Removal of Phosphorus from Hypolimnetic Lake Water by Reactive Filter Material in a Recirculating System—Laboratory Trial

Round 1

Reviewer 1 Report

This paper includes a simple but clear research on phosphorus removal from bottom P-rich water by sort of extracorporal dialysis technology. This research design is very clear, concise and straightforward, yet problem-application-oriented.  Although still many problems should be addressed in the future full-scale application, such as long term effect, stratification dynamics, algal and even other metal interactions. As preliminary results, this paper still has its own merits. The results are also straightforward and concise and discussion is reasonable.

Several questions here are given for revise:

1. Tank A and B should be equal at the start point by the design. But it seems still some discrepancy in pH, DO and other parameters. Should you explain why?
2. The background of physico-chemical parameters for bottom water should be included. such as the DO, pH, phosphorus, and other nutrients. This is very important for make correct design of the experiment system. For example, why need to bubble the N2 in the water to maintain a very low oxygen condition? Why the phosphorus concentrations represent the in-situ condition?
3. Why  Polonite filter is a proper reactive medium? It would be interesting to show how the authors choose this reactive medium. As there are so many commercial reactive medium, like the paper said, there should be some principle to choose. Does phosphorus is the only problem for lake Hönsan? Is this Polonite effect exclusive? As we known, there are many, like Phos-Lock.
4.  L184， To the best of our knowledge, this is the first time that Polonite was tested for P- removal from hypolimnetic water. If this is indeed the first time to test Polonite, comparison should be included for different RM. The effective and disavantages should be useful for future selection of RM.
5. in section 3.2, pH arise was found in this experimental system. If pH could arise as high as 9.5, the sedimental phosphours compostion should have changes, especially Fe-P, org-P. These species were recognized as pH-sensitive. Normally, high pH will make more labile P. 
6. Table 2, 1-12 represent sampling intervals. please clarify which day for the number
7. please clarify what exact meaning for Table 2. The DO seems co-controlled by N2 bubble and sediment oxygen demand. I am not sure why the DO changes means something.

Author Response

1. We tried to get exactly the same conditions in both containers. However, the biggest problem was the sediment that we did not mix because we were afraid to disturb the natural conditions and oxygenate it. We included in the revised version, data on the starting conditions (as other reviewer asked for)
2. It is included. We needed to bubble some N2 to keep anoxic conditions on the bottom. (we learned from an experiment with students!) If not, the P release from sediment would change  and not represent conditions in the lake from where it was taken. P concentration, we wanted to have a similar of that in the lake.
3. Of course we could have tested at the same time several filter materials. The problem was that we decided to have real hypolimnetic water and real sediment from the lake Hönsan (185 km from the university). The transport and our economy at that time of the experiment did not allow us to do so. Yes, phosphorus is the only problem in lake Hönsan as well as many other lakes close to that and other areas in central Sweden. We took Polonite because we have all background about this material. In the semi-scale test now on-going at Lake Hönsan we have also a slag material (AOD). Phoslock is produced as powder as far as I know and cannot be used as a granular filter media.  It is also very expensive (3x Polonite)
4. We removed this sentence.
5. Yes, we know that. We have improved discussion about this issue. In this simple experiment we did not see any effect of the pH change. Maybe if the pumping continued one month more we would have seen some change.
6. We revised tables and decided to use bed volumes throughout
7. We revised text about DO. However DO is important to check, as we said, oxygen in the bottom sediment would have stopped/decreased the P release

Reviewer 2 Report

This study presented the phosphorus removal efficiency of the reactive filter material Polonit when used in the treatment of hypolimnetic water in a closed-circuit system. I like the interesting work. It is a good idea.

Author Response

Thank you for your kind respons on our manuscript.

Reviewer 3 Report

The article deals with an important topic of eutrophic lakes restoration. The subject of the article is within the scope of the journal. I believe some improvements need to be done in the manuscript. I listed them below as general and more specific comments.

General comments

• The proposed method needs to be described more precisely just at the very beginning when it is first mentioned. I believe the method was not put in the whole context of the newly forming approach to hypolimnetic withdrawal as a lake restoration technology. The advantages of the method proposed were not clearly described.
• I believe the results should be complemented and presented in a more comprehensive and consistent manner. 1. I believe the results for both, P removal efficiency and PO4-P concentrations (in the containers and in the water treated on the column (before it reached the container)) should be presented, to provide a comprehensive picture of the system operation. The more that the Authors refer to some threshold values (line 195-197) and compare them with water treated on the column (but these data are not shown). 2) I suggest to describe and discuss all the results for P (removal efficiency, resulting concentrations and so on) in one section. In the current manuscript the PO4-P data are splited into 2 sections which makes the reader confused (at least it confused me). 3) The pH and the other data (DO, temperature) should be rather shown and described in a separate section(s). The data for pH for the container (averages) are presented twice (Fig. 3 and Table 2). This is superfluous. 4) The terms effluent and influent were mistaken (at least once, see specific comments).
• I suggest to reorganize the text in section R&D somehow to systematize discussion elements related to the result implications. In lines 297 – 316 the conclusions are not results based and could be drawn without conducting the research presented in the manuscript. So they are not really implications although they were placed in section 3.4. entitled “….and implications”. Implications of the results were discussed in previous sections (eg. lines 198-205, 272-280). So I suggest to either group the implications (resulting from the results presented in this study) into one section called “Implications…” or make the lines 297-316 a section devoted to problems which need to be solved.

Specific comments

• Line 28 – `or` instead of `and`?
• Line 29-31 - This sentence is intended to introduce the subject of phosphorus control methods in lakes. However, it refers to the rarely used method of introducing coagulants directly into the sediments. In practice, however, it is typically the dosing of coagulants or other phosphorus-binding materials into the water column (eg. Dondajewska et al., 2002, https://doi.org/10.1007/978-3-030-12139-6_17; Dithmer et al., 2016, https://doi.org/10.1016/j.watres.2016.02.011; Zamparas & Zacharias 2014, https://doi.org/10.1016/j.scitotenv.2014.07.076). I would suggest to mention first of all the widely used methods, as this would be more representative for the state of the art.
• Lines 33 – The same approach was described by Nurnberg, 2020, https://doi.org/10.1007/s10750-019-04094-z and Silvonen et al., 2021, https://doi.org/10.1016/j.scitotenv.2020.143202. Silvonen et al., 2021 called it “closed circuit approach” and it would be worth to mention this term when introducing the methodological approach. Similar method, without redirecting the hypolimnetic water back into the lake, as an improvement of the traditional HW method, was also discussed by Łożyńska et al., 2021, https://doi.org/10.3390/min11020098, together with focus on P recycling. These works should be mentioned to show the full context of the newly developing approach to HW method and P stripping from hypolimnion.
• Line 44-48 – Please specify the solution being proposed. What is the difference between the newly proposed solution and the solutions previously proposed for hypolimnetic water treatment (lines 31-33)? What are the advantages? Please specify in more detail what is supposed to happen with the treated water. This is not clearly claimed here and I was a bit confused. Does the method assume discharge of the treated water into the downstream water body or back to the lake? Back to the hypolimnion? Should the method, when upscaled for the field, be based on hypolimnion water pumping? This should be cleared here in this paragraph. In this text piece the importance of P recycling is stressed. However, the paper does not deal with this topic at all. How do the Authors assess P recycling potential taking into account the Polonite material used and amount of P possible to be fixed under hypolimnion relevant conditions?
• Line 60-62 – Why is the value “below 1000 micrograms” important if the hypolimnetic water has very often higher concentration of P-PO4?
• Line 67 – There was an earlier study on P stripping from hypolimnetic water on different materials which could be mentioned in this literature review, too (Łożyńska et al., 2020, https://doi.org/10.1016/j.ecohyd.2019.12.003).
• Line 71-73 – The information on the diffusion adjusted hypolimnetic withdrawal (HW) and its potential does not refer to the problem of P stripping from the hypolimnetic water (and this problem together with latest studies is described in this paragraph). Therefore I suggest to add the information on diffusion adjusted HW somewhere in line 35, where the method of HW is introduced.
• Line 76 – As a reader I missed in this point some very short information on the Polonite material. Even though the material is well studied and generally known, I recommend adding here a short info about this material, how it is produced, just to give some general idea on that in this point of the paper.
• Line 78 – Couldn`t be predicted without this study that pH will increase if an alkaline material will be used for treatment of hypolimnetic water which is by definition more or less neutral (from slightly acidic to slightly alkaline)?
• Line 80-81 – The sentence is unclear. Reference is missing?
• Line 82-83 – Not clear results of which study are presented in this paper (field or laboratory?). Adding some calculations for up-scaling the system would add extra value to the manuscript.
• Line 84 – Please consider splitting section 2 into three subsections – one for the experiment, one for the lake and sampling, one for the methods of measurements.
• Line 86 – Please add a reference to Figure 1 somewhere here.
• Line 88 – What was the aim of using glass beads in a separate column? It is given in line 171-174. Still, I was thinking about that already in line 88.
• Line 91 – How many times did the total volume of water was treated? Please add this information. What was the height of the water column above the sediment?
• Line 99 – Please add more detailed information on the Polonite material – how its is produced, from what source material and so on.
• Line 110-111 - Sentence is not fully clear (remove the “returned water” or replace it with “(effluent is returned to the container)”?).
• Line 149 – According to this equation effluent is the water treated on the columns (coming into containers) and influent is what comes to the column (from the container). This is not in line with results presented in tables where influent is obviously the water after it passed trough the column and before it was mixed with the water in the container. The same is said in line 255. See also the previous comment to lines 110-111. Please check it throughout the text.
• Line 127 – Please correct brackets.
• Lines 127-131 – Please add reference to the lake data and characteristics.
• Line 141 – Table2 (and 3) – please add an information what was the depth of water (or height from the sediment) of the water portion corresponding to “surface/middle/bottom”.
• Line 145 – Please add more detail on the method of P-PO4 determination.
• Line 169 – Still, I recommend adding some short information on the mechanisms involved in P stripping by Polonite somewhere in the Introduction.
• Line 179 – I can not see the negative values in Fig. 1
• Line 180 – Please add the reference to Fig. 1 in the first line You start describing the relevant results.
• Line 192-193 – Please check the language.
• Line 197 – Add “PO4-P” after “average”
• Line 203 – I would in any case consider redirecting the hypolimnetic water after treatment into the surface water as hypolimnetic water is rich not only in PO4, but also in NH4+, and deoxygenated. Adding this water to the lake surface can have harmful effects, i.e. stimulate primary production in the productive zone.
• Line 205 – Why? Please specify.
• Line 218-219 – Was the CaCO3 present on the sediment surface prior to experiment or it formed during the experiment? If the first case is right, CaCO3 would have to dissolve to make the pH to increase. Still, adding N2 has surely removed CO2 and thus caused the pH to increase so I`m not sure if the CaCO3 would dissolve. If the latter is right (CaCO3 formed onto the sediment surface _during the experiment_) it is worth to mention, that a coupling could take place – pH increased due to N2 addition and, together with increased pH and Ca due to contact with Polonite, it caused CaCO3 to precipitate. In that case however, CaCO3 precipitation could not be a source of increased pH.
• Line 218 – Please add a reference to Table 2.
• Line 220 – Correct “precipitates” to “precipitation”.
• Line 222 – Why? Please specify why You expect CaSO4 formation in hypolimnion. Would calcium-phosphates precipitation be also possible in hypolimnion due to Ca release from the Polonite?
• Line 222 – I suggest something like “precipitates of (CaCO3 and CaSO4?) forming in situ ….” Instead of “in situ precipitates”.
• Line 219-225 – Comment 1. In line 219-222 precipitation of CaCO3 and/or CaSO4 is mentioned as possible to take place _in the returned water_ in situ. In line 224 precipitation of calcium-phosphates is mentioned as possible to occur _“in the filter”_. What is the importance of the latter one for the topic of the paper? If I correctly understand the Authors` conclusions or predictions, I believe these are two different issues which need be clearly named and discussed separately: a) precipitation of any new phases in the lake water due to changed physio-chemical and chemical conditions in this water (higher Ca, higher pH, possibly higher temperature which are caused by the treatment on a column), which can be regarded a non-target effect, in some cases leading to further P control in the lake after the water was returned to the lake, and b) precipitation of any new phases in the column which is in my opinion rather a question of P removal mechanism and possibly also a kind of disturbing effect leading to clogging (?).
• Line 223-225 – The effect of humic substances can be somehow predicted based on previous studies (eg. Alvarez et al., 2004; https://doi.org/10.12911/22998993/67103). Please note that in boreal lakes the high level of humic substances can also affect the CaCO3 formation mentioned (eg. Reddy & Hoch, 2000, https://doi.org/10.1007/0-306-46924-3_8).
• Line 225 – Is the citation No 28 correct? Should it not be No 27?
• Line 233 – 1) Is the citation in line 223 (No 27) correct? Should it not be 26? 2) I see here some analogy to previous studies on Ca(OH2) additions to induce CaCO3 precipitation (eg. Koschel et al., 2011, doi: 10.1016/j.scitotenv.2011.01.006). It would be worth to mention that CaCO3 and gypsum itself have been proven to effectively control P release if used for sediment capping (eg. Salonen et al. 2001, https://jukuri.luke.fi/bitstream/handle/10024/450632/Salonen.pdf?sequence=1&isAllowed=y; Bartoszek, 2017, https://doi.org/10.12911/22998993/67103; Berg et al., 2004, https://doi.org/10.1016/j.apgeochem.2004.05.004).
• Line 234 – It has to be stressed the sediments will release the more PO4, the lower is the concentration in the overlying water, as the gradient between the pore water and overlying water will increase. So, it should be expected that reducing the PO4 level in the container (and in the real life in the lake), due to column activity, will enhance the P release. The same effect will have the increased temperature and the alkaline pH. Therefore, the treated hypolimnetic water should not be redirected to the hypolimnion unless the pH is not corrected. Please discuss on that.
• Line 235 –In Fig. 2 the legend is simple and directly shows the two types of filter material (Polonite and glass beads). In Fig. 3 the reader has to check what is container A and B – Why not simply add name of the filter materials in the figure? The same comment is valid for Table 2 and 3 and the text of the paper (eg. line 209). Please revise this throughout the text.
• Line 243 – The term “sampling occasion” (in the table) is not clear. Please provide the number of the day of the experiment. The same comment refers to Table 3.
• Line 246 – Please provide results for the temperature in a table or a graph.
• Line 257 – “factor 8” is not clear.
• Line 264 – I suggest “which is supported” instead of “indicated”.
• Line 273 – What can be the effects of “oxygenation, increased temperature and pH” in that case? Please shortly explain.
• Line 276 – Could other materials be used which do not result in such a substantial pH rise?
• Line 286 – I suggest to split the section 3.4 into 2 sections – one for the other results and one for the implications.
• Line 281-283 – Please check the table caption for typos.
• Line 294 – Please check the citation (No 31). Should it not be 30?
• Line 298-300 - During lake circulation periods there is no hypolimnion. It is obvious that treatment of the bottom waters is needed in summer and winter as P accumulates there in these periods.
• Line 311 – Please check the sentence. Eutrophication can not be a resource.
• Line 316 – Taking all the factors together – pumping cost, pH correction, efforts to maintain a desired temperature and PO4 level – how do the Authors asses the feasibility of the method implementation in practice? Is it higher as compared to the simpler approach where the withdrawn hypolimnetic water is treated to remove P and discharged downstream the lake as was proposed in some earlier studies? Some realistic comment on that is needed.
• Line 320-321 – Please support this sentence with citations. Would these materials cause similar problems with pH?

Author Response

Respons to reviewer #3, see red colored text

Open Review

(x) I would not like to sign my review report
( ) I would like to sign my review report

English language and style

( ) Extensive editing of English language and style required
( ) Moderate English changes required
(x) English language and style are fine/minor spell check required
( ) I don't feel qualified to judge about the English language and style

Comments and Suggestions for Authors

The article deals with an important topic of eutrophic lakes restoration. The subject of the article is within the scope of the journal. I believe some improvements need to be done in the manuscript. I listed them below as general and more specific comments.

General comments

• The proposed method needs to be described more precisely just at the very beginning when it is first mentioned. I believe the method was not put in the whole context of the newly forming approach to hypolimnetic withdrawal as a lake restoration technology. The advantages of the method proposed were not clearly described. We improved Introduction and followed your advice.
• I believe the results should be complemented and presented in a more comprehensive and consistent manner. 1. I believe the results for both, P removal efficiency and PO4-P concentrations (in the containers and in the water treated on the column (before it reached the container)) should be presented, to provide a comprehensive picture of the system operation. The more that the Authors refer to some threshold values (line 195-197) and compare them with water treated on the column (but these data are not shown). 2) I suggest to describe and discuss all the results for P (removal efficiency, resulting concentrations and so on) in one section. DONE In the current manuscript the PO4-P data are splited into 2 sections which makes the reader confused (at least it confused me). 3) The pH and the other data (DO, temperature) should be rather shown and described in a separate section(s) DONE. The data for pH for the container (averages) are presented twice (Fig. 3 and Table 2) Table 2 removed. This is superfluous. 4) The terms effluent and influent were mistaken (at least once, see specific comments). Ok, fixed
• I suggest to reorganize the text in section R&D somehow to systematize discussion elements related to the result implications. In lines 297 – 316 the conclusions are not results based and could be drawn without conducting the research presented in the manuscript. So they are not really implications although they were placed in section 3.4. entitled “….and implications”. Implications of the results were discussed in previous sections (eg. lines 198-205, 272-280). So I suggest to either group the implications (resulting from the results presented in this study) into one section called “Implications…” or make the lines 297-316 a section devoted to problems which need to be solved. We have reorganized a lot and hopefully it is to your satisfaction

Specific comments

• Line 28 – `or` instead of `and`? Fixed
• Line 29-31 - This sentence is intended to introduce the subject of phosphorus control methods in lakes. However, it refers to the rarely used method of introducing coagulants directly into the sediments. In practice, however, it is typically the dosing of coagulants or other phosphorus-binding materials into the water column (eg. Dondajewska et al., 2002, https://doi.org/10.1007/978-3-030-12139-6_17; Dithmer et al., 2016, https://doi.org/10.1016/j.watres.2016.02.011; Zamparas & Zacharias 2014, https://doi.org/10.1016/j.scitotenv.2014.07.076). I would suggest to mention first of all the widely used methods, as this would be more representative for the state of the art. We have improved and the three references now occur in the ms.
• Lines 33 – The same approach was described by Nurnberg, 2020, https://doi.org/10.1007/s10750-019-04094-z and Silvonen et al., 2021, https://doi.org/10.1016/j.scitotenv.2020.143202. Silvonen et al., 2021 called it “closed circuit approach” and it would be worth to mention this term when introducing the methodological approach. Similar method, without redirecting the hypolimnetic water back into the lake, as an improvement of the traditional HW method, was also discussed by Łożyńska et al., 2021, https://doi.org/10.3390/min11020098, together with focus on P recycling. These works should be mentioned to show the full context of the newly developing approach to HW method and P stripping from hypolimnion. Yes, you are right. DONE
• Line 44-48 – Please specify the solution being proposed. What is the difference between the newly proposed solution and the solutions previously proposed for hypolimnetic water treatment (lines 31-33)? What are the advantages? Please specify in more detail what is supposed to happen with the treated water. This is not clearly claimed here and I was a bit confused. Does the method assume discharge of the treated water into the downstream water body or back to the lake? Back to the hypolimnion? Should the method, when upscaled for the field, be based on hypolimnion water pumping? This should be cleared here in this paragraph. In this text piece the importance of P recycling is stressed. However, the paper does not deal with this topic at all. How do the Authors assess P recycling potential taking into account the Polonite material used and amount of P possible to be fixed under hypolimnion relevant conditions? We have improved and tried to make it more clear. Assessing the P recycling potential: We just touch this issue because it is important in a future large-scale test and implementation in society. There are lots of research already done on this issue so why not mention it?
• Line 60-62 – Why is the value “below 1000 micrograms” important if the hypolimnetic water has very often higher concentration of P-PO4? We removed this value and improved
• Line 67 – There was an earlier study on P stripping from hypolimnetic water on different materials which could be mentioned in this literature review, too (Łożyńska et al., 2020, https://doi.org/10.1016/j.ecohyd.2019.12.003). Yes, citation to their work is there now
• Line 71-73 – The information on the diffusion adjusted hypolimnetic withdrawal (HW) and its potential does not refer to the problem of P stripping from the hypolimnetic water (and this problem together with latest studies is described in this paragraph). Therefore I suggest to add the information on diffusion adjusted HW somewhere in line 35, where the method of HW is introduced. Correct, we removed from these lines
• Line 76 – As a reader I missed in this point some very short information on the Polonite material. Even though the material is well studied and generally known, I recommend adding here a short info about this material, how it is produced, just to give some general idea on that in this point of the paper. We added a little more and gave references where to find information.
• Line 78 – Couldn`t be predicted without this study that pH will increase if an alkaline material will be used for treatment of hypolimnetic water which is by definition more or less neutral (from slightly acidic to slightly alkaline)? Of course, but the development in the water body and other parameters.
• Line 80-81 – The sentence is unclear. Reference is missing? OK, fixed
• Line 82-83 – Not clear results of which study are presented in this paper (field or laboratory?). Adding some calculations for up-scaling the system would add extra value to the manuscript. We made it clear and calculations from this small experiment should not been done (as engineers we are)
• Line 84 – Please consider splitting section 2 into three subsections – one for the experiment, one for the lake and sampling, one for the methods of measurements. Done
• Line 86 – Please add a reference to Figure 1 somewhere here. Fixed
• Line 88 – What was the aim of using glass beads in a separate column? It is given in line 171-174. Still, I was thinking about that already in line 88. They served as control for what can happen with only pumping through an inert medium.
• Line 91 – How many times did the total volume of water was treated? Please add this information. What was the height of the water column above the sediment? Added and all information about heights are in fig. 1.
• Line 99 – Please add more detailed information on the Polonite material – how its is produced, from what source material and so on. Detailed information is found in references but we added a little. Exactly how it is produced is a question for the manufacturer. Usually they don´t disclose everything, doesn´t matter which producer we talk about
• Line 110-111 - Sentence is not fully clear (remove the “returned water” or replace it with “(effluent is returned to the container)”?). Improved
• Line 149 – According to this equation effluent is the water treated on the columns (coming into containers) and influent is what comes to the column (from the container). This is not in line with results presented in tables where influent is obviously the water after it passed trough the column and before it was mixed with the water in the container. The same is said in line 255. See also the previous comment to lines 110-111. Please check it throughout the text. Ok, fixed.
• Line 127 – Please correct brackets. ??
• Lines 127-131 – Please add reference to the lake data and characteristics. Done
• Line 141 – Table2 (and 3) – please add an information what was the depth of water (or height from the sediment) of the water portion corresponding to “surface/middle/bottom”. In figure 1
• Line 145 – Please add more detail on the method of P-PO4 determination. Done
• Line 169 – Still, I recommend adding some short information on the mechanisms involved in P stripping by Polonite somewhere in the Introduction. Done, but information is found in reference
• Line 179 – I can not see the negative values in Fig. 1 Why fig 1? Negative values in tab. 2
• Line 180 – Please add the reference to Fig. 1 in the first line You start describing the relevant results. Done
• Line 192-193 – Please check the language. ??
• Line 197 – Add “PO4-P” after “average” Ok.
• Line 203 – I would in any case consider redirecting the hypolimnetic water after treatment into the surface water as hypolimnetic water is rich not only in PO4, but also in NH4+, and deoxygenated. Adding this water to the lake surface can have harmful effects, i.e. stimulate primary production in the productive zone. That´s why we want to redirect to the hypolimnion in a full-scale system
• Line 205 – Why? Please specify. Because we will keep the conc. at a non-eutrophication level
• Line 218-219 – Was the CaCO3 present on the sediment surface prior to experiment or it formed during the experiment? No calcite prior exp. If the first case is right, CaCO3 would have to dissolve to make the pH to increase. Still, adding N2 has surely removed CO2 and thus caused the pH to increase so I`m not sure if the CaCO3 would dissolve. If the latter is right (CaCO3 formed onto the sediment surface _during the experiment_) it is worth to mention, that a coupling could take place – pH increased due to N2 addition and, together with increased pH and Ca due to contact with Polonite, it caused CaCO3 to precipitate. In that case however, CaCO3 precipitation could not be a source of increased pH. For us it will turn into speculation the effect of the few times we bubbled N2.
• Line 218 – Please add a reference to Table 2. Done.
• Line 220 – Correct “precipitates” to “precipitation”. Done
• Line 222 – Why? Please specify why You expect CaSO4 formation in hypolimnion. Would calcium-phosphates precipitation be also possible in hypolimnion due to Ca release from the Polonite? Revised. Yes, Ca-P formation is possible but the dilution in a real lake will probably lead to only local precipitation, i.e. at the discharge point
• Line 222 – I suggest something like “precipitates of (CaCO3 and CaSO4?) forming in situ ….” Instead of “in situ precipitates”. OK.
• Line 219-225 – Comment 1. In line 219-222 precipitation of CaCO3 and/or CaSO4 is mentioned as possible to take place _in the returned water_ in situ. In line 224 precipitation of calcium-phosphates is mentioned as possible to occur _“in the filter”_. What is the importance of the latter one for the topic of the paper? If I correctly understand the Authors` conclusions or predictions, I believe these are two different issues which need be clearly named and discussed separately: a) precipitation of any new phases in the lake water due to changed physio-chemical and chemical conditions in this water (higher Ca, higher pH, possibly higher temperature which are caused by the treatment on a column), which can be regarded a non-target effect, in some cases leading to further P control in the lake after the water was returned to the lake, and b) precipitation of any new phases in the column which is in my opinion rather a question of P removal mechanism and possibly also a kind of disturbing effect leading to clogging (?). Ok, thanks for these suggestions. We improved discussion on your comments.
• Line 223-225 – The effect of humic substances can be somehow predicted based on previous studies (eg. Alvarez et al., 2004; https://doi.org/10.12911/22998993/67103). Please note that in boreal lakes the high level of humic substances can also affect the CaCO3 formation mentioned (eg. Reddy & Hoch, 2000, https://doi.org/10.1007/0-306-46924-3_8). Yes, we know these refs. But did not include because we wanted to restrict discussion. However, we included them now.
• Line 225 – Is the citation No 28 correct? Should it not be No 27? Yes, we miss one ref. and got disorder in numbering. FIXED
• Line 233 – 1) Is the citation in line 223 (No 27) correct? Should it not be 26? 2) I see here some analogy to previous studies on Ca(OH2) additions to induce CaCO3 precipitation (eg. Koschel et al., 2011, doi: 10.1016/j.scitotenv.2011.01.006). It would be worth to mention that CaCO3 and gypsum itself have been proven to effectively control P release if used for sediment capping (eg. Salonen et al. 2001, https://jukuri.luke.fi/bitstream/handle/10024/450632/Salonen.pdf?sequence=1&isAllowed=y; Bartoszek, 2017, https://doi.org/10.12911/22998993/67103; Berg et al., 2004, https://doi.org/10.1016/j.apgeochem.2004.05.004). It is about capping, not the topic here
• Line 234 – It has to be stressed the sediments will release the more PO4, the lower is the concentration in the overlying water, as the gradient between the pore water and overlying water will increase. So, it should be expected that reducing the PO4 level in the container (and in the real life in the lake), due to column activity, will enhance the P release. The same effect will have the increased temperature and the alkaline pH. Therefore, the treated hypolimnetic water should not be redirected to the hypolimnion unless the pH is not corrected. Please discuss on that. It is now improved and more clear discussed
• Line 235 –In Fig. 2 the legend is simple and directly shows the two types of filter material (Polonite and glass beads). In Fig. 3 the reader has to check what is container A and B – Why not simply add name of the filter materials in the figure? The same comment is valid for Table 2 and 3 and the text of the paper (eg. line 209). Please revise this throughout the text. We did it in other way by using bed volumes throughout the paper
• Line 243 – The term “sampling occasion” (in the table) is not clear. Please provide the number of the day of the experiment. The same comment refers to Table 3. Yes, we removed sampling occasion . Day of sampling not appropriate and interesting for a lab experiment
• Line 246 – Please provide results for the temperature in a table or a graph. We give data in text instead, temp graph only boring.
• Line 257 – “factor 8” is not clear. Removed and written in a different way
• Line 264 – I suggest “which is supported” instead of “indicated”. OK, fixed
• Line 273 – What can be the effects of “oxygenation, increased temperature and pH” in that case? Please shortly explain. Revised
• Line 276 – Could other materials be used which do not result in such a substantial pH rise? Yes, revised
• Line 286 – I suggest to split the section 3.4 into 2 sections – one for the other results and one for the implications. Done
• Line 281-283 – Please check the table caption for typos. Fixed
• Line 294 – Please check the citation (No 31). Should it not be 30? Ok, fixed
• Line 298-300 - During lake circulation periods there is no hypolimnion. It is obvious that treatment of the bottom waters is needed in summer and winter as P accumulates there in these periods. The problem is that it is not easy to know when  it starts and when it ends. We  develop a technical solution which will help Municipalities to monitor turn-over periods or better, the pumping will start automatically when O2 is below 2 mg/L
• Line 311 – Please check the sentence. Eutrophication can not be a resource. Yes, it can! Read the fully sentence, then you will know what is meant

1. Line 316 – Taking all the factors together – pumping cost, pH correction, efforts to maintain a desired temperature and PO4 level – how do the Authors asses the feasibility of the method implementation in practice? Is it higher as compared to the simpler approach where the withdrawn hypolimnetic water is treated to remove P and discharged downstream the lake as was proposed in some earlier studies? Downstream washing of P-polluted bottom water will NOT be allowed in the Nordic countries!  Some realistic comment on that is needed. We will not speculate here – papers will be published later when we have data from the current project at Lake Hönsan.

• Line 320-321 – Please support this sentence with citations. Would these materials cause similar problems with pH?  See comment above. We will see after the semi-scale test at lake Hönsan.

Reviewer 4 Report

Review of Manuscript ID: water-1605877

Eutrophication is one of main issues in the environment protection nowadays. There are a lot of proposed methods for lakes’ restoration. But every lake has unique environmental conditions, and proposed restoration methods are not suitable in every case. One of the oldest methods used for lake water quality improvement is hypolimnetic withdrawal. This method can be applied for deep, stratified lakes, but the removal of hypolimnetic water, which is rich in nutrients (N and P) into the outflow, has some undesirable aspects. Main problem is the export of N and P load from the lake ecosystem beyond drainage basin. Then the seeking solutions for decrease P concentrations in the removed hypolimnetic water is very valuable.

The reviewed manuscript subject concerns method for P elimination for the hypolimnetic water, which is directed from the lake restored by hypolimnetic withdrawal into outflow. Authors tested usability of adsorbing material Polonite® for P removal from the hypolimnetic water. The research was made in laboratory conditions. The effectiveness of reactive Polonite® bed for P removal was compared to bed filled by glass beads. Th experiment showed the effectiveness of Polonite® material in P removal from water. In my opinion the experiment was well designed, and prove the usefulness of tested reactive material.

Using solid materials based on calcium compounds creates increase in pH of effluent. Authors also observed this effect and proposed solutions for solving this problem. High pH can have influence on calcium -carbonates equilibrium in lake water (in the case of recirculation of effluent into the lake), giving some effects, which were observed by authors (eg. precipitation of Ca compounds on sediment). Also pH elevation in the water-sediment interface, if occurs, can stimulate the internal P loading from bottom sediment. The P pools bound with organic matter and Fe and Al oxides and hydroxides can be released into the near bottom water. That fact could be discussed by Authors.

Another thing is the nature of P sorption by tested material. Important information is, what kind of adsorption is dominant. Physical adsorption is quick, but can be relatively reversible, which is important for further desorption processes. The Ca-P fraction in the lake sediment is hardly soluble, because P is mainly chemisorbed. Chemical adsorption is more durable, and hard to reverse. It could be some problems with later P desorption. In my opinion Authors should take this issue for discussion.

I think that manuscript is interesting and important for development of lake restoration issue, then I recommend the paper for publication after minor revision.

Minor comment:

Line 170 – the word ”Phosfate” should be replaced by “Phosphate”

Author Response

1. Concerning high pH in effluent from reactive media. We have improved our discussion on this topic and also this issue about increased release of P from sediment in the case pH increase in the water body. However  we did not see any effect in our small experiment - maybe duration of experiment was too short
2. Concerning Ca-P fraction. We have added some discussion on that topic as well and we know from previous research (in the case of Polonite) that P is precipitated in several insoluble fractions. It can however be released if pH in Polonite decrease to 7.5-8 but we always replace the filter media before this pH is reached.
3. Minor comment is fixed.

Round 2

Reviewer 3 Report

The review is attached.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf