!!!!!!!!!!!!!!!!!!!!!!!!! UNDER CONSTRUCTION !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
What is the objective of this page?
To help users to improve S2S ECMWF MARS requests performance via the WebAPI.
A good understanding of the MARS efficiency issues is essential especially for users that are interested in downloading large amounts of data.
How is the S2S reforecast data organised in MARS?
In general it is organised, as a huge tree, with the indentation below, showing different levels down that tree:
- centre (BoM, ECMWF, NCEP, JMA, ...)
- realtime or reforecast
- type of data (control forecast or perturbed forecast)
- type of level (single level or pressure level or potential temperature)
- model version date (2014-01-01 or ...)
- hindcast dates (2013-09-01, 2013-09-06, 2013-09-11, 2013-09-16, 2013-09-21, 2013-09-26, ...)
- time and steps
- members (for perturbed forecast)
- levels (for pl or pt)
- parameters
- levels (for pl or pt)
- members (for perturbed forecast)
- time and steps
- hindcast dates (2013-09-01, 2013-09-06, 2013-09-11, 2013-09-16, 2013-09-21, 2013-09-26, ...)
- model version date (2014-01-01 or ...)
- type of level (single level or pressure level or potential temperature)
- type of data (control forecast or perturbed forecast)
- realtime or reforecast
What would be the natural way to group requests?
The idea is to request as much data as possible from the same tape file. The natural way to group requests would be:
all parameters, all levels, all members, all time-steps for 1 hindcast date for a type of level for a type
Note the following:
- 'all' means 'all' that the user wants. It doesn't have to be all parameters.
- If a user is interested only on z500, he may request more hindcast dates in one go, since the overall request will not be so big.
An ECMWF reforecast request for all the available hindcastDates of an ECMWF model version date
- The objective of this example is to demonstrate how to iterate efficiently over all the available hincastDates for an ECMWF model version date.
- To start writing your request you firstly need to select a model version date and to find the corresponding Hindcast Dates. Please check ECMWF availability
Keep in mind that the ECMWF reforecast configuration is "on the fly". For more information read a brief description of reforecasts.
- The requests below can be used as a starting point, however you need to keep in mind that you have to adapt it to your needs eg to set the keyword values according to your requirements ("param", "levtype", "step" etc).
- In this way you can extend this request to download the whole S2S ECMWF reforecast. Don't forget to check ECMWF availability
You can use the variable target to organise the requested data in files as you wish.
#!/usr/bin/env python from ecmwfapi import ECMWFDataServer server = ECMWFDataServer() origin = "ecmf" modelVersionDate = "2015-12-03" hindcastDates = ["1995-12-03", "1996-12-03", "1997-12-03", "1998-12-03", "1999-12-03", "2000-12-03", "2001-12-03", "2002-12-03", "2003-12-03", "2004-12-03", "2005-12-03", "2006-12-03", "2007-12-03", "2008-12-03", "2009-12-03", "2010-12-03", "2011-12-03", "2012-12-03", "2013-12-03", "2014-12-03"] def retrieve_ECMWF_reforecast(): """ A function to demonstrate how to iterate efficiently over all hindcastDates for a particular ECMWF reforecast model version. Change the variables below to adapt the iteration to your needs """ # Please note that the "sfc" and "pl" requests below could run in parallel # Step 1: Get pressure level data pfplTarget = "%s_%s_%s.grb" % (origin, "pfpl", modelVersionDate) ECMWF_reforecast_pf_pl_request("/".join(hindcastDates), pfplTarget) # Step 2: Get surface data pfsfcTarget = "%s_%s_%s.grb" % (origin, "pfsfc", modelVersionDate) ECMWF_reforecast_pf_sfc_request("/".join(hindcastDates), pfsfcTarget) def ECMWF_reforecast_pf_pl_request(hindcastDate, target): """ A ECMWF reforecast, perturbed forecast, pressure level, request. The cost of this request is fields and Gbytes Change the keywords below to adapt it to your needs. """ server.retrieve({ "class": "s2", "dataset": "s2s", "date": modelVersionDate, "expver": "prod", "hdate": hindcastDates, "levtype": "pl", "levelist": "10/50/100/200/300/500/700/850/925/1000", "origin": origin, "param": "130/131/132/133/135/156", "step": "24/to/768/by/24", "stream": "enfh", "target": target, "time": "00", "number": "1/2/3/4", "type": "pf", }) #!/usr/bin/env python from ecmwfapi import ECMWFDataServer server = ECMWFDataServer() origin = "ecmf" modelVersionDate = "2015-12-03" hindcastDates = ["1995-12-03", "1996-12-03", "1997-12-03", "1998-12-03", "1999-12-03", "2000-12-03", "2001-12-03", "2002-12-03", "2003-12-03", "2004-12-03", "2005-12-03", "2006-12-03", "2007-12-03", "2008-12-03", "2009-12-03", "2010-12-03", "2011-12-03", "2012-12-03", "2013-12-03", "2014-12-03"] def retrieve_ECMWF_reforecast(): """ A function to demonstrate how to iterate efficiently over all hindcastDates for a particular ECMWF reforecast model version. Change the variables below to adapt the iteration to your needs """ #Please note that the "sfc" and "pl" requests below could run in parallel #Step 1: Get pressure level data pfplTarget = "%s_%s_%s.grb" % (origin, "pfpl", modelVersionDate) ECMWF_reforecast_pf_pl_request("/".join(hindcastDates), pfplTarget) #Step 2: Get surface data pfsfcTarget = "%s_%s_%s.grb" % (origin, "pfsfc", modelVersionDate) ECMWF_reforecast_pf_sfc_request("/".join(hindcastDates), pfsfcTarget) def ECMWF_reforecast_pf_pl_request(hindcastDate, target): """ A ECMWF reforecast, perturbed forecast, pressure level, request. The cost of this request is fields and Gbytes Change the keywords below to adapt it to your needs. """ server.retrieve({ "class": "s2", "dataset": "s2s", "date": modelVersionDate, "expver": "prod", "hdate": hindcastDates, "levtype": "pl", "levelist": "10/50/100/200/300/500/700/850/925/1000", "origin": origin, "param": "130/131/132/133/135/156", "step": "24/to/768/by/24", "stream": "enfh", "target": target, "time": "00", "number": "1/2/3/4", "type": "pf", }) def ECMWF_reforecast_pf_sfc_request(hindcastDate, target): """ A ECMWFreforecast, perturbed forecast, sfc request. The cost of this request is 383,040 fields and 7.1 GB Change the keywords below to adapt it to your needs. """ server.retrieve({ "class": "s2", "dataset": "s2s", "date": modelVersionDate, "expver": "prod", "hdate": hindcastDate, "levtype": "sfc", "origin": origin, "param": "1/33/34/59/121/122/134/136/146/147/151/165/166/167/168/169/175/176/177/179/180/181/235", "step": "24/to/744/by/24", "stream": "enfh", "target": target, "time": "00", "number": "1/2/3/4", "type": "pf", }) if __name__ == '__main__': retrieve_ECMWF_reforecast()
Useful links