NCEP Loading
Here we provide global grids of loading derived from the NCEP (National Center for Environmental Protection) Reanlysis surface Pressure data set.
This is a GGFC Operational Product. Data sets here are updated as often as the NCEP data are updated.
The 3-dimensional displacements have been derived using the method originally outlined in van Dam and Wahr [1987].
van Dam, T. and J. Wahr, Displacements of the Earth's surface due to atmospheric loading: Effects on Gravity and Baseline Measurements, J. Geophys. Res., 92, 1281-1286, 1987.)
Deviations from that procedure include:
- Determination of the ocean mask: In van Dam and Wahr [1987], a 2.5 X 2.5 degree land/ocean mask was used. Here, we use a 0.25 X 0.25 degree land/ocean mask. Units that contain only land are assigned the surface pressure defined by the original NCEP gridded file. Those with only water are assigned the modified inverted barometer (IB) pressure, defined in van Dam and Wahr [1987]. 2.5 degree grid units containing water and land are subdivided into 0.25 X 0.25 units, each assigned either the land value or the IB value as appropriate.
- The "average" pressure: In van Dam and Wahr [1987], the surface pressure from an arbitrary day was removed from all data used in that paper. Here we use a average pressure determined using a 20 year mean.
- The pressure data are low pass filtered to remove the erroneous atmospheric tides in the surface pressure data. The filtering means that the online data are always 3-4 days behind the actual date.
The data are first organized by reference frame. Within each reference frame directory the data are organized by year, then by day. Each day file contains 4 epochs of data at the following epochs: 00, 06, 12, 18. Each individual file contains the global (2.5 deg. X 2.5 deg) 3-dimensional surface displacement (in mm) for the epoch corresponding to the file name.
Each file can be read using the following algorithm (or something equivalent) where dvt is surface displacement to the north, dvl is surface displacement to the east, and du is surface displacement in the radial direction:
do ihr=1,4
read(iunit,1001) chyr,cdoy,chhr(ihr),ref
1001 format(a4,1x,a3,1x,a2,1x,a2)
do ilon=1,144
rlon=(ilon-1)*2.5
do ilat=1,73
rlat=90.-(ilat-1)*2.5
read(unit,1002) vt(ilon,ilat),vl(ilon,ilat),du(ilon,ilat)
1002 format(3(f10.4,2x))
end do
end do
end do
chyr | year in character format |
cdoy | doy in character format |
chhr(4) | hour epoch in character format |
ref | reference frame in character format |
rlon | longitude of data point |
rlat | latitude of data point |
You may use the program atml_tseries.f to extract the data for a single latitude and longitude. The program will need to be adapted to your particular needs. However, before altering the program, please compile on your system and confirm that you obtain the results similar to those in stjo.atml. Details are inserted as comments in the fortran file.
Anyone using this data set is requested to use the following citation:
van Dam, T., 2010, Updated October 2010. NCEP Derived 6-hourly, global surface displacements at 2.5 x 2.5 degree spacing. Data set accessed YYYY-MM-DD at http://geophy.uni.lu/atmosphere-downloads/
For questions regarding this data set, please contact T. van Dam, tonie.vandamuni.lu