Climatological data


Main page



Summer surface salinity field constructed on a 167x63 grid.
From 34.7 psu (blue) to 39.8 psu (red).

Table of Contents


1. Introduction

This page provides access to the Mediterranean seasonal climatology reconstructed from an historical hydrological data base. The resulting 3D fields (salinity, temperature and velocity) are stored as 3D matrices, with description of the grid location. Direct access to these fields and a description of the file format are also provided on this page.

If you prefer to load the files manually, please use the ftp server instead.

If there is any problem with this data set (reading, importing, ...) please contact the MODB support (

(up to Table of Contents)

2. Three successive releases

Three versions of this seasonal gridded data set have been released (MED2, MED4 and MED5). They differ in the data base and the methods that have been used to build them. These data sets are available on the last part of this page by clicking on the convenient (MED2, MED4 or MED5) button.


The MED2 Gridded Data Set is the result of the application of the VIM analysis method (1) to an historical data bank resulting from the merging of two main files:

  • the french BNDO file (Bureau National des donnees oceanographiques, Brest, France) containing Bottles profiles collected since the beginning of the century.
  • the U.S. NODC file (National Oceanographic Data Center, Washington, USA) which is a subset of a world ocean data bank. (CTD and BOTTLES profiles only)

A full description of this data set, the analysis technique and the results obtained at the seasonal time scale is available in (2). The printed version of an atlas (3) can be ordered by sending an e-mail to

  1. Brasseur P., 1991, A Variational Inverse Method for the reconstruction of General Circulation Fields in the Northern Bering Sea, J. Geophys. Res., 96(C3), 4891-4907.
  2. Brasseur P., Beckers J.M., Brankart J.M. and Schoenauen R., 1996, Seasonal Temperature and Salinity Fields in the Mediterranean Sea: Climatological Analyses of a Historical Data Set. Deep Sea Research. 43(2), 159-192.
  3. Brasseur P., Brankart J.M., Beckers J.M., 1994, Seasonal Variability of the General Circulation of the Mediterranean : results of climatological analyses. (B/W)


The MED4 Gridded Data Set was computed using the first released MODB data base which differs from the previous one mainly in the addition of the bathythermographs from the U.S. NODC world ocean data bank.

This version of the climatology does not include the full MODB data base yet. Nevertheless, it has been used as a reference climatology in several scientific works (see examples below).

  1. Brankart J.M. and Brasseur P., 1996. The general circulation in the Mediterranean Sea: a climatological approach. Submitted to the Journal of Marine Systems.
  2. Wu, P and Haines, K., 1996. A modelling study of deep water formation in the Mediterranean. Journal of Geophysical Research. In press.


This may be considered as the final version of the Mediterranean climatology released by the MODB project. The final report include an atlas representing this climatology. We are preparing both paper and online versions of this document.

(up to Table of Contents)

3. Analysed Data Fields

There are four seasons for the data set :

  1. WINTER : January, February, March
  2. SPRING : April, May, June
  3. SUMMER : July, August, September
  4. AUTUMN : October, November, December

There are four files for each season :


Units are 'psu' for salinity, 'degree Celsius' for temperature and 'm/s' for velocity.

(up to Table of Contents)

4. Matrix description

To understand the way the data is stored, please refer to the figure below. The files are stored in a 3D array (matrix) with 3 indexes I, J, K, going from 1 to IMAX, 1 to JMAX, 1 to KMAX. The origin is located at the 'bottom left deepest corner'.


There are KMAX horizontal slices from the bottom (K=1) to the surface (K=KMAX). Increasing I's go from W to E and increasing J's go from S to N.

Islands, coasts and places where there are no values are represented by an exclusion value.

(up to Table of Contents)

5. Localisation

The point (i,j,k) is located at (-5.5 + i*0.25 , i=1,167) long and (30.0 + j*0.25 , j=1,63) lat at a depth given by the Kth line of the file MED2.zed , MED4.zed or MED5.zed .

Example :
The point (3,4,1) is located at -4.75 long and 31.0 lat at the bottom deepest layer.
The grid of the MED4 climatology has been temporarily extended westwards.
The point (i,j,k) is located at (-9.5 + i*0.25, i=1,184) lon and (30.0 + j*0.25, j=1,63 ) lat
Depth is given by the kth line of the file MED4.zed.

(up to Table of Contents)

6. Reading the data

The data sets are stored in compressed ASCII files. First you must 'uncompress' the files with the Unix command 'uncompress'. The size of your file will increase to about 4.3 MB.

The first line of the ASCII file contains the dimension of the array : IMAX, JMAX, KMAX. The integers are stored with the FORTRAN format I5.

The second line contains the exclusion value, the minimum and the maximum value of this file. These numbers are stored with the FORTRAN format E12.5.

The remaining lines contains the data of the array, with 5 numbers per line (except the last line for each I-line).

The array is stored in horizontal slices from sea surface to sea bottom and from north to south. So the indexes go from :

   DO K = KMAX to 1
       DO J = JMAX to 1
           DO I = 1 to IMAX

There are two examples of extracting routines written in FORTRAN : readata1.f , readata2.f

(up to Table of Contents)

7. Access to the Data Fields

Here is the list of the available matrices :

  1. Winter
  2. Spring
  3. Summer
  4. Automn

(up to Table of Contents)  


[Atlas] [Historical hydrographic DB ] [ Climatological data ] [Charts Generator ] [ Bathymetric Data ] [Diva] [ModbWin] [Publications] [Pictures] [MPEG Animations]
[Main page]

 Last modified 09/26/01