Anammox-Based Processes for Mature Leachate Treatment in SBR: A Modelling Study

Round 1

Reviewer 1 Report

In this work two real scale sequencing batch reactors (SBRs) operated under different conditions (PN/A and SND processes), for the treatment of high-strength ammonium urban landfill leachate, were modelled using BioWin 6.0. It is interesting and meaningful. The manuscript has been generally well structured. Research motivations and the novelties of the study have been well defined. Materials and methods section have given enough information; experimental results have been discussed in more details with relevant references. However, a minor revision is still needed before publication.

1. Why there are not comparison results between PN/A and SND processes in both abstract and conclusions, which is emphasized in the introduction. Please add it if possible.
2. There are many errors in the article, please check, Line 193, Line 355-358, Line 362, Line 406, Line 441, Line 518.

Author Response

Dear Reviewer #1, thanks for your interest in reviewing our manuscript. Please see the attachment. 

Author Response File: Author Response.docx

Reviewer 2 Report

Processes 1284292 “Anammox-based processes for mature leachate treatment in SBR: a modelling study”

This manuscript performed model simulation with actual measurement data of full-scale leachate treatment plant. The data of full-scale plants would be valuable and interesting topic for the audience, but I think that the information of the experimental method is partial, and it is difficult to make an accurate assessment of the interpretation of the results.

Here I am writing some comments for improving.

There is no big difference between SBRa and SBRb for nitrogen removal performance, although operation condition was much different, and I'm not sure the reason of it  and why anamox can be retained in SBRb. Is it the effect of the past DEMON process? 

Was the sludge granule form that tolerate to high DO concentration?

Could the model explain this phenomenon?

L111

The plants were operated with DEMON process since 2012 to 2015, then they were operated PN/AMX or nitrification/denitrification mode, then investigated from 2016 to 2017 for this study. Could you describe operation condition DEMON process? SBRb was operated with high DO concentration but it maintained high concentration of anammox (as simulation result). I suppose that the past DEMON process accumulated anammox and effected SBRb performance in the test periods.

I also have a question about initial biomass assumption for simulation. How did you set it with consideration for anammox biomass?

L139

If influent volume 29.7 m3/cycle was add to total volume 906 m3, and if two cycle/day was operated, HRT should be 15 day????

L362-

I’m interested in FA concentrations in actual data. Could you add the FA data and discussion with inhibition parameters used in BioWin?

L442-445

Just repeat of L407-409?

L542

Was the sCOD checked reproductivity by simulation?

Table 1

NLR is used with unit (kgN/m3/day) in text.

(m3/cycle) is lacking for Buttermilk, in SBRb.

There is no explanation for abbreviations for Vfill, Aeration_i, Mixing_i.

Table 3

Fbs in III period is 10 times higher than I and II period, please add the reason of this value. Is “n. d.” in Table 2 “not done” or “not detected”?

Table 5

Are parameters for anammox not modified??

Figure 2, 3

I fell strange with these figures,

In Fig 2-b, d , even DO concentration is high as 2-3 mg/L, nitrate concentration was very low.

On the contrary, Fig 2-a, c Fig 3-a nitrite concentration was very low, why??

I’m interested in change of DO concentrations in these track study. Could you show it?

And, show the legend about arrows with C mean.

Fig. 5-c

N2 production peak in SBRa is not consistent with anoxic period in Fig. 3-a. check it again, please.

There are many strange words “In Error! Not a valid.......”. I couldn't read the text correctly.

Author Response

Dear Reviewer #2, thanks for your critical comment on our manuscript. Please see the attachment. 

Author Response File: Author Response.docx

Reviewer 3 Report

Anammox-based processes are being intensively studied for energy-efficient and sustainable nitrogen removal from wastewater. In this study, two full-scale sequencing batch reactors (SBRs) operated under different conditions, for the treatment of high-strength ammonium urban landfill leachate, were presented and modeled using a commercial software. The model was calibrated and validated, based on which the contribution of anammox and heterotrophic denitrifiers on N revmoal was further discussed, as well as the potential N2O emission. The manuscript is overall well written and clear. However, some critical information is missing, and the discussions should be further elaborated, as detailed below.

Major comments:

1. On model calibration: it was not clear how the best-fit parameters were identified. The authors should add more information on this aspect.
2. The models were calibrated with two track studies (one SBR cycle) in period I and II. Where is the data of the long-term reactor performance of both SBRs? Could the model capture the trend of these data?
3. Could the authors justify the significantly lower decay rate of AOB, NOB, and AOA used for the calibrated model?
4. Was the PN/A (SBR_A) modelled as granular sludge process? If yes, better give some brief description on
5. Based on the model prediction, the biomass concentration (and thus the fraction of) OHO increased significantly during period III in both SBRs (Fig. 4). Since period III only lasted for around 40 days, would the authors expect the washout of anammox bacteria under the conditions of period III (lower temperate and elevated COD/TAN) for a longer period? If yes, could PN/A still be used for mature leachate treatment under these conditions?
6. The authors referred to possible FA inhibition for NOB suppression in the studied SBRs. Was FA inhibition modelled in your model? If yes, better give more information on this.
7. SBR_B was intended for the SPND process, but the model predicted a few times higher contribution of anammox than denitrification on N2 production (Table 6). How does this prediction relate to the real data from the corresponding reactor (e.g., did the authors detect the presence of anammox bacteria by microbial community analysis or anammox activity by batch tests?)?
8. The authors tried to calibrate and validate the model, a logical and more interesting next step seems to be revealing the impact of operating conditions (e.g., cycle length, DO, temperature, COD/TAN) on the process performance. It is just a suggestion for consideration.
9. The default N2O model in BioWin 6.0 is not entirely mechanistic; the three major pathways are not modelled explicitly and in detail. The authors also did not present any N2O measurement data, so it is better also to point out the potential uncertainty of these simulation results.
10. The N2O was either produced by OHO or AOB in the model, so it should be possible the estimate the contribution of AOB in the modelled N2O production, just as what the authors did for OHO (Line 524-531). Right?
11. Other research on PN/A for leachate treatment should also be included in the introduction and discussion.

Minor comments:

11. Line 193: Error. Also check the error throughout the manuscript
12. Line 228: Define TNN first.
13. Figure 3: There are no Fig. 3c and 3d.
14. 5: Adding the corresponding operation phase or conditions within the cycle would make the figure much easier to read and follow.

Comments for author File: Comments.docx

Author Response

Dear Reviewer, thanks for your interest in reviewing our manuscript. Please see the attachment. 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I confirmed the answer from the author.
It is acceptable, but in new Table 6, I think that several data of INH3 do not fit to value calculated form NH3 concentrations and 0.075 mmol/L (KiNH3). Check them again.
Table S2 is not referenced in the main text, I think it should be described in L 417.

Author Response

Thanks for the careful review.

Point 1

It is acceptable, but in new Table 6, I think that several data of INH3 do not fit to value calculated form NH3 concentrations and 0.075 mmol/L (KiNH3). Check them again.

Dear Reviewer #2, thanks for your comment. In the original paper in Table 6 we reported the minimum, medium and maximum value of the peaks of FA concentration in SBRs cycle, while we calculated the I_NH3 coonsidering the minimum, medium and maximum values of the FA average value of each cycle. We agree with you that this make confusion. Thus, in the revised paper we calculated the INH3 starting from the FA concentration reported in the table. We also modified text before table 6.

Point 2

Table S2 is not referenced in the main text, I think it should be described in L 417.

Thanks. In the reviewed paper we reported reference of Table S2 in L 417.

Reviewer 3 Report

The authors have adequately addressed the comments. Congrats.

Author Response

Thanks for the careful review