ESSENCE (Ensemble SimulationS of Extreme weather events under Nonlinear Climate changE) is a DEISA (Distributed European Infrastructure for Supercomputing Applications) project that is conducted by CKO (Centrum voor Klimaatonderzoek) to explore the changing climate and its variability. A large-member ensemble of runs with a state-of-the-art climate model is performed to investigate the range of possible future climate change. The project is led by Henk Dijkstra from Utrecht University. More information can be found here.

The model used is the ECHAM5/MPI-OM coupled climate model developed at the Max-Planck-Institute for Meteorology (MPI-Met) in Hamburg. The model runs are performed on the NEC SX-8 at the Höchstleistungsrechenzentrum Stuttgart (HLRS).

Model runs start in 1950 and end in 2100. During the historical period (1950-2000) they are forced by observed concentrations of greenhouse gases and anthropogenic aerosols. For the future the forcing follows the SRES A1b scenario. The whole ensemble is subdivided into a standard run and several experiments according to the following table:

run no.	experiment	PI	state	postprocessing
001-020	standard; constant forcing	-	stopped	finished
021-036	standard; SRES A1b	-	finished	finished
037	standard; SRES A1b; extra output	-	finished	finished
041-045	HOSING 1	Hazeleger	finished	finished
046-050	HOSING 2	Dijkstra	finished	finished
051-057	SOIL	v/d Hurk	finished	finished
101-220	HEAT	Vossepoel	finished	finished

The results from the model runs are stored at SARA in Amsterdam. To get access to the data a special login is needed. Please send me an e-mail containing your name, institution and e-mail address to obtain such a login.

If you do not belong to the ESSENCE project group yet please also give a short description of the research that you want to perform. The ESSENCE steering group will then decide on your request.

The raw model output from ECHAM5 is in GRIB format, that from MPI-OM in EXTRA format. The organization of the files and their contents also differs between the models. At SARA we transform the data to netCDF format and reorganize them as to have one variable per file per run (ensemble member). As some files would become very large this way we split them into pieces of 5 years (atmosphere) and 10 years (ocean). For some runs from the first series (runs 001 - 020) the files have a different length. Follow the links in the right margin for the actual organization of the data. For technical reasons the splitting is also done for those files that would allow to have them in one file for the whole run. These files are now being concatenated. The following link leads to a list with the current state of this action:

Variables form the atmosphere model (ECHAM5) are on a regular lat/lon grid. Those from the ocean model (MPI-OM) are on the original curvilinear model grid. To display them properly they have to be inerpolated to a regular grid. Vector components are given with respect to the curviliniear model grid. They have to be rotated to geographical north/south and east/west directions before interpolation to achieve a proper representation. More about how to do this can be found at

Meridional overturning can be computed using the follwoing package:

The files are named as quantity_run_period.nc, where
quantity is an abbreviation for the physical quantity. The abbreviations can be found by clicking on the links in the table below.
run is the run number according to the table above, and
period = begin_year-end_year is the period for which the file contains data.

The files are stored in directory /archive/essence/DATA on HUYGENS at SARA. The DATA directory contains one subdirectory for each run, conveniently labled as run???, with ??? being the run nubmer. Each of the run???-directories is subdivided into twelve subdirectories, one for ocean output and the other eleven for output from the atmosphere model:

oce	monthly means of ocean fields (2D and 3D)
2D_3hrly	2D surface fields with a resolution of 3 hours
2D_daymean	daily means of 2D surface fields
2D_monmean	monthly means of 2D surface fields (from 6-hourly output)
2D_monmean_3h	monthly means of 2D surface fields (from 3-hourly output)
3D_daymean	daily means of 3D fields (17 pressure levels)
3D_daymean_500	daily means of selected quantities at 500 hPa height
3D_daymean_gh	daily means of geopotential at 5 selected heights
3D_monmean	monthly means of 3D fields (31 model levels)
ATM	monthly means of 3D fields (17 pressure levels)
strimau	monthly means of several cloud-related variables (31 model levels)
ZO_ozone	monthly zonal means of Ozone

An example of a complete file name is /archive/essence/DATA/run005/oce/THO_005_1950-1960.nc