataset', sm=4)
layout.add_widget(row=0, content='start_date', sm=4)
layout.add_widget(row=0, content='end_date', sm=4)
layout.add_widget(row=1, content='Lat', sm=4)
layout.add_widget(row=1, content='Lon', sm=4)
layout.add_widget(row=2, content='[submit]',sm=2)
layout.add_widget(row=3, content='output-0', widget_align='center')
layout.add_widget(row=3, content='output-1', widget_align='center')

@ct.application(layout=layout,autorun=False)

@ct.input.dropdown('Dataset', values=DATASETS, default='Consolidated reanalysis')
@ct.input.text(name="Lat",default="51.71,52")
@ct.input.text(name="Lon",default="4.87,6")
@ct.input.text(name = "start_date",label='Start date', default='2018-01-01')
@ct.input.text(name="end_date",label='End date', default='2018-04-30')

@ct.output.markdown()
@ct.output.download()
def viewer_application(Dataset= None,
                       Lat=None,
                       Lon=None,
                       start_date=None,
                       end_date=None):
    

   
    
    dates = compute_dates_range(start_date,end_date)
    

    data = []
    
    for date in dates:

  
        _,n_days = calendar.monthrange(date.year,date.month)
    
        request =             {
            'system_version': 'version_2_1',
            'variable': 'river_discharge_in_the_last_24_hours',
            'hyear': date.strftime("%Y"),
            'hmonth': date.strftime("%m"),
            'hday': list(map(lambda x: '%02d'%(x) ,range(1,n_days+1))),
            'dataset': DATASETS[Dataset],
        }


        temp = ct.catalogue.retrieve('cems-glofas-historical', request)

        
        data.append(temp)
        

    data = ct.cube.concat(data,dim="forecast_reference_time")
    
    data =ct.cdm.update_attributes(data,attrs={'cds_magics_style_name':'discharge-global'})

    lat,lon= list(map(float,Lat.split(","))),list(map(float,Lon.split(",")))

    
    if len(lat)> 1:
        point = [ct.shapes.point(lat=y,lon=x,label= f'Lon: {x}\n Lat: {y}') for x,y in zip(lon,lat)]
    else:
        point = [ct.shapes.point(lat=lat, lon=lon, label= f'Lon: {lon}\n Lat: {lat}')]

    ts = []

    for x,y in zip(lon,lat):

        
        ts.append(ct.geo.extract_point(data,lon=x, lat=y))
    
    if len(ts) >1:
     
        ret =  ct.cube.concat(ts, dim='station')
    
    else:
        ret = ts[0]
    
    
    text = f"#Download {len(lat)} time series, from {start_date} to {end_date}"
    
    
    return text,ret
    


# === FUNCTIONS ====

def compute_dates_range(start_date,end_date):
    start_date = datetime.date(*[int(x) for x in start_date.split('-')])
    
    end_date = datetime.date(*[int(x) for x in end_date.split('-')])
    
    end_date = end_date + datetime.timedelta(1)
    
    date_range =  end_date - start_date
    
    dates = []
    for d in range(date_range.days):
        dates.append(start_date + datetime.timedelta(d))
    
    dates = [i for i in dates if i.day == 1]
    return dates

Page tree

CDS Toolbox