|Reviewer comments to revised manuscript by J. Hieronymus, K. Eilola, M. Hieronymus, H.E.M. Meier, and S. Saraiva renamed now as “Causes of simulated long-term changes in phytoplankton biomass in the Baltic Proper: A wavelet analysis” and re-submitted to “Biogeosciences”|
The authors have done an extensive revision accepting most of my minor and language suggestions and I am rather satisfied with these results. Unfortunately, the following major concerns had been addressed less satisfactorily. Therefore, I would recommend a further major revision.
1. In the revised version, it became even clearer that the manuscript deals with a study of the model itself rather than with a usage of the model for studying the marine system. As the authors admitted: “Much of the JUSTIFACATION certainly boils down to the USE of a relatively NEW TOOL. …we feel that an ILLUSTRATION OF ITS USES is valuable.” As deficit of geographical and biogeochemical plausibility cannot be compensated by methodological novelty of the implemented statistical tool (wavelet analysis), such study is more appropriate, perhaps, for the journals dealing with modelling technics rather than in Biogeosciences (see my initial comments).
2. The revisions concerning nutrient limitation are still confusing and misleading in respect to the real Baltic as we know it from both observations and simulations with other models. No consideration is given to how an appropriate re-parameterization of internally inconsistent N-limitation > 1 would change relations between NLIM and PLIM for different phytoplankton groups. More importantly, the chosen method compares N- and P-limitation as determined by a specific set of constants prescribed in result of calibration of this specific model, while a sensitivity of simulation to such a choice is not even mentioned. The manuscript still contains a little addition in- , perhaps, even directly contradicts to existing knowledge about the relationships between nutrient and phytoplankton dynamics in the Baltic Proper. In that respect, particularly surprisingly and unconvincingly sound such expressions as “… a shift towards less limited conditions… “ (Line223), “Phosphate is still limiting during winter…” (L230), “… the phase shifts from NUTLIM preceding diatoms by three months to diatoms preceding nutlim by the same amount” (L245-246) and the following considerations of maximal NUTLIM on Fig.11 (shouldn’t minimum NUTLIM be more interesting?), “The spring bloom is phosphate limited throughout the run except for a few years after 1990 where diatoms display nitrogen limitation” (L304-305). Correspondingly, the entire Section 3.2, including Figs. 6-11 still looks just as exercise with a new tool, being more concerned with the tool rather than the Baltic. Perhaps, it can be focused only on analysis of N:P instead of NUTLIM and substantially shortened, absorbing and compressing much from the Summary and Conclusions Section as well.
3. The laconic Section on river loads looks now better. Although, the strong inter-annual coherence (with only 1 year lag?) between local riverine input and DIN in the mixed layer (L255) deserves more consideration and explanation, remembering although about open boundaries with the gulfs and the south-western Baltic. On the other hand, the anti-phase coherence between salinity and DIN on periodicities >1 yr, might has the same reason as the in-phase coherence between salinity and phosphate, that is the upward transport of deep waters due to hypoxia enriched with DIP and depleted in DIN. Could be worth mentioning as pertinent to the vicious circle of Baltic Sea eutrophication that is more appropriate here than in the Section on limitation.
4. The “Summary and conclusion” Section contains too much repetition or just a prolongation of the awkward discussion from the preceding “Results and discussion” Section, including even some contradictions with dates and chronology. It must be effectively cut down to just the conclusions.
L32 – “internal loads”?!
L58 – Should it be: “All variables, COMPUTED WITH 3D RESOLUTION, have thus been horizontally averaged over the study area.”, as in a post-processing? Then, perhaps, it better begin with the models (present Section 2.2) and after that indicate the study area
L84-85 “The rate of nitrogen fixation as a function of the phosphate concentration and temperature” should, perhaps, read “…fixation NFIX IS a function…”. Besides, what is more important, NFIX is also a function of N:P ratio (see, e.g. Eilola et al., 2009)
L124 – “…, which accounts for photo-inhibition” could be a good addition
L140-141 – The description of Fig. 2 does not correspond to Fig. 2 itself
L153, 162 – “References” do not explain or highlight, studies and papers do
L295-296 - The combined effect results in nitrogen limitation after the spring bloom which benefits cyanobacteria” – nitrogen limitation does not benefit any algae including blue-greens, phosphorus excess does.