SOCIB CAIB Micinn

modelling and forecasting facility
SOCIB / Wmop operational validation

Operational Validation Menu

The performances of WMOP are monitored by performing routine validation against real-world measurements. Such validation allows us to quickly identify problems in the model as well as increasing our understanding of the model's abilities and limitations. A more complete hindcast validation (see section "Wmop historical validation") helps us to fine-tune the model for optimal performance.
You will find in this section quantitative model skill metrics between wmop and observations as well as some direct comparions. We are currently using data from remote sensing, Argo T/S data and in situ buoys.

Sea Surface Temperature

In this sub-section, the yesterday SST from WMOP is compared to a L3 satellite SST product (http://oceandata.sci.gsfc.nasa.gov/MODISA/L3SMI/ ) . Figure 1 and 2 represent respectively the daily mean SST from product and WMOP. Figure 3 represents the spatial differences between WMOP and product and Figure 4 displays the histogramm of those differences. Just click on the pictures to enlarge them.

Figure 1: Yesterday Satellite SST Figure 2: Yesterday WMOP SST
Figure 3: WMOP minus satellite SST Figure 4: Histogram of Model minus satellite SST

Geostrophic current from Altimetry

Here, the geostrophic currents are computed from WMOP and from the AVISO SALTO/DUACS Absolute Dynanic Topography altimetry product (ftp://ftp.aviso.oceanobs.com//donnees/ftpsedr/ DUACS/regionalmfstep/dt/upd/madt/merged/h/, see Le Traon et al., 1998; Pujol and Larnicol, 2005 for further details). Figure 1 and Figure 2 represent respectively the geostrophic current from AVISO and WMOP for yesterday. The differences between those are displayed in Figure 3.
Figure 1: AVISO geostrophic currents (m/s) Figure 2: WMOP geostrophic sea surface currents (m/s) Figure 3: Differences WMOP minus AVISO (m/s)

Argo T/S profiles

If available, each day, the yesterday Argo T/S profiles are downloaded from the ifremer ftp (ftp://ftp.ifremer.fr/ifremer/argo/latest_data/). Figure 1 represents the position of the available(s) ARGO float(s) whereas Figure 2 and 3 represent respectively the temperature and salinity from the ARGO float and from WMOP. Figure 4 displays the T/S diagram from both ARGO (red line) and WMOP (black line).
Figure 1: Argo position yesterday
Figure 2: Temperature profiles (degrees) FIgure 3: Salinity profiles (PSU) Figure 4: T/S diagram

Argo profiles weekly statistic

soon available

Moorings

In this section, the WMOP output are compared to the following buoys:
  • Gulf of Lion (from Météo France, red point on Figure 1)
  • Tarragona (from Puertos del Estado, pink point on Figure 1)
  • Valencia (from Puertos del Estado, white point on Figure 1)
  • Dragonera (from IMEDEA, black point on Figure 1)
  • Maó (from IMEDEA, cyan point on Figure 1)
  • Cabo de Palos (from Puertos del Estado, yellow point on Figure 1)
  • Cabo de Gata (from Puertos del Estado, green point on Figure 1)
Comparisons are made over the last month. Statistics are also computed (correlations, std, stddiff, misfit, scatterplot). Note the current are smoothed (fc=24h) in order to remove the inertial frequency.
By clicking on the different locations on Figure 1, the comparisons between WMOP and the selected buoys will be displayed.

carte


Moorings longterm comparisons

In this section, in order to follow the performance of the WMOP forecasting system, we compare the WMOP output to four following buoys:
  • Gulf of Lion (from Météo France, red point on Figure 1 Mooring sub-section)
  • Tarragona (from Puertos del Estado, pink point on Figure 1 Mooring sub-section)
  • Dragonera (from IMEDEA, black point on Figure 1 Mooring sub-section)
  • Cabo de Palos (from Puertos del Estado, yellow point on Figure 1 Mooring sub-section)
The blue curves represent the one month running-mean-bias (bias estimated over a window of 1 month) between wmop output and observations in term of Sea Surface Temperature and/or Sea Surface Salinity. The grey shaded envelop represent the one month running RMS-difference (+/- RMSdiff added the the blue curve).
Figure 1: Gulf of Lion Figure 2: Tarragona
Figure 3: Dragonera Figure 4: Cabo de Palos

Intercomparirons

In this section, the WMOP Sea Surface Temperature is compared to the MFS (Mediterranean Forecasting System) outputs, to the former version of WMOP and to the following buoys :
  • Gulf of Lion (from Météo France, red point Mooring sectionon Figure 1)
  • Tarragona (from Puertos del Estado, pink point on Figure 1)
  • Valencia (from Puertos del Estado, white point on Figure 1)
  • Dragonera (from IMEDEA, black point on Figure 1)
  • Cabo de Palos (from Puertos del Estado, yellow point on Figure 1)
  • Cabo de Gata (from Puertos del Estado, green point on Figure 1)
Comparisons are made over the last month. Statistics are also computed (correlations, std, stddiff, misfit, scatterplot)
Note the current WMOP version, in this subsection is called WMOPV2.

carte

Figure 1: WMOP bathymetry, the colored
points represent the considered buoys.


Figure 2a: Gulf of Lion, timeseries Figure 2b: Gulf of Lion, scaterplot
Figure 3a: Tarragona, timeseries Figure 3b: Tarragona, scaterplot
Figure 4a: Dragonera, timeseries Figure 4b: Dragonera, scaterplot
Figure 5a: Cabo de Palos, timeseries Figure 5b: Cabo de Palos, scaterplot
Figure 6a: Cabo Gata, timeseries Figure 6b: Cabo Gata, scaterplot
Figure 7a: Valencia, timeseries Figure 7b: Valence, scaterplot
RedIcts
©2010 ICTS SOCIB - Sistema d’observació i predicció costaner de les Illes Balears. Parc Bit, Naorte, Bloc A 2ºp. pta. 3
Palma de Mallorca SPAIN. E-07121. Tel: +034 971 43 99 98. CIF: Q0700535H.