We thank the reviewer for her/his time and positive comments on the manuscript.

We agree that the study is a good example of the benefits of operational oceanography and we will reinforce this idea in the revised manuscript. We will also carefully check the English language.

Finally, we do also recognize the importance of analyzing these events in the longer term when multi-decadal kilometer-scale reanalysis simulations become available in the study area.

We are very grateful to the reviewer Dr Ricardo Torres for his time taken to carefully review the manuscript and for his constructive comments and suggestions.

We agree that both intensity and duration are important parameters for the operational definition of upwelling events. As illustrated in Figure 4, the intensity of an upwelling event may differ depending on the index used for its characterization, which constitutes a major difficulty for their identification. The intensity might be measured by the quantification of surface temperature gradients, offshore Ekman transport, offshore total transport, or alongshore currents, as described by the four indices considered in this study. In this short manuscript, we do not aim at providing a general operational definition of these upwelling events, which we think would require a dedicated analysis to carefully define specific thresholds for the different indices. Instead, we rather want to illustrate the sensitivity of this identification to the selected index, focusing on the comparison of the most extreme storm-induced cases detected over the last 9 years. As suggested, we will extend the discussion of this sensitivity in the conclusion of the revised manuscript. We will also add duration metrics of the three main extreme events (Jan 2017, Jan 2020 and Nov 2021), computed from the daily-averaged indices used in this work.  This will allow to provide elements for the future potential operational definition of such upwelling events.

Concerning the interpretation of ocean color images, we agree with the reviewer that they are certainly not exempt of uncertainties given the very coastal and cloudy situation which is considered. Unfortunately, there are no error estimates associated with these Sentinel 3 – OLCI CHL observations. The only uncertainty metric that is provided with the data product is the so-called quality index (QI), which measures, pixel by pixel, the difference between the observation and the climatological average, normalized by the climatolgical standard deviation. The QI maps for 4 and 7 November 2021 are included in the attached pdf.

On 7 November, in the area of enhanced Chla on the northwestern coast of Mallorca Island, the QI values range from 1 to 2, with a peak from 4 to 5 in the more coastal zone. Given the extreme conditions considered here and the very likely limitation of the climatological estimates in these coastal areas (the 300m-resolution Sentinel3-OLCI data are only available since 2016), these values seem realistic and, in our opinion, do not allow to identify evident errors in these observations.

Moeover, we agree with the reviewer that CHL observations in shallow coastal environments can be significantly affected by sediment resuspension. However, in our opinion, this is unlikely to be the case here due to the steep local topography that deepens to more than 500m a few km from the coast.  

We will include these elements in the revised manuscript, also tempering our interpretation of these ocean color images and highlighting the limitations due to the lack of any other complementary observations in the area during the study period.

In addition, following the reviewer´s suggestion, we have performed a short Lagrangian analysis to evaluate the hypothesis of a lateral transport of Chla from the northern Bay in Mallorca Island (Alcudia Bay) towards the upwelling area. The plots are provided in the attached pdf. The model didn’t allow any surface transport exporting particles outside the Alcudia Bay on the days before and during the upwelling event. These new inputs allow us to reasonably reject this hypothesis. We will add this additional information in a supplementary material associated with the revised manuscript.

With these elements at hand and within the limitations of both the model and observations, we believe that the spatial consistency between the observed patch of enhanced CHL and the model upwelling area is most probably indicative of an upwelling-induced local enhancement of CHL.

Re: “L233: I don't think that Figure 2 provides confidence in the realism of the numerical results. Comparison with HF radar would.”

The WMOP operational model predictions are assessed on a daily basis in the areas with available HF radar observations (Ibiza Channel, Ebro Delta and Gibraltar Strait; https://www.socib.es/?seccion=modelling&facility=wmedvalidation, Mourre et al., 2018). In particular, these predictions were proven to be realistic in the Ebro Delta region during storm Gloria (Sotillo et al., 2021). Unfortunately here, no HF radar data are available off the northwestern coast of Mallorca Island to perform such analysis.

Again, within the limitations of both the model and observations, we believe that the spatial consistency between the CHL enhancement area and the model upwelling area suggests that the model probably provides a reasonable representation of the spatial extension of the upwelling. We will clarify our statement in the revised manuscipt.

Finally, we will also incorporate all suggested corrections and carefully check the English language.