the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Sep 2026
Upper-ocean physical and biogeochemical response to Hurricane Norbert in the eastern tropical Pacific
Abstract. Tropical cyclones can strongly perturb the upper ocean through intense wind stress, vertical mixing, Ekman pumping, sea surface cooling, and nutrient injection into the euphotic zone. These processes enhance phytoplankton biomass generation and produce delayed chlorophyll-a responses after storm passage. Here, we investigate the physical and biogeochemical response of the Eastern Tropical Pacific to Hurricane Norbert, a category 4 hurricane that crossed the Mexican Pacific in October 2008. We combine tropical cyclone best-track data from IBTrACS, ERA5 wind forcing, CROCO physical-biogeochemical simulations, satellite-derived sea surface temperature and chlorophyll-a products, and Argo float profiles located near the hurricane track. The results show that Norbert generated strong positive Ekman pumping, mixed-layer deepening, surface nitrate enrichment, and a delayed increase in chlorophyll-a concentration along portions of the storm trajectory. The model reproduced the main structure of sea surface temperature cooling more clearly than the chlorophyll-a response, indicating that the physical response was more directly controlled by hurricane forcing, whereas the biological response was modulated by nutrient availability, pre-existing stratification, light conditions, and regional circulation. Argo profiles confirmed post-storm changes in the vertical thermal structure and mixed-layer depth. These results indicate that major hurricanes in the eastern tropical Pacific can generate significant but spatially heterogeneous biogeochemical responses through the coupling between wind-driven mixing, nutrient entrainment, and phytoplankton growth.
- Preprint
(24341 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4874', Anonymous Referee #1, 03 Oct 2026
-
RC2: 'Comment on egusphere-2026-4874', Anonymous Referee #2, 04 Oct 2026
This paper using a regional circulation-biogeochemical model (CROCO + bgc) to examine the biogeochemical response in the upper-ocean to passing tropical cyclones. The study focuses in the Eastern Tropical Pacific, targeting Hurricanes Fausto and Norbet, with the latter being the primary focus of the study. The authors commendably compare the simulated upper-ocean response to satellite SST and Chl-a and ARGO-profiles (T,S, density).
My primary issue with the paper is that I am not sure what the contribution is besides illustrating what seems to be a version of an expected response in a particular region. This issue is amplified by the many descriptive claims of controlling mechanisms that are not backed up by quantitative analysis; that is, the paper is mainly ‘descriptive speculation’. For example, diagnostics of a BGC tracer budgets or diagnosing the variability of nutrient and light limitations in the model would provide more clarity on the actual mechanisms driving the surface Chl response.
More generally, I was not sure what (new, mechanistic, model improvement) research question that manuscript sets out to answer, beyond what is stated on L52-53, which is too specific to an event, as opposed to seeking new mechanistic understanding or improving models / prediction. The comparison to observations is commendable, but the authors do not necessarily pinpoint where the model goes wrong, which could chart a path to improving simulations of the upper-ocean biogeochemical response to hurricanes. For example, the (rather vague)`reasons' for model-data disagreement on L286-289 are never actually investigated.
Recommendation: Based on the above summary, I think that the authors need to significantly revise the study, noting that these revisions need to include quantitative analysis to back up the authors’ claims. I believe this will take substantial work. I offer some primary suggestions for these revisions below, as well as other comments that I hope help to improve the study.
Primary suggestions / major comments
1. Replace descriptive speculation with quantitive analysis: There are many instances of descriptive speculation of controlling mechanism. These include: L191-192, L200-201, L262-263, L270, L283-284, L286-289, L305-306. Many of these speculative claims could be justified by more quantitative analysis in the simulation. For example, diagnosing the dominant contributions in the nitrate budget or diagnosing nutrient and light limitation in the model. This would provide more clarity on the simulated mechanisms, which could then provide new mechanistic insights or point to reasons for model-observation disagreement.
2. Re-frame study to make the novel contribution more clear: As mentioned above, I think the authors need to attempt to do more than what is stated on L52-53, which effectively sets out to tell a descriptive story of a particular event in a particular place. I think the authors could be well positioned to highlight some new or interesting mechanisms and/or areas for model improvements; I think framing the study in this manner would be more impactful, but it requires addressing the first major comment.
Other comments
1. The model description (both physics and bgc) needs more detail; the description provided does not enable reproducibility. For example, what are the boundary conditions? I would guess radiation-nudging, but this is not specified. Further, the choices for the model configuration, particularly resolution, are not justified. It is clear the authors want to resolve mesoscale variability; do the authors think that resolving submesoscale variability would change the simulated response? Also, are there caveats to using ERA5 for the wind-field? These choices need to justified more explicitly.
2. The Discussion section is one long paragraph. This should be broken up to help the reader.
3. I believe that CROCO uses potential temperature, not conservative temperature. So, the comparison with conservative temperature from the ARGO profiles (L115) may not be strictly “apples to apples” (although the differences between potential and conservative temperature here may be very small). I suggest the authors check / clarify this just to be sure.
4. Some of the writing feels like AI — a whiff of over-confidence in a summary claim that links all the pieces of together — although this is difficult to tell. Some of the lines that popped out to me are: L293-296, L325-327.
5. L242: typo “ranging”
6. L275-276: what else could the interpretation be? When do blooms occur in “unfavorable” oceanic conditions?
Citation: https://doi.org/10.5194/egusphere-2026-4874-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 292 | 93 | 53 | 438 | 50 | 45 |
- HTML: 292
- PDF: 93
- XML: 53
- Total: 438
- BibTeX: 50
- EndNote: 45
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General assessment
The manuscript investigates the upper-ocean physical and biogeochemical response to Hurricane Norbert (October 2008) using IBTrACS track data, ERA5 winds, CROCO physical–biogeochemical simulations, satellite SST and chlorophyll-a, and Argo profiles. The authors report positive Ekman pumping, mixed-layer deepening (~40 to ~70 m), surface nitrate enrichment (~6 mmol m⁻³), and a delayed chlorophyll-a response.
The topic is relevant and suitable for the journal, particularly because coupled hurricane–biogeochemical responses in the eastern tropical Pacific remain relatively under-studied. The combination of model, satellite, and Argo data is also useful. However, several of the main conclusions are based primarily on temporal ordering and spatial co-location rather than quantitative analysis. Some statements are also inconsistent with the figures. I therefore suggest that the following issues be addressed.
Major comments
The statement that SST is reproduced better than chlorophyll-a needs stronger support. For Norbert, the correlations are 0.59 for SST and 0.445 for chlorophyll-a, but for Elida–Fausto they are 0.71 and 0.705, respectively, which are essentially similar. Also, RMSE values expressed in different units cannot be directly compared. Please provide normalized metrics and/or uncertainty estimates, or soften this conclusion.
The landfall description appears inconsistent. Lines 66–67 mention landfall between Sinaloa and Sonora, whereas later the manuscript refers to the Baja California Peninsula, and Figs. 1–2 show the storm passing across/near the peninsula. Please clearly list the landfall locations and dates.
The CROCO configuration requires more information. Please provide details of the atmospheric forcing, surface fluxes, vertical mixing scheme, spin-up, boundary conditions, output frequency, and physical–biogeochemical coupling.
Please provide the exact OSTIA SST product, spatial resolution, and DOI. For chlorophyll-a, specify whether the product is L3, L4, or gap-filled/interpolated. Gap-free values during hurricane-related cloud cover should be clearly distinguished from direct satellite observations. Also define the SST climatological period used to calculate anomalies and clarify whether the same climatology is applied to the model.
The manuscript interprets the sequence of Ekman pumping/mixing → nitrate increase → chlorophyll-a increase as evidence of a hurricane-induced biological response. However, this is mainly based on spatial and temporal correspondence. Please provide a quantitative pre-/post-storm anomaly analysis or other statistical support. Otherwise, the causal statements should be softened.
The suggestion that Elida and Fausto preconditioned the ocean before Norbert is interesting but remains speculative. The authors themselves note that the contribution cannot be isolated. Please provide quantitative evidence of persistence from July to October or present this more clearly as a possible mechanism.
This statement is not quantitatively demonstrated. Norbert and Elida–Fausto differ in intensity, translation speed, timing, and background conditions. Please provide metrics such as affected area, maximum anomaly, duration, or integrated nitrate/chlorophyll response.
The Argo section needs clarification and more quantitative information. The sentence describing temperature changes “from ~23 to 28 °C” as a decrease is unclear. I suggest adding a small table containing float ID, date, distance and side relative to the hurricane track, time after passage, ΔSST/upper-ocean temperature, and MLD before and after the storm. Figure 9 also appears to show the strongest thermal response at approximately 20–60 m rather than only at the surface.
Specific comments
Recommendation
The study has potential and uses a useful combination of modelling and observations. However, the main physical–biogeochemical interpretations need stronger quantitative support, and several methodological, figure, and consistency issues should be addressed before the conclusions can be fully supported.