Here, we provided point-by-point responses to the comments of Reviewer 1.

Response to Reviewer 1

This study presents a valuable investigation into the impact of Mediterranean cyclones on biogeochemical dynamics, specifically Chl-a. The integrated multi-platform approach (reanalysis, ARGO, Sentinel-2) enables a reasonable examination of both surface and subsurface responses. The core finding is DCM uplift driven by slow-moving cyclones and its governing physical mechanisms, which has been well studied in many other ocean basins, though it is important for understanding Mediterranean productivity. The manuscript is generally well-focused, but requires substantial review from former studies to clarify the creativity.

Answer. Many thanks for your valuable comments that allowed us to improve the manuscript. Here, we reported a step-by-step response to your comments, and we have highlighted the associated changes in the main text.

Point 1. The most important issue is the fundamental dynamics, like the cyclone-induced dynamics and pre-cyclone condition. This topic has been widely discussed and investigated in former studies, but the current study was not well reviewed for these mechanisms. It is important to incorporate the findings from former studies and highlight the creativity of the current work.

Answer 1. We revised the manuscript considering former studies, mainly those considered for cyclone-induced dynamics and pre-cyclone condition. In particular, a discussion for DCM and MLD depths prior the cyclone onset was inserted, also considering the season for the different cyclones (lines 408-416).

In contrast, weak stratification during late autumn and winter facilitated easier mixing, but the response was less pronounced when the pre-storm MLD was already deep or nitrate levels were low, as observed during medicane Helios(D’Adderio et al., 2023). MLD shoaling was most pronounced in areas with initially shallow or seasonally variable MLD (Ricchi et al., 2019, 2020; Vargas-Yáñez et al., 2022). For instance, medicane Apollo (October 2021) induced a 25 m MLD shoaling in the central Mediterranean, where summer stratification was transitioning (Menna et al., 2023).

The most substantial Chl-a increases occurred in regions with abundant subsurface nitrate reservoirs, like in the Ionian Sea (Lazzari et al., 2016)), particularly during slow-moving medicanes like Zorbas, Apollo, and Ianos (Jangir et al., 2024; Kotta and Kitsiou, 2019b). By way of example, medicane Zorbas produced a strong biogeochemical response in the Ionian Sea, where moderate pre-storm stratification and high subsurface nitrate concentrations prevailed, but weaker effects occurred in the Peloponnese due to downwelling-favorable currents (Kassis & Varlas, 2021). Conversely, cyclone Ciprian triggered coastal upwelling near Cyprus (Fig. 14), where nitrate-rich waters enhanced the Chl-a response . In contrast, weaker or faster-moving cyclones (e.g., Erik-2015) had minimal effects in highly stratified or oligotrophic areas, demonstrating how pre-cyclone ocean conditions, combined with cyclone evolution, critically shape biogeochemical responses (Macías et al., 2014; Mélin et al., 2017; Menkes et al., 2016).”

Point 2. Clearly define the criteria used to classify cyclones as "Tropical-like" (TLCs) vs. "Extratropical" (ETCs) in this specific study context. Were hybrid types included? Referencing a standard classification scheme would be helpful.

Answer 2. The classification for each cyclone, in terms of TLCs vs ETCs, derives from literature works (see Table 1). Hence, we did not apply a criterion to differentiate them, but we only considered the oceanographic parameters of the cyclones that influenced the Chl-a dynamics. Note that the definition of medicanes has been provided recently (Miglietta et al., 2025), which includes hybrid cyclones.

Point 3. Justify the selection criteria for the 20 cyclones. What thresholds (e.g., intensity, duration, size) were applied? Were specific seasons or years favored? A table listing the cyclones, their type, key characteristics (max wind, min pressure, translation speed), and impacted regions would significantly enhance transparency and reproducibility.

Answer 3. To justify the criteria for the 20 cyclones, we considered the strongest medicanes and some of  the most intense extratropical cyclones already studied and discussed in literature (see Table 1). We inserted another table (new Table A1) in the appendix A1 that reports the key characteristics of the cyclones assessed from ERA5 (e.g., max wind, MSLP minima, translation speed). Furthermore, the impacted areas are shown in Supplementary Material S1, with a gray-shaded radius indicating the region influenced by wind stress.

Point 4. Provide explicit, quantitative definitions for "offshore" and "nearshore" zones used in the analysis (e.g., distance from coast, bathymetry thresholds). This is crucial for interpreting the comparative results.

Answer 4. A quantitative definition was inserted in subsection 2.2 (line 138). In this study, the nearshore was defined as the buffer zone from the coastline to the global average depth of closure of 13 m (Athanasiou et al., 2019). The coastline data were obtained from the European Environment Agency, and the depth data were extracted from the GEBCO bathymetric grid (GEBCO - The General Bathymetric Chart of the Oceans, 2020).

Point 5. The abstract highlights comparing offshore and nearshore responses, but the summary focuses primarily on DCM uplift (typically an offshore phenomenon). Elaborate on the key differences or similarities observed between these zones regarding Chl-a response (surface and subsurface), nitrate supply, and MLD changes. Was nearshore response dominated by different processes (e.g., sediment resuspension, terrestrial input)?

Answer 5. As discussed in Section 3.4, nearshore dynamics are primarily driven by runoff and surface currents. While upwelling also modulates Chl-a variability in this zone, satellite imagery clearly underscores the significant role of sediment resuspension and terrestrial inputs. This is evident in Figs. 9c–d, which shows a prominent plume generated by terrestrial discharge. Although terrestrial inputs appeared to be relevant in the nearshore, their influence has been reflected in the offshore too, in which the surface currents drove the transportation of floating sediments for several kilometers seaward.

Point 6. While wind-driven upwelling and air-sea heat exchange are identified as critical, can you quantify their relative contributions to the DCM uplift and resultant Chl-a changes (e.g., using correlation analysis, idealized model runs, or budget analysis within the reanalysis/ARGO data)?

Answer 6. A revised correlation analysis was reported in the Fig. 10, where the correlation with DCM uplift was calculated with  the duration of slow-moving phases, the Total Energy due to heat fluxes, and the total duration of the events. The main results of correlation analysis are discussed in section 4, where a positive correlation was observed between slow-moving phases and DCM uplift (Fig.10a), while a negative correlation was observed between total energy of heat flux and DCM uplift (Fig.10b). The positive correlation reflects a prolonged persistence of strong cyclones that can determine persistent upwelling, heavy rainfall in a given location, favoring vertical mixing in seawater and increasing phytoplankton blooms. Conversely, the negative correlation is related to cooling processes occurring during the intense phase of cyclones, such as upwelling, radiative heat loss, and vertical mixing.

Point 7. Discuss the potential influence of pre-storm oceanographic conditions (stratification strength, initial DCM depth/nitrate content, background MLD) on the magnitude of the observed response. Was the response consistent, or did it depend heavily on the pre-cyclone state?

Answer 7. A discussion concerning pre-cyclone conditions was inserted from line 396 to line 407:

“The concentration of Chl-a and DCM responses to Mediterranean cyclones was strongly modulated by pre-storm oceanographic conditions of the Mediterranean basin. Strong stratification during summer and early autumn typically inhibited vertical mixing, limiting nutrient uplift (D'Ortenzio & Ribera d'Alcalà, 2009). However, slow-moving cyclones (e.g., Zorbas, Ianos) overcame this barrier through prolonged wind-driven upwelling and turbulent mixing, leading to significant DCM uplift (40–60 m) and Chl-a increases (Jangir et al., 2024). In contrast, weak stratification during late autumn and winter facilitated easier mixing, but the response was less pronounced when the pre-storm MLD was already deep or nitrate levels were low, as observed during medicane Helios (February 2023; D'Adderio et al., 2023). MLD shoaling was most pronounced in areas with initially shallow or seasonally variable MLD. For instance, medicane Apollo (October 2021) induced a 25 m MLD shoaling in the central Mediterranean, where summer stratification was transitioning (Menna et al., 2023)”.

Point 8. Detail the typical temporal lag between cyclone passage, MLD deepening, DCM uplift, and the observable surface Chl-a response. Sentinel-2 coverage is weather-limited; how did this impact capturing the peak response? Did ARGO floats provide insight into the evolution?

Answer 8. A new sentence was inserted in the section 2 (line 100):

“To capture the full temporal evolution of biogeochemical parameters during the cyclone passage, we analyzed Sentinel-2 and Argo data alongside ERA5 (hourly data) and CMEMS (daily data) reanalysis products. This approach provided a comprehensive overview of the event’s progression, particularly revealing time-series variations in mixed layer depth (MLD) and deep chlorophyll maximum (DCM) dynamics”.

Sentinel-2 coverage is weather-limited, and ARGO floats provide point data on the water vertical profile. So, the integration with reanalysis dataset allowed us to obtain a full temporal evolution of the cyclone passage. However, the outputs are provided as daily data, so quantifying the time lag between cyclone passage and DCM uplift (from the CMEMS reanalysis) is not possible. This is especially true given the intensity parameters and slow-moving phases assessed from the ERA5 hourly data.

Point 9. Discuss the persistence of the Chl-a signal. Was the uplift transient, or did it lead to a sustained bloom?

Answer 9. Chl-a signal resulted to be characterized by a sustained bloom, as also highlighted in other studies (Kotta and Kitsiou, 2009; Kotta et al. 2017). On the other hand, the ARGO data revealed that DCM uplift persisted also after the cyclolysis.

Point 10. Acknowledge the limitations of Sentinel-2 for Chl-a retrieval in the Mediterranean, particularly in offshore oligotrophic waters where algorithms are less robust and in the immediate aftermath of storms (high turbidity, cloud cover, sun glint). How were these challenges addressed (e.g., specific algorithm choice, masking, validation against ARGO near-surface data)? How did data availability impact the analysis for the 20 events?

Answer 10. As described in Section 2.2, chlorophyll-a (Chl-a) was retrieved from satellite imagery using the C2RCC processor in ESA's SNAP software. This processor employs a neural-network-based atmospheric correction and inherent optical property (IOP) inversion algorithm to estimate water constituents in optically complex waters. Prior to Chl-a retrieval, we preprocessed Sentinel-2 Level-1C top-of-atmosphere (TOA) radiance data with C2RCC, which utilizes a bio-optical model to derive water-leaving reflectances before estimating Chl-a concentrations through IOP inversion. To minimize artifacts from high turbidity, cloud cover, sun glint, and other disturbances, we selected images from periods before cyclone onset and after cyclosis to isolate changes attributable to cyclone passage.

Point 11. Consider replacing "Hurricanes" with "Cyclones" for broader scientific accuracy, as "Medicane" is the common term for the TLC subset, and ETCs are also studied. (e.g., "Chlorophyll-a Response to Mediterranean Cyclones").

Answer 11. Correction was made as you suggested.

Point 12. Distinguish between the physical uplift of the existing DCM community and potential new production fueled by the injected nutrients. The observed Chl-a increase could be due to either or both. Can the data shed light on this?

Answer 12. Satellite imagery highlights Chl-a variations induced by nutrient injection, mainly in the coastal areas characterized by river plumes, which determined resuspension of sediments nearshore waters (see Fig. 9d). In contrast, the observed DCM uplift in offshore waters stems exclusively from oceanographic forcing by cyclone passage. Notably, satellite-derived Chl-a levels were strongly influenced by suspended organic matter (reaching up to 20 mg/m³), which masked the Chl-a signature associated with purely oceanographic changes.

Point 13. Briefly discuss how the observed responses compare to those documented for tropical cyclones in other basins. What makes the Mediterranean response unique or similar?

Answer 13. New sentences were inserted in the discussion section about this aspect in lines 431-440:

“Like tropical cyclones, Mediterranean cyclones induce DCM uplift through wind-driven mixing and upwelling, bringing nutrients to the euphotic zone. This aligns with observations in the Pacific and Atlantic, where cyclones enhance primary productivity via vertical mixing (Lin, 2012; Zhao et al., 2008). The dome-shaped uplift of biogeochemical parameters (e.g., MLD shoaling, nitrate enrichment) near the cyclone center mirrors patterns seen in tropical cyclones (Price, 1981; Sanford et al., 1987; Walker et al., 2005). Mediterranean cyclones (especially medicanes) are smaller and shorter-lived than tropical cyclones, yet their slow movement (e.g., Zorbas: 11.5 km/h) prolongs mixing, creating localized but intense biogeochemical responses. Tropical cyclones typically cover larger areas with faster translation speeds (e.g., 20–30 km/h; (Fu et al., 2014)).Mediterranean coastal upwelling (e.g., during Blas or Ciprian) is often wind-driven and localized, unlike large-scale equatorial upwelling zones affected by tropical cyclones”.

Point 14. Expand on the "significant implications for nutrient cycling and primary productivity." How might these pulsed events contribute to annual budgets? Could they influence community structure?

Answer 14. Some sentence are inserted in the following lines (464-476):

“Furthermore, the Mediterranean Sea is nutrient-limited, with surface Chl-a typically ranging from 0.05–0.3 mg/m³ (Marty and Chiavérini, 2002; Moutzouris-Sidiris and Topouzelis, 2021; Teruzzi et al., 2021). Cyclone occurrence could disrupt the water column stratification, injecting nitrate (0.5–0.8 mmol/m³) and phosphate into the euphotic zone (Fig. 5, 7). This pulsed nutrient supply can be particularly significant in late summer and autumn, when the Mediterranean’s stratified surface waters are most nutrient-depleted (D’Ortenzio et al., 2005; D’Ortenzio and Ribera d’Alcalà, 2009). For instance, slow-moving medicanes, like Ianos (September 2020), drove 30–60 m uplifts of DCM, enhancing Chl-a concentrations by 0.01–0.02 mg/m³—comparable to the effects of seasonal riverine inputs in coastal zones(Ludwig et al., 2009).

These nutrient pulses can shift plankton community composition, favoring larger, fast-growing diatoms over smaller picoplankton that dominate under oligotrophic conditions (Marty and Chiavérini, 2002). Such shifts have been observed following medicane Apollo (2021), where diatom blooms coincided with DCM shoaling (Menna et al., 2023). This transition from microbial loop-dominated recycling to a classical diatom-copepod food web can enhance carbon export efficiency (Chen et al., 2017; Cullen, 2015), potentially offsetting productivity declines associated with marine heatwaves (Androulidakis and Pytharoulis, 2025; Darmaraki et al., 2019; Izquierdo et al., 2022; Li et al., 2024a).

The cumulative impact of cyclones on annual productivity budgets is likely modest (≤5–10% in affected areas), but their role in late-season nutrient injection and carbon export could grow under climate change, especially if warming SSTs intensify cyclone-driven mixing (Avolio et al., 2024; González-Alemán et al., 2019). Future studies should quantify how these event-scale perturbations integrate into broader biogeochemical cycles and whether they enhance ecosystem resilience to stratification.”

Point 15. Consider to add a schematic of mechanism figure and a composite map of Chl-a/MLD anomalies.

Answer 15. A schematic representation of DCM, MLD anomalies and currents is reported in the graphical abstract. A figure for DCM and Chl-a changes was inserted in Fig. 9 and a schematic representation was already inserted in Fig. 12. Here, you can observe the component of MLD anomalies in the cross-section (Fig. 12b) and the DCM uplift (both in the cross-section and also in the vertical profiles).

Point 16. The methods section must provide sufficient detail on data processing, matchup procedures, and statistical analyses.

Answer 16. The Methods section has been refined to more clearly present the data processing and matchup procedures. Key improvements include detailed descriptions of the methodology used to identify slow-moving cyclone phases and calculate translation speeds from ERA5 reanalysis data, with specific criteria defined for slow movement phases (displacements <36 km over 4 hours). The analysis of physical parameters was expanded to incorporate heat fluxes, mixed layer depth, and current velocities from Copernicus reanalysis products, along with the calculation of Total Energy through double integration of surface heat flux to quantify air-sea interactions. Statistical validation procedures were enhanced to systematically present Pearson correlation analyses (e.g., r = -0.64 for DCM uplift versus heat loss) and include RMSE validation against ARGO float observations, ensuring robust verification of the integrated reanalysis-observational approach. In our opinion, these methodological refinements improve both the clarity and reproducibility of the study's analytical framework.

Point 17. The legends for colormap in Figure 1 and many other figures are wrong because there is continuous color being used in the figure.

Answer 17. The legends in the figures have been updated.

Point 18. Some text are presented in different fonts.

Answer 18. The fonts in the text have been updated.

Here, we reported point-by-point responses to the Reviewer 2.

Response to Reviewer 2

In this study, the authors assess the response of biogeochemistry to Mediterranean cyclones, examining the response of the nutricline, MLD, and DCM to these storm events and relating changes to physical drivers. The manuscript represents an impressive body of work bringing together reanalysis products, multispectral satellite imagery, and ARGO measurements from twenty storms across the Mediterranean basin.



In summarising the variability in responses of biogeochemistry to Mediterranean cyclones (both between cyclones and across sections of individual storms), I feel the manuscript has the potential to make a valuable contribution to the field. Although the description of varied responses to these storms is not novel, nor the description of dome-shaped cross-sections of biogeochemical parameters, the manuscript makes a strong case for the importance of heat fluxes as a major driver, and makes an interesting comparison between nearshore and offshore responses.

At present, I feel there are aspects of the methodology that would benefit from clarification, and elements of the introduction and discussion that could be expanded on to better emphasise the key results of this work.

Answer . Many thanks for your revisions that allowed us to improve the manuscript. We followed your comments to improve the methodology description and the introduction and discussion sections.

Major comments

Methodology

Point 1. In general, explicitly defining several aspects of the methodology would strengthen transparency. I can see R1 has already mentioned pre-cyclone conditions, classifying storms, why storms were chosen, offshore/nearshore. I agree that these are important details and that a table might be a useful place to present some of these.

Answer 1. Table 1 reported the features of Mediterranean cyclones considered in this study. To better highlight the physical features of the Mediterranean cyclones, we reported a new Table in appendix A1 where some important parameters were assessed from ERA5. In particular, we focused on the total duration of cyclone lifetime and number of hours related only to the slow-movement phases.

Point 2. I would also ask how exactly is a cyclone-impacted area defined- where is the border or edge? The exact area used will be important for reproducing your energy change calculations.

Answer 2. The cyclone-impacted areas were selected considered the literature reported in Table 1. The DCM and Chl-a changes were applied at a regional scale, following the track of the Mediterranean cyclones reported in table 1. Subsequently, the maximum 10-meter wind speed within a 200 km radius from the cyclone center was extracted for each time step, in order to assess the cyclone-affected areas. A new sentence was inserted to specify the area selection (line 101).

Point 3. Some discussion of the uncertainties and limitations associated with the various data products used would be welcome.

Answer 3. Limitations and uncertainties of the various products were discussed through RMSE assessment of between Chl-a reported in reanalysis products and Chl-a profiles measured from ARGO-float (Supplementary Table S1). The RMSE values were below the reanalysis vertical profile resolution (3 m). Nevertheless, for medicanes Ianos and Blas, the RMSE showed larger discrepancies in DCM depth, which were greater than 1 m.

Discussion

Point 4. With the impressive and wide-ranging dataset there are several implications and points of discussion one could pursue. I understand that the authors may consider some of these questions out of the scope of this manuscript, but some points that could perhaps be elaborated on without too much further data analysis:

Answer 4. We revised several points in the discussion following your comments.

Point 5. Discuss the persistence of the Chl-a signal. How long after the passing of the cyclone does elevated Chl-a and a shallower DCM persist?

Answer 5. Analysis of ARGO-float and CMEMS reanalysis revealed that the DCM uplift persisted after the cyclone’ passage for for several days. Some discussion sentences were inserted also citing the works of Kotta et al. 2023; Kotta et al., 2017.

Point 6. Regarding the observation of offshore transport of organic matter, for instance with medicane Zorbas. Is this a general feature of mediterranean cyclones-, something that would happen always, or something specific to Zorbas.

Answer 6. This feature appeared to be specific only to Zorbas. The offshore transport appeared to be related to some oceanographic conditions caused by current patterns in southeastern Sicily in response to the medicane Zorbas impact. A system of gyre-like currents developed offshore of southeastern Sicily determining the Chl-a increase in the offshore area.

Point 7. In general, from L264-275, the nearshore responses are presented as general features following cyclones either with or without rain, but only two examples are presented (Zorbas and Daniel). Could you clarify this section to make clear either that these responses sometimes occur, or that rain almost always = plume, and without rain = offshore transport.

Answer 7. While Zorbas and Daniel highlight contrasting responses, intermediate behaviors were also observed during other events (e.g., Blas and Ciprian; Fig. 14), where wind-driven upwelling combined with moderate rainfall enhanced Chl-a locally. Thus, nearshore responses depend on the balance between precipitation-driven runoff and wind-driven transport, with plume dominance under heavy rainfall and offshore transport prevailing in arid regions or dry cyclone phases.

Point 8. Can you provide further discussion as to why your result for Zorbas differs to Kassis and Varlas (2021) in terms of MLD? They report a clear deepening of MLD and you report a shoaling of DCM in centre at least. Do they consider different timescales, spatial area affected by cyclone, or are the differences methodological? L246-250 I think you touch on this briefly but do not cite Kassis and Varlas here. Resolving any differences in results, if any, seems important since MLD and DCM changes are a key aspect of the paper.

Answer 8. Reanalysis and ARGO validation support DCM uplift in the core, while Kassis & Varlas (2021) highlight regional MLD variability influenced by broader oceanic processes during medicane Zorbas.

Suggested text revisions are reported from the line 285:

“Kassis and Varlas (2021) noted that Atlantic water intrusion could have disrupted local upwelling, leading to net MLD deepening in offshore areas of southeastern Sicily and the Pelopponese. Chl-a signal resulted to be characterized by a sustained bloom, as also highlighted in other studies (Kotta, 2023; Kotta et al., 2017; Kotta and Kitsiou, 2019b). On the other hand, the ARGO data revealed that DCM uplift persisted also after the cyclolysis in the center of the cyclone.”

Other general comments

Point 10. Change title to encompass both ECs and TLCs

Answer 10. Title was revised.

Point 11. In introduction: L55 I would change “This is due to the fact” to something along the lines of “one reason/one source of uncertainty”. Phytoplankton dynamics is a broad term and there are many sources of uncertainty, not just the role played by cyclones.

Answer 11. The sentence was revised.



 Point 12. In addition, I felt the section could be slightly rearranged to emphasise the variable responses to cyclones as a source of uncertainty. When I read L55, I find myself questioning why the response of cyclones are not understood, and it is not until I reach line 60 “Severe weather events…”  that I understand why this is. I would merge these sentences and shift text between re. DCM, nutricline, nutrients upwards into the section describing Mediterranean Sea

Answer 12. Changes in the sentences were performed following your comment.

Point 13. Figure 10: Is clear, but some sort of line at 0m uplift would really help highlight which unusual cyclones see DCM uplift, and that these are the ones associated with lower heat transfer.

Answer 13. The correlation graphs of Figure 10 were revised.