The following MARS requests can be used as a tutorial. The links into the MARS Catalogue might be helpful to create a valid retrieve script. However, the dates might need to be adapted.
As one can very easily retrieve large amounts of data it is best to start mars in a location where plenty of disk space is available, e.g. in $SCRATCH:
The solutions below provide valid MARS requests which can be passed to the mars command as
% mars < myrequestfile
or
% mars myrequestfile
Simple retrieves
Retrieve 1000,850,700 and 500 hPa temperatures from the 15th of February 1999 12 UTC Analysis > see in MARS Catalogue
solution
retrieve,
class = od,
type = an,
expver = 1,
date = 19990215,
time = 12,
param = t,
levtype = pressure level,
levelist = 1000/850/700/500,
target = "t.grb"
Retrieve wind components for tomorrow at 12:00 and 18:00 from yesterday's 12Z forecast at all 137 model levels > see in MARS Catalogue
solution
retrieve,
type = forecast,
param = u/v,
levtype = model level,
levelist = 1/to/137,
date = -1,
time = 12,
step = 48/54,
target = "wind"
Retrieve geopotential and surface pressure for yesterday's analysis cycles (00,06,12,18) > see in MARS Catalogue
solution
retrieve,
class = od,
type = analysis,
levtype = surface,
date = -1,
time = 00/06/12/18,
param = z/sp,
target = "all.an"
Retrieve surface pressure and surface thermal radiation downwards at every 12-hour time-step for the 12Z forecast from two days ago > see in MARS Catalogue
solution
retrieve,
type = forecast,
param = sp/str,
date = -2,
step = 12/to/240/by/12,
time = 12,
levtype = sfc,
target = "sfc.grb"
Retrieve the operational monthly mean model level temperature (all levels) from the 06:00 analysis for February 1997 > see in MARS Catalogue
solution
retrieve,
class = od,
stream = mnth,
type = analysis,
time = 06,
date = 19970200,
levtype = model level,
param = t,
levelist = 1/to/31,
target = "monthly"
Retrieve the temperature on model levels 1, 2 and 3 for yesterday's 12 hour forecast. > see in MARS Catalogue
solution
retrieve,
type = fc,
date = -1,
time = 12,
step = 12,
levtype = ml,
level = 1/2/3,
param = t,
target = "t_ml_1_2_3"
Time and step
Retrieve the 1000 hPa geopotential from yesterday's 12 UTC forecast (> see in MARS Catalogue) for:
the 12 hour forecast
solution
retrieve,
type = fc,
date = -1,
time = 12,
step = 12,
levtype = pl,
level = 1000,
param = z,
target = "z1000_a"
the 6, 12, 24 and 48 hour forecasts
solution
retrieve,
type = fc,
date = -1,
time = 12,
step = 6/12/24/48,
levtype = pl,
level = 1000,
param = z,
target = "z1000_b"
the 12 hour to 240 hour forecasts at 12 hour intervals
solution
retrieve,
type = fc,
date = -1,
time = 12,
step = 12/to/240/by/12,
levtype = pl,
level = 1000,
param = z,
target = "z1000_c"
all forecast time-steps
solution
retrieve,
type = fc,
date = -1,
time = 12,
# step = all, # The value "all" is convenient to use but MARS can only report on missing steps if
# the full list of steps is specified:
step = 0/1/2/3/4/5/6/7/8/9/10/11/12/13/14/15/16/17/18/19/20/21/22/23/24/25/26/27/28/29/30/31/32/33/34/35/36/37/38/39/40/41/42/43/44/45/46/47/48/49/50/51/52/53/54/55/56/57/58/59/60/61/62/63/64/65/66/67/68/69/70/71/72/73/74/75/76/77/78/79/80/81/82/83/84/85/86/87/88/89/90/93/96/99/102/105/108/111/114/117/120/123/126/129/132/135/138/141/144/150/156/162/168/174/180/186/192/198/204/210/216/222/228/234/240,
levtype = pl,
level = 1000,
param = z,
target = "z1000_d"
ENS Overlap
Retrieve Total Precipitation, perturbed forecast overlap
solution
RETRIEVE,
STREAM = EFOV,
CLASS = OD,
DATE = -1,
TIME = 00/12,
EXPVER = 1,
STEP = 240,
TYPE = PF,
NUMBER = 1/TO/50
LEVTYPE = SFC,
PARAM = TPVAR,
TARGET = TP_EFOV
See also FAQ: "How can I compute the accumulation using data from the overlap stream?"
Post-processing
Retrieve the 1000,500 and 300 hPa temperatures for today at noon from the 12 UTC forecast of two days ago on an N48 Gaussian Grid > see in MARS Catalogue
solution
retrieve,
type = forecast,
param = t,
date = -2,
time = 12,
step = 48,
grid = N48,
levelist = 1000/500/300,
levtype = pl,
target = "t.gg.grb"
Retrieve the same on a latitude/longitude grid, 1.5 by 1.5 degrees. > see in MARS Catalogue
solution
retrieve,
type = forecast,
param = t,
date = -2,
time = 12,
levelist = 1000/500/300,
levtype = pl,
step = 48,
grid = 1.5/1.5,
target = "t.ll.grb"
Retrieve Europe in the same way as in the previous retrieval. Compare the size of the resulting file with the previous one. > see in MARS Catalogue
solution
retrieve,
type = fc,
param = t,
date = -2,
time = 12,
levelist = 1000/500/300,
levtype = pl,
step = 48,
grid = 1.5/1.5,
area = e,
target = "t.europe"
Retrieve the 1000 and 500 hPa geopotential for 1200 yesterday from the analysis in the formats below (one job for each form): > see in MARS Catalogue
converting the data to a global N48 Gaussian grid
solution
retrieve,
type = an,
date = -1,
param = z,
levtype = pressure level,
level = 1000/500,
grid = N48,
target = "myfile1"
converting the data to a regular 5.0 by 5.0 degree global latitude/longitude grid
solution
retrieve,
type = an,
date = -1,
param = z,
levtype = pressure level,
level = 1000/500,
grid = 5.0/5.0,
target = "myfile1"
converting the data to an N80 Gaussian grid for the European area
solution
retrieve,
type = an,
date = -1,
param = z,
levtype = pressure level,
level = 1000/500,
grid = N80,
area = europe,
target = "myfile1"
converting the data to an N80 Gaussian grid for the area from 50 degrees North to 10 degrees North, and 20 degrees West to 20 degrees East
solution
retrieve,
type = an,
date = -1,
param = z,
levtype = pressure level,
level = 1000/500,
grid = N80,
area = 50/-20/10/20,
target = "myfile1"
converting the data to the same area but convert it to a 2.5 degree by 2.5 degree latitude/longitude grid
solution
retrieve,
type = an,
date = -1,
param = z,
levtype = pressure level,
level = 1000/500,
grid = 2.5/2.5,
area = 50/-20/10/20,
target = "myfile1"
Retrieve surface winds for yesterday's 24 hour forecast on the European area converting to a 0.57 by 0.57 degrees latitude/longitude grid. > see in MARS Catalogue
What happens to the conversion?
solution
retrieve,
type = fc,
date = -1,
time = 12,
step = 24,
levtype = sfc,
param = 10u/10v,
area = europe, # As the area is not compatible with the grid,
grid = 0.57/0.57, # MARS will expand the area to make it fit.
target = "10u_10v"
Observations
Retrieve all the sea surface observations for the 1st February 2015. > see in MARS Catalogue
solution
retrieve,
type = ob,
obsgroup= con,
obstype = ssd,
date = 20150201,
time = 00,
range = 1439,
target = "obs.bufr"
Ensemble forecasts
Retrieve the surface temperature fields of the first 10 members of yesterday's 12 UTC ensemble prediction (type: perturbed forecast) for the next 3 days at noon. > see in MARS Catalogue
solution
retrieve,
class = od,
stream = ef,
type = perturbed forecast,
levtype = sfc,
repres = gg,
date = -1,
time = 12,
ensemble = 1/to/10,
step = 48/to/96/by/24,
param = st
,
target = "ensemble"
Retrieve the same for the control forecast. > see in MARS Catalogue
solution
retrieve,
class = od,
stream = ef,
type = control forecast,
levtype = sfc,
repres = gg,
date = -1,
time = 12,
step = 48/to/96/by/24,
param = st,
target = "ensemble.cf"
ERA 40
Retrieve the vertical integral of eastward and northward heat flux as monthly means of daily means on a 1.125 by 1.125 degree latitude/longitude grid over Europe from the ERA 40 archive for January 1964 > see in MARS Catalogue
solution
retrieve,
class = e4,
stream = moda,
type = an,
levtype = sfc,
date = 19640101,
param = 69.162/70.162,
area = europe,
grid = 1.125/1.125,
target = "e4_moda_fluxes.grb"
Retrieval of snow depth from the ERA-40 archive for November 1993, for all analysis base times > see in MARS Catalogue. How many fields are requested?
solution
retrieve,
class = e4,
stream = oper,
expver = 1,
date = 19931101/to/19931130,
time = 00/06/12/18,
type = an,
levtype = sfc,
param = sd,
repres = gg,
target = "era40.199311.sd"
# It retrieves 30 x 4 = 120 fields.
See also these Web API retrieve examples for ERA-Interim, S2S, TIGGE, ERA-20C etc.