The use of circulation weather types to predict upwelling activity along the western Iberian Peninsula coast

Abstract

Coastal upwelling is a phenomenon that occurs in most western oceanic coasts, associated with coastal surface water divergence and consequent ascension of colder and nutrient-rich waters from deeper levels. In this work, we evaluate the intra- and interannual variability of the upwelling index (UI) off the western coast of the Iberian Peninsula considering six locations at various latitudes along the 10°W meridian: Rias Baixas (42°N), Aveiro (41°N), Figueira da Foz (40°N), Cabo da Roca (39°N), Sines (38°N) and Sagres (37°N). In addition, the relationship between the variability of the occurrence of several circulation weather types (CWTs) and the UI variability along this coast was assessed in detail, allowing to discriminate which types are frequently associated with strong and weak upwelling activity. It is shown that upwelling activity is mostly driven by wind flow from the northern quadrant, for which the obtained correlation coefficients (for the N and NE types) are higher than 0.5 for the six considered locations.Taking into account these significant relationships, we then developed statistical multi-linear regression models to hindcast upwelling series (April-September) at the referred locations, using monthly CWTs frequencies as predictors. Modeled monthly series reproduce quite accurately observational data, explaining more than 60% of the total variance, presenting skill-scores against the climatology also above 60%, and having relatively small absolute errors. However, despite the ability of our models in representing the interannual variability of UI, they do not reproduce accurately most UI peaks, that occur typically in July. This may be due to the role played by mesoscale phenomena not represented in the statistical models, namely sea breezes that result from the intensified thermal low, which enhances coastal meridional winds and hence upwelling.