A 1 year sea surface heat budget in the northeastern Atlantic basin during the POMME experiment: 2. Flux optimization

Abstract

Gridded air‐sea fluxes were computed during a 1 year period by applying a synthesis approach (Caniaux et al., 2005) as part of the Programme Océan Multidisciplinaire Méso Echelle (POMME) (September 2000–October 2001 in the northeast Atlantic 21°–15°W, 38°–45°N). In the present paper we address the question of adjusting the precipitation and turbulent surface fluxes so that the corresponding heat, salt, and momentum budgets are in agreement with the evolution of the thermal and salt contents deduced from the data collected during the experiment. The proposed method is based on the one‐dimensional modeling of the whole POMME area by solving specific temperature and salinity equations (including advective terms) and on the optimization of adjustable coefficients with a genetic algorithm. Adjusted fluxes allow us to simulate the average sea surface temperature and salinity with errors less than 0.02°C and 0.02 psu, respectively, over 1 year. The corresponding new surface heat budget indicates that the adjusted annual net heat (+17 W m−2) is weaker than previously calculated (+33 W m−2) but remains positive, meaning a heat gain for the ocean. The adjusted annual salt budget reveals a net evaporation of 0.75 mm d−1 for the POMME region. In contrast with climatology we find that the zone is located southward of the null buoyancy flux area during September 2000–September 2001.

A high‐resolution simulation of the ocean during the POMME experiment: Simulation results and comparison with observations

“…After this period, da‐MLD does not change much, except for a slight temporary increase in depth in response to wind stress maxima on days 114 (24 April) and 123 (3 May) (see Figure 4). At the end of the simulation, the da‐MLD is close to 20 m. If the simulation is continued after day 140 (20 May), da‐MLD remains close to 20 m until the Fall of 2001, as in other simulations using a 1‐D model [ Caniaux et al , 2005b], or a simplified 3‐D model assimilating geostrophic currents [ Giordani et al , 2005a, 2005b], therefore after day 140 the restratification period has ended in the POMME domain. The MLD standard deviation is also shown in Figure 7, and demonstrates that the absolute variation of MLD in the POMME domain is smaller after the restratification period than before.…”

“…The fluxes obtained are corrected using an optimization method based on a one dimensional oceanic model forced by lateral advection terms reconstructed from data, in order to equilibrate the annual mixed layer heat content deduced from the POMME data set [ Caniaux et al , 2005b]. Then, the daily fluxes produced are time interpolated at each time step of the model.…”

Mixed layer heat budget in the Iceland Basin from Argo

“…Thus, according to the residuals estimated in this study, monitoring the variability of the mixed layer processes from Argo data only remains a great challenge. As done in the POMME experiment [Caniaux et al, 2005a[Caniaux et al, , 2005b, the use of a calibrated flux product may enhance the precision of a mixed layer heat budget, but this cannot be achieved at basin scale. Our investigations proved that the Argo data set is suitable for the monitoring of the processes ruling the seasonal cycle of the mixed layer in the Iceland Basin.…”

“…Other estimates of the main terms of the surface heat budget have been obtained over smaller areas from different types of data but with large residuals and uncertainties [Wang and McPhaden, 1999;Swenson and Hansen, 1983;Foltz et al, 2003]. It is worth noting that observations made over the recent POMME experiment [Memery et al, 2005], which took place in a very well sampled 8°square area of the Northeast Atlantic, led to a more precise upper layer heat budget [Gaillard et al, 2005;Caniaux et al, 2005aCaniaux et al, , 2005b. However, such a high-resolution array is not available at a basin scale.…”

Two‐way one‐dimensional high‐resolution air–sea coupled modelling applied to Mediterranean heavy rain events

“…The thermocline is diagnosed as the first level from the top of the ocean that experiences a density difference larger than 0.02 kg m −3 from the reference ocean level (5 m). This method has been applied successfully in the past by Paci et al (2005) and Giordani et al (2005Giordani et al ( , 2006 the POMME experiment datasets (Weill et al, 2003;Caniaux et al, 2005aCaniaux et al, , 2005bMémery et al, 2005).…”