Review Reports - Treatment of Winery Wastewater Using Bench-Scale Columns Simulating Vertical Flow Constructed Wetlands with Adsorption Media

Round 1

Reviewer 1 Report

The manuscript "Treatment of Winery Wastewater Using a Vertical Flow Constructed Wetland with Adsorption Media is a good scientific proposal for the removal of several contaminants from winery wastewater, using an innovative technology with high removal capacity.

Following are the suggestions to improve the manuscript:

1) line 41-42: check the specified quantity in liters. The quantities of wastewater are 45.4 million liters!

2) Are there other works that deal with the treatment of the same wastewaters? It is necessary to implement the bibliography in the introduction.

3) line 52: I suggest inserting references such as “Biomass - Importance, Chemistry, Classification, and Conversion (2019). Biofuel Research Journal, 22, 962-979”, which generally deals with the collection, production of various biomasses and the resulting byproducts.

4) line 122: change the verbs "produces", used twice in the same sentence.

5) line 278: there is an introductory and comparison part with other methods used for the removal of the COD. These comparisons should be accommodated in the introduction and not in the results section.

6) line 395: I suggest adding the following reference which deals with synthetic wastewater prepared and used by monitoring the COD/N/P ratio for adsorption experiments: “Removal of endocrine disrupting chemicals from water: adsorption of bisphenol-A by biobased hydrophobic functionalized cellulose. (2018). International Journal of Environmental Research and Public Health, 15 (11), 2419.”

7) Is this system scalable on industrial level? What are the main advantages compared to the other methods currently used?

8) I would propose to add a final synthetic paragraph showing a comparison between this type of treatment for the removal of COD, Nitrogen and Phosphorus respectively, and the conventional methods used for this type of contaminants in wastewaters.

9) The experimental data do not report errors, please add them.

Author Response

Thank you for your kind words. We extensively edited the manuscript using the grammar software “Grammarly Premium” to improve English language and style.

1. line 41-42: check the specified quantity in liters. The quantities of wastewater are 45.4 million liters!

Thank you for pointing this out this mistake. We changed it to “45.4 million liters”.

2. Are there other works that deal with the treatment of the same wastewaters? It is necessary to implement the bibliography in the introduction.

To better address this, we separated the paragraph beginning on Line 73 into two paragraphs and added information about other works. It now reads: “Regulations have been set to mitigate the impacts of wastewater discharges, and new, more restrictive regulations are driving the development of technologies for winery wastewater treatment [6,43,44]. Many wineries are located rurally and do not have access to public sewers. Those that do often face high surcharges due to the acidic pH and high chemical oxygen demand (COD) in the wastewater [6,13], making onsite treatment an attractive alternative. Historically, onsite treatment of winery wastewater at small wineries has been accomplished with land application [6], but stricter regulations have increased the land needed for treatment, reducing that available for vineyards. Currently, activated sludge systems represent the majority of treatment systems at European wineries [6], but are complex to operate for small wineries and are expensive due to high energy use [7,9]. Emerging treatment systems for winery wastewater include membrane bioreactors, jet-loop activated sludge, and air micro-bubble reactors, as summarized in Mosse et al. (2011) [9]. However, these technologies may not be applicable to small-scale wineries due to their cost, complexity, and limited demonstrated applications [9].”

3. line 52: I suggest inserting references such as “Biomass - Importance, Chemistry, Classification, and Conversion (2019). Biofuel Research Journal, 22, 962-979”, which generally deals with the collection, production of various biomasses and the resulting byproducts.

Thank you for the suggestion - we examined the article and while it is quite interesting and well researched, we do not feel that it contributes to this research manuscript.

4. line 122: change the verbs "produces", used twice in the same sentence.

Reworded sentence to “This winery produces both red and white wine, totaling approximately 2,400 cases per year.” to eliminate repeated word.

5. Line 278: there is an introductory and comparison part with other methods used for the removal of the COD. These comparisons should be accommodated in the introduction and not in the results section.

We appreciate that there is a considerable amount of new literature presented in this section, however, this was done to easily compare the surface loading rate of the SVFCW investigated in this study with previous SVFCW studies.

6. line 395: I suggest adding the following reference which deals with synthetic wastewater prepared and used by monitoring the COD/N/P ratio for adsorption experiments: “Removal of endocrine disrupting chemicals from water: adsorption of bisphenol-A by biobased hydrophobic functionalized cellulose. (2018). International Journal of Environmental Research and Public Health, 15 (11), 2419.”

Thank you for sharing this article with us. We have added the reference to Line 443: “The ratio of COD:N in winery wastewater is often 100, an order of magnitude larger than that in municipal wastewater [6,74].”

7. Is this system scalable on industrial level? What are the main advantages compared to the other methods currently used?

We elaborated on the advantages of this system; Lines 491-499 now read: “Using the permitted Michigan surface loading rate of 50 pounds of BOD5/acre/day [77] for land application as an example, SVFCWs require approximately 80% less treatment area than land application. In addition to the potential economic benefit from the sale of wetland plants [78], the use of aesthetically pleasing vegetation may be beneficial to wineries where tourism is important. Additionally, SVFCWs are substantially more energy-efficient than activated sludge [7]. Further, the characteristics of the discharged wastewater from SVFCWs is comparable to or better than wastewater treated by conventional methods such as activated sludge [79]. These benefits make this technology a viable option for implementation at small-scale industrial wineries.”

8. I would propose to add a final synthetic paragraph showing a comparison between this type of treatment for the removal of COD, Nitrogen and Phosphorus respectively, and the conventional methods used for this type of contaminants in wastewaters.

Please see response to comment 7.

9. The experimental data do not report errors, please add them.

We have added a summary of our Quality Assurance and Quality Control (QAQC) plan to section 2.4 (Wastewater Sampling and Analysis) as well as the results of the QAQC conducted throughout the study.

Reviewer 2 Report

The use of constructed wetland for the treatment of wine wastewater is a very good idea. Especially in the aspect of significant seasonal variation of sewage inflow. Absorbed elements can be used for the rest of the year by plants to increase their growth. It is worth thinking in such cases about the associated farming - the use of plants for economic purposes. Unfortunately, the authors did not touch on this aspect. Another problem concerns the very definition of constructed wetland. Lack of reference to the soil-plant-atmosphere system means that the authors practically dealt with the biofilter only. So I have doubts whether the title of the article is correct.

In my opinion after the harvest period, the proposed sewage treatment plant should work on social and domestic sewage. This will ensure sustained plant growth.

Tab. 1 - converted min and max values. Can the average values be lower than the minimum values?

Should the constructed wetland use the primary settling tank / septic tank for removing the suspension? - I am worried about the phenomenon of bed clogging in the long term, especially in the aspect of using off-season household and domestic sewage

Conclusions:

The first sentence should only refer to the purification potential. As the authors themselves state later, the laboratory conditions require field tests.
The authors did not consider SVFCW's work in the longer term, with the supply of social and household sewage and low temperatures during the year.

Author Response

1. Absorbed elements can be used for the rest of the year by plants to increase their growth. It is worth thinking in such cases about the associated farming - the use of plants for economic purposes. Unfortunately, the authors did not touch on this aspect.

Thank you for bringing this up. One of the authors actually did a study with a company, Aquatic Plants of Florida, who harvested mangroves from wetlands. An economic analysis demonstrated that the sales of wetland plants made the system economically beneficial (even though it was a small-scale system). We’ve addressed this potential benefit on Line 494.

2. Another problem concerns the very definition of constructed wetland. Lack of reference to the soil-plant-atmosphere system means that the authors practically dealt with the biofilter only. So I have doubts whether the title of the article is correct.

Excellent point, the authors debated this and decided to keep the title as a full-scale application of this treatment system would meet the definition of constructed wetland. To clarify the specifics of this research, we changed Lines 135-139 to: “Additionally, the subsurface application of the wastewater used during cold weather is below the root zone of most wetland-style plants [47,57]. As such, it was determined that excluding plants and only investigating the biofilter portion of the constructed wetland would result in the most conservative experimental design.”

3. In my opinion after the harvest period, the proposed sewage treatment plant should work on social and domestic sewage. This will ensure sustained plant growth.

This research was focused on applications in the United States where it is strongly discouraged to combine domestic sewage with industrial waste as it changes the management techniques. For this application, the industrial (winery) wastewater is treated by the proposed system and domestic wastewater is treated in a conventional septic system. However, this is an excellent suggestion and we have included it as potential future research at the end of the manuscript: “Additionally, this application was focused in a region where combining domestic wastewater with industrial wastewater is strongly discouraged. However, in regions where co-treatment is allowed, domestic wastewater could provide a year-round source of substrate to microbial communities, justifying further investigation.” (Lines 518-521)

4. Tab 1 - converted min and max values. Can the average values be lower than the minimum values?

For clarification, we changed “mean” to “average” in the table where “average” represents the mathematical average and can never be lower than the minimum or higher than the maximum.

5. Should the constructed wetland use the primary settling tank / septic tank for removing the suspension? - I am worried about the phenomenon of bed clogging in the long term, especially in the aspect of using off-season household and domestic sewage

Thank you for the great observation – sedimentation and pretreatment are absolutely essential and we added this statement to Lines 95-98: “An extensive pretreatment system that included a septic tank and an effluent filter mitigated issues related to bed clogging [47]. Following pretreatment, wastewater was distributed into the system at the vegetated surface during warm months or below the soil layer during cold months [47].”

Conclusions:

6. The first sentence should only refer to the purification potential. As the authors themselves state later, the laboratory conditions require field tests.

To clarify, we changed Line 480 to “The bench-scale SVFCW shows potential to sufficiently treat winery wastewater under variations in loading rates, frequency, and temperature without the aid of nitrogen adsorption media or a pH buffer.”

7. The authors did not consider SVFCW's work in the longer term, with the supply of social and household sewage and low temperatures during the year.

Please see response to comment 3.

Reviewer 3 Report

This manuscript by Skornia et al. aims to assess the potential of a subsurface vertical flow constructed wetland for the purification of winery wastewater.

In this work, which is very clearly presented, all the experiments have been diligently conducted and the results are correctly discussed with the literature.

I recommend to accept this manuscript after minor modifications.

Specific comments are given below,

L12: WW is often "high-strength", could you explain this word?

L41-42: "45.4 liters" maybe it misses something?

L61-62: I'm not convinced by the separation in two tables. Considering classical wastewater parameters, N and P can also be considered as contaminants. I therefore suggest to merge these two tables for clarity.

L98: "do not leach harmful products", as you certainly know, tires are one of the main producer of µplastics. So i agree with you on the cheapness af the used material, but i suggest to moderate the last part of the sentence.

Please check the units in your figures, as for example in Figure 5 (mg/L N) is not correct.

Author Response

1. In this work, which is very clearly presented, all the experiments have been diligently conducted and the results are correctly discussed with the literature. I recommend to accept this manuscript after minor modifications.

Thank you for your kind words.

Specific comments are given below,

L12: WW is often "high-strength", could you explain this word?

The authors agree that the use of “high-strength” is ambiguous. For clarification, we changed Lines 12-14 to “Wastewater produced during the wine-making process often contains an order of magnitude greater chemical oxygen demand (COD) concentration than is typical of domestic wastewater. This waste stream is also highly variable in flow and composition due to the seasonality of wine-making.”

Additionally, on Line 51, we clarified “highest strength” by changing it to “highest chemical oxygen demand (COD) strength wastewater”.

L41-42: "45.4 liters" maybe it misses something?

Thank you for pointing this out this mistake. We changed it to “45.4 million liters”.

Thank you for the suggestion, we have merged Tables 1 and 2 into one table titled “Conventional wastewater pollutants in winery wastewater as reported in literature”.

We agree that very small particles that are worn from tires are a major source of micro-plastics. However, the larger scrap tire particles used in this study have not been shown to have significant toxic effects. The following revision was made regarding scrap tire chips: “In a prior study, scrap tire chips (1-1.5 cm particle size) were found to leach small amounts of bioavailable organic carbon that supported denitrification [51]. Although low concentrations of zinc, selenium, manganese, antimony, and cobalt were detected, other metals of concern were below detection limits (see Krayzelova et al. [51] for review of toxicology studies). Moreover, tire chips have been approved for use in onsite drain field applications in several states [55].”

Please check the units in your figures, as for example in Figure 5 (mg/L N) is not correct.

We changed the y-axis lab to “Cumulative Concentration (mg/L N)” and added to Lines 430-432: “Organic nitrogen was calculated by subtracting ammonia and nitrate from total nitrogen, assuming no other forms of inorganic nitrogen were present with the given operating conditions and influent ammonia concentrations [72,68].”

Reviewer 4 Report

General comments:

The results of this study do not provide significant suggestions regarding the benefits of using the applied material and methods for the treatment of winery wastewater. The removal of phosphorus by adsorption on a commercial adsorbent is not innovative and does not add to the value of this manuscript. This study lacks proof for the conclusions made.

Specific Comments:

Materials and Methods:

Lines 126-128: The authors monitored the concentrations of COD, total nitrogen, and total phosphorus as the key wastewater parameters. However, it would have been more accurate to monitor the BOD concentration of wastewater, as this parameter better represents the organic content of wastewater. At least, several BOD₅ measurements should have been conducted during the course of operation in order to develop a relationship between COD and BOD₅. Please comment. Table 3: How was the average BOD concentration estimated in the absence of the minimum and maximum concentrations? Lines 148-150: How did the top-down flow of wastewater promote aerobic conditions? What was the DO concentration in the wastewater? What was the DO concentration at various heights and locations of the column during the operation? Lines 155-156: The third column of experimental systems is reported to also have aerobic conditions. Were the first and third columns aerated? Figure 1 does not show any aeration into the columns. Please comment on the air flow rates and DO concentrations in these columns. Also, the DO concentration and, preferably, the ORP in the anaerobic columns should be reported. Line 162: the authors report that the first system served as the control and contained only gravel. How do the authors define the control system? Was the growth of microorganisms prevented in this column? Did the authors remove the possibility of contaminant adsorption in this column? Line 167-168: The third column of each system was used as a polishing column with only gravel present. How was this column different from the columns in the first system? How was polishing accomplished in the third column of each system, and not in the first system? Line 184: the reference should be reported as Krayzelova et al. (2014). Line 189: What were the characteristics of the secondary effluent wastewater that was used as the inoculum? What was the biomass concentration, BOD, TN and TP in this effluent? Do the authors refer to the effluent of a wastewater treatment plant, or the raw wastewater which was used as the inoculums? Line 202: What was the reason for the applied operation mode? Why was the wastewater distributed into the SVFCWs four times per day? Why not operating the system in continuous mode? Lines 225-226: What was the “specific influent concentration of total phosphorus” suggested by the manusfactureres? Lines 247-249: Did the authors verify the accuracy of the applied preservation method, by analyzing some samples both immediately and also after 28 days following the preservation? Ammonia removal is accomplished by adsorption/ion exchange, using a cation removal resin. Why not removing nitrate by ion exchange, using an anion-removal resin? In fact, the removal of nitrate from water by ion exchange is a common method for the removal of this contaminant.

Results and Discussion:

Figure 3 shows that the concentration of all parameters in the control system was very low. Normally, the control system operates in such a way that no removal of any contaminant takes place. The authors should clarify the role of control system in their study, and the reasons behind the removal of all contaminants to very low levels in this system. Actually, on lines 273-275, the authors state that no significant differences in COD removal were observed between the control system and the treatment systems with adsorption media. Therefore, the suggested method for the removal of contaminants offers no particular value. The reason behind this observation should be stated and discussed. The reference on line 279 should be reported as Serrano et al. Similarly, the reference on line 282 should be reported as Rozema et al. (2016). The year for the reference Campbell & Safferman (line 289) is missing. Lines 298-303: How can the authors maintain that in their study COD was removed by biological processes, and refer to their wetland as a biological treatment system? What is the proof for the biological activities in their system? Did the authors monitor biomass concentration and activity? Was any microbiological test conducted during the study? This issue is particularly important since even the control system continuously removed a major fraction of COD and reached the same removal efficiency after 13 days. The increased removal of COD in the inoculated column could have been due to other abiotic mechanisms, e.g. adsorption onto the biomass, complexation, and precipitation promoted by the presence of biomass in the column, and not by biological degradation. Lines 322-323: It is stated that treatment during the reduced temperature phase was not significantly different. This statement cast doubt on the removal of COD by biological processes, since temperature significantly affects the activity of biomass and controls the kinetics of biological processes. Section 3.1.2.: What were the mechanisms of ammonia and nitrate removal in the control system that did not contain any clinoptilolite? Lines 355-356: The authors state that the high concentrations of potassium in winery wastewater may have impacted ammonia removal with clinoptilolite. However, in the Materials and Methods section, the estimation of the amount of clinoptilolite was reported to have taken into consideration the potassium competition. Thus, this conclusion is not correct. Line 357-358: How can higher nitrogen loading rates mitigate competition from competing ions, especially when the competing ions are also present at high concentrations? This is confusing. Lines 375: Where is the proof for the presence of denitrification in this column? Figure 5: The assumption that there is no other form of inorganic nitrogen other than ammonia and nitrate may not be correct since ammonia oxidation produces nitrite as well as nitrate, and nitrite may persist depending on the DO level in the system. Line 406: if there was no difference in the pH of solution with or without oyster shells as a pH buffer, why was a buffer included in the system? Figure 6: Why is phosphorus concentration in Figure 6b decreasing continuously? The authors should comment on the possible removal of phosphorus by precipitation, especially in light of using calcium carbonate that can potentially precipitate phosphorus as its calcium salt. The Conclusions and the Abstract sections of the manuscript state that the treatment systems with nitrogen adsorption media did not enhance nitrogen removal. So, what was the mechanism of nitrogen removal in the treatment systems examined in this study? In other parts of the manuscript, the authors mention the removal of nitrogen by adsorption on tire chips. Were adsorption tests performed on this media?

Author Response

The removal of phosphorus by adsorption on a commercial adsorbent is not innovative and does not add to the value of this manuscript. This study lacks proof for the conclusions made.

We appreciate your concern, however, in wastewater treatment systems it is uncommon to use adsorption media for phosphorus removal; in treatment of winery wastewater, it is even rarer as explained in Mosse et al. (2011). Further, it is essential for this manuscript as it shows complete treatment for key wastewater characteristics. If phosphorus removal is not required at a winery, the design described in this manuscript makes it simple to remove the phosphorus treatment mechanism. We added this explanation to Line 251: “This treatment system design allows for flexibility in full-scale implementation as the tertiary treatment is easily removed if phosphorus removal is not necessary at a winery.”

Specific Comments:

Materials and Methods:

Unfortunately, resources did not allow for routine BOD₅ measurements, however, the ratio is established in literature and our ratio was within the reported range. We clarified this by adding to Lines 151-155: “A random wastewater sample was analyzed for biochemical oxygen demand (BOD5) and was found to have a COD: BOD5 ratio of 1.57. This ratio is within the range of 1.45–1.76 described by Table 1 and a review paper by Mosse et al. (2011) [9], indicating that the wastewater used in this study was representative of typical winery wastewater.”

Table 3: How was the average BOD concentration estimated in the absence of the minimum and maximum concentrations?

We took a random wastewater sample and measured the BOD₅ concentration once during the experiment (as explained in Lines 274-275) to check that our COD:BOD ratio was within the range reported in literature. To clarify that this was a single measurement, we removed the measurement from the table and added the clarification described in the response to comment 2.

Lines 148-150: How did the top-down flow of wastewater promote aerobic conditions? What was the DO concentration in the wastewater? What was the DO concentration at various heights and locations of the column during the operation? Lines 155-156: The third column of experimental systems is reported to also have aerobic conditions. Were the first and third columns aerated? Figure 1 does not show any aeration into the columns. Please comment on the air flow rates and DO concentrations in these columns. Also, the DO concentration and, preferably, the ORP in the anaerobic columns should be reported.

We agree that these columns are not exclusively aerobic and instead, contain both aerobic and anoxic zones, similar to a trickling filter (Tchobanoglous & Burton, 1991, p. 404). Additionally, the systems were not actively aerated. To emphasize this, we changed Lines 173-176 to: “Influent winery wastewater from a settling tank was pumped to an elevated reservoir and then flowed via gravity into the first column of each system through a barbed inlet fitting, 0.45 meter (1.5 feet) below the top of the column. This column was open to the atmosphere and passively aerated.”

Similarly, Lines 181-183 were changed to: “Effluent wastewater from the second column was then pumped to the top inlet of the third column. This column served as a polishing column and was also open to the atmosphere and passively aerated.”

Unfortunately, we did not measure the DO concentrations, air flow rates, or the ORP in the treatment systems.

We are confident that the second column was anoxic as it remained saturated throughout the entire project period and any DO in the effluent of column 1 could have been consumed near the inlet by aerobic microbial activity. We qualified this in the manuscript by changing Lines 178-180 to: “Filling the second column from the bottom resulted in water saturation, therefore, it was assumed that an anoxic environment developed within the second column due to aerobic microbial activity near the inlet.”

Line 162: the authors report that the first system served as the control and contained only gravel. How do the authors define the control system? Was the growth of microorganisms prevented in this column?

The control system was included to investigate the impact of the nitrogen adsorption media on system performance. Microbial growth was not prevented in this system and it was not expected that COD removal would be impacted by nitrogen adsorption media, however, we monitored this to confirm our expectations. We clarified the definition of the control system on Line 189: “The first system served as the control for determining the impact of nitrogen adsorption media and contained only gravel, which was approximately 0.64-centimeter (0.25-inch) in diameter.”

Did the authors remove the possibility of contaminant adsorption in this column?

This was an excellent point – at the concentrations used in this study, gravel’s ammonia adsorption capacity is very low at 10 mg/kg (Kadlec & Wallace, 2008, p. 275). As the decomposition of organic nitrogen readily changes the form of nitrogen to ammonia (Tchobanoglous & Burton, 1991, p. 86), it can be assumed that the organic nitrogen removed in the first column was converted to ammonia. Using an ammonia concentration of 25 mg/L and an adsorption capacity of 10 mg/kg, we would have exhausted the ammonia holding capacity within approximately 22 days. It is reasonable to expect that this time would have been further reduced by the presence of competing ions in the wastewater, such as potassium. However, we did not observe any breakthrough of ammonia, indicating that biological removal of nitrogen was taking place.

Line 167-168: The third column of each system was used as a polishing column with only gravel present. How was this column different from the columns in the first system? How was polishing accomplished in the third column of each system, and not in the first system?

We clarified the purpose of the Polishing cell on Lines 167-169: “Generally, the final cell is not required for acceptable treatment in a SVFCW, but is included to provide operational flexibility on a full-scale system and remove residual carbon and nitrogen.”

Line 184: the reference should be reported as Krayzelova et al. (2014).

Thank you, this has been corrected.

Line 189: What were the characteristics of the secondary effluent wastewater that was used as the inoculum? What was the biomass concentration, BOD, TN and TP in this effluent? Do the authors refer to the effluent of a wastewater treatment plant, or the raw wastewater which was used as the inoculums?

This information is provided on Lines 218-222. Additionally, we have added the WWTP’s limited for carbonaceous BOD₅ and TP. Please note that this treatment plant has not had any violations so we are confident that the inoculum characteristics are below the stated concentrations.

Line 202: What was the reason for the applied operation mode? Why was the wastewater distributed into the SVFCWs four times per day? Why not operating the system in continuous mode?

We have reworded Lines 233-235 to read: “This schedule was chosen to simulate the frequency of wastewater production at a small winery where wastewater is produced in batches rather than continuously.”

Lines 225-226: What was the “specific influent concentration of total phosphorus” suggested by the manufacturers?

We reworded the sentence on Line 258 and added the influent concentration range. It now reads: “The quantity of PO4Sponge was determined following the manufacturer recommendation for an empty bed contact time of 30 minutes for influent concentrations of 10–20 mg/L total phosphorus [61].”

Lines 247-249: Did the authors verify the accuracy of the applied preservation method, by analyzing some samples both immediately and also after 28 days following the preservation?

Line 281 now reads: “Testing, preservation, and neutralizing methods followed HACH standard procedures, summarized in Table 5. All of these methods are compliant with United States Environmental Protection Agency (USEPA) testing standards, with exception of total nitrogen [62].”

Ammonia removal is accomplished by adsorption/ion exchange, using a cation removal resin. Why not removing nitrate by ion exchange, using an anion-removal resin? In fact, the removal of nitrate from water by ion exchange is a common method for the removal of this contaminant.

We agree. The reason that the scrap tire chips were added is because they were shown to serve as a low-cost anion exchange resin with a moderate affinity for nitrate (Krayzelova et al., 2014). This is discussed on Lines 107-112.

Results and Discussion:

Please refer to the response to comment 5 that clarified the purpose of the control system.

Actually, on lines 273-275, the authors state that no significant differences in COD removal were observed between the control system and the treatment systems with adsorption media. Therefore, the suggested method for the removal of contaminants offers no particular value. The reason behind this observation should be stated and discussed.

We agree and made this clearer in the conclusions on Lines 482-484.

The reference on line 279 should be reported as Serrano et al. Similarly, the reference on line 282 should be reported as Rozema et al. (2016). The year for the reference Campbell & Safferman (line 289) is missing.

Thank you, these references have been corrected.

Lines 298-303: How can the authors maintain that in their study COD was removed by biological processes, and refer to their wetland as a biological treatment system? What is the proof for the biological activities in their system? Did the authors monitor biomass concentration and activity? Was any microbiological test conducted during the study? This issue is particularly important since even the control system continuously removed a major fraction of COD and reached the same removal efficiency after 13 days. The increased removal of COD in the inoculated column could have been due to other abiotic mechanisms, e.g. adsorption onto the biomass, complexation, and precipitation promoted by the presence of biomass in the column, and not by biological degradation.

This system is similar to a trickling filter where “The organic material present in the wastewater is degraded by a population of microorganisms attached to the filter media” (Tchobanoglous & Burton, 1991, p. 404). Additionally, the start-up periods examined in this study appear to show microbial growth. Although this was not confirmed with microbiological tests, the increase in COD removal over time suggests that a population was growing in monitored start-up periods. If the removal mechanism for COD had been adsorption to the gravel, this same acclimation period would not have been observed.

However, it is important to note that constituents of the wastewater must adsorb through the biofilm layer to be utilized by the microbial population (Tchobanoglous & Burton, 1991, p. 404). To avoid confusion, “degradation” has been changed to “treatment” on Line 448.

In regards to the inoculated and non-inoculated start-up periods, Line 348 states: “The similar start-up periods indicate that there is an active microbial community present in the winery wastewater that is capable of colonizing the media.” This statement is corroborated by a study by Malandra et al. (2002) entitled Microbiology of a biological contactor for winery wastewater treatment. We have added this reference to the statement on Line 349.

Lines 322-323: It is stated that treatment during the reduced temperature phase was not significantly different. This statement cast doubt on the removal of COD by biological processes, since temperature significantly affects the activity of biomass and controls the kinetics of biological processes.

The surface loading rate applied in this system was based on prior studies conducted at low temperatures so it was not surprising that high treatment performance was observed at reduced temperatures. To clarify that the surface-loading rate was sized for cold-weather operation, we have added to Line 237, it now reads: “Wastewater was distributed at an approximate surface loading rate of 5.18 kg COD/m²/d (1.06e-2 lb COD/ft²/d); this loading rate was previously determined to be optimum for a cold-weather SVFCW treating milking facility wastewater [45] but had not been tested for winery wastewater.”

Section 3.1.2.: What were the mechanisms of ammonia and nitrate removal in the control system that did not contain any clinoptilolite?

Removal was a result of biological activity, specifically microbial uptake, ammonification and nitrification/denitrification. Please see responses to comments 6 and 17.

Lines 355-356: The authors state that the high concentrations of potassium in winery wastewater may have impacted ammonia removal with clinoptilolite. However, in the Materials and Methods section, the estimation of the amount of clinoptilolite was reported to have taken into consideration the potassium competition. Thus, this conclusion is not correct.

The initial design of the treatment systems used the adsorption capacity previously determined by Rodriguez-Gonzalez (2017) [60] that accounted for the presence of competing ions. However, the results of this study are in contrast with results of previous studies using the nitrogen adsorption media; as such, further investigation is needed to better understand the impacts of competing ions in winery wastewater. We have clarified and added information to Lines 394-401: “The high concentrations of potassium in winery wastewater may also have impacted ammonia removal, as clinoptilolite is known to have a higher affinity for potassium than ammonium when the two ions are present in equimolar concentrations. However, when ammonium concentrations are higher than potassium this effect is reduced, as was observed in a 2017 study that demonstrated 80–90% ammonia removal efficiency by clinoptilolite when treating swine wastewater with a NH4+ concentration of 54 M and a K+ concentration of 3.0 M [65]. Consequently, more research is needed to better understand the impact of potassium and other competing ions in winery wastewater.”

Line 357-358: How can higher nitrogen loading rates mitigate competition from competing ions, especially when the competing ions are also present at high concentrations? This is confusing.

Please see response to comment 20.

Lines 375: Where is the proof for the presence of denitrification in this column?

The saturated second column exhibited simultaneous removal of COD and nitrate, which serve as electron donor and electron acceptor, respectively, for denitrification. We therefore assume that denitrification was occurring. Moreover, although columns 2 and 3 of Treatment 2a and 2b contained nitrate adsorption media (tire chips), we are unaware of processes that would have exhibited the same results in the control column.

Figure 5: The assumption that there is no other form of inorganic nitrogen other than ammonia and nitrate may not be correct since ammonia oxidation produces nitrite as well as nitrate, and nitrite may persist depending on the DO level in the system.

In addition to the DO level within a system, ammonia concentration, pH, and temperature also play a large role in nitrite accumulation (Aponte-Morales, Payne, Cunningham, & Ergas, 2018; Bae, Baek, Chung, & Lee, 2001) [68,72]. It is unlikely that nitrite accumulated in the bench-scale SVFCWs in this system as ammonia concentrations greater than 50 mg/L as N (Aponte-Morales et al., 2018), a basic pH, and high temperatures (30 °C) are needed (Bae et al., 2001). The maximum ammonia concentration observed in this study was below this threshold and the systems investigated were only moderately basic through the third cell and operated at temperatures of 20 °C and 10 °C.

We have clarified the reasons for our assumption in Lines 430-432.

Line 406: if there was no difference in the pH of solution with or without oyster shells as a pH buffer, why was a buffer included in the system?

It was originally hypothesized that it would be needed but we found the wastewater was sufficiently treated without the buffer. We clarified this by adding to the conclusions on Lines 482-484: “Although this contradicts the original hypothesis that a SVFCW would not provide sufficient treatment without amendments, this represents a more cost-effective solution for small-scale industrial wineries.”

Additionally, we have added to our pH discussion on Lines 458-460: “While oyster shells were not needed in this study, oyster shells might be useful as a source of alkalinity in the nitrification stage of biofilters treating low alkalinity wastewaters because nitrification consumes alkalinity.”

Figure 6: Why is phosphorus concentration in Figure 6b decreasing continuously? The authors should comment on the possible removal of phosphorus by precipitation, especially in light of using calcium carbonate that can potentially precipitate phosphorus as its calcium salt.

We added a comment on the initial increase in PO4Sponge performance on Line 467-469: “It is not known why there was an initial increase in phosphorus removal by the PO4Sponge, however, the higher levels at the beginning of the study were not concerning as they were still quite low.”

We have also added the following statement to Line 473 to address your comment on recovering the phosphorus: “Following treatment, the phosphorus can be removed from the PO4Sponge by precipitation and used for beneficial purposes.”

The Conclusions and the Abstract sections of the manuscript state that the treatment systems with nitrogen adsorption media did not enhance nitrogen removal. So, what was the mechanism of nitrogen removal in the treatment systems examined in this study? In other parts of the manuscript, the authors mention the removal of nitrogen by adsorption on tire chips. Were adsorption tests performed on this media?

Nitrogen removal was assumed to be a result of biological activity. Please see responses to comments 6 and 17 as well as the discussion on the C:N:P ratio on Lines 438-448.

The adsorption capacity of the tire chips was based on isotherms performed by Krayzelova et al. (2014) [51], as stated in Lines 210-212.

Round 2

Reviewer 1 Report

Dear Authors,

the manuscript is improved in the last revision and all my requests are satisfied, for this reason I think that the manuscript is now available for the publication.

Author Response

Thank you for your original comments that enabled us to greatly improve the manuscript.

Reviewer 4 Report

1. In response to my comment related to lines 148-150 of the original manuscript (How did the top-down flow of wastewater promote aerobic conditions? What was the DO concentration in the wastewater? What was the DO concentration at various heights and locations of the column during the operation?), the authors state on lines 174-175 (revised manuscript) that the column was open to the atmosphere and passively aerated. This response indicates that the DO concentration inside the column was not measured, and the establishment of aerobic condition in the column is simply an assumption that cannot be verified. This is important because nitrification requires a relatively high DO concentration. Therefore, the simple presence of oxygen as a result of open-top setup of the experimental column will not provide an adequate DO concentration inside the column.

Further down, on lines 178-179 (revised manuscript), the authors state that “it was assumed that an anoxic environment developed within the second column due to aerobic microbial activity near the inlet.” Although the establishment of anoxic condition inside a column that is not aerated is plausible, the lack of measurement of environmental parameters, such as dissolved oxygen (DO) concentration and oxidation-reduction potential (ORP), to verify the environmental conditions in a reactor (column), which is a common engineering practice, casts doubt on the developed assumptions and the conclusions made.

2. The authors provide the same reasoning on lines 181-182 (revised manuscript) about the establishment of aerobic condition in the third column of experimental systems. They state that the polishing column was also open to the atmosphere and passively aerated. The open-top set-up of column does not produce aerobic condition inside the column. The condition at various heights in the column was not verified and cannot be assumed to be aerobic.

The authors should repeat part of the experiments, or simply create a setup similar to the experimental installation by filling a column and leaving the top part of the column open to the atmosphere, and then taking samples from various locations (heights) inside the column and checking the DO and ORP concentrations. This simple test, although does not fully reproduce the experimental conditions, will provide an estimation of the environmental condition inside the experimental column.

3. Figure 3 shows that the concentration of all parameters in the control system was very low. Normally, the control system operates in such a way that no removal of any contaminant takes place. The authors should clarify the role of control system in their study, and the reasons behind the removal of all contaminants to very low levels in this system. Actually, on lines 312-314 (revised manuscript), the authors state that no significant differences in COD removal were observed between the control system and the treatment systems with adsorption media. Therefore, the suggested method for the removal of contaminants offers no particular value. The reason behind this observation should be stated and discussed.

4. Lines 361-362 (revised manuscript): It is stated that treatment during the reduced temperature phase was not significantly different. It is well known that temperature significantly affects the activity of biomass and controls the kinetics of biological processes. The authors should comment on the lack of temperature impact on the treatment process, which is suggested to be biological.

5. Section 3.1.2.: What were the mechanisms of ammonia and nitrate removal in the control system that did not contain any clinoptilolite?

6. Lines 419-421: The authors state that “nitrate was removed immediately in the first column of each system, indicating that both aerobic and anoxic zones were present in the columns or biofilms. This immediate removal was also observed in the Start-Up test, where columns were not inoculated.” What was the mechanism of nitrate removal in the Start-Up test, where columns were not inoculated? Based on this observation, how can the authors claim that the removal of nitrate was accomplished by the biofilm in the first column of each system?

Author Response

The design of the columns used in this study is most similar to a fixed film, free-flowing gravel bed. To take multiple samples and measure DO at various heights throughout the columns disrupts flow patterns. Additionally, we are not aware of an ORP meter specifically for packed bed reactors.

We understand your concern and agree that because the DO and ORP were not measured, we are assuming that aerobic and anoxic conditions were present. We have made it clearer that this was an assumption on Lines 174-175. However, it has been well-established that trickling filters, the conventional treatment system that is most similar to the columns tested in this study, use aerobic process to treat wastewater (Tchobanoglous et al., 2014). Additionally, intermittent loading, such as was used in this study, leads to increased oxygen transfer in subsurface vertical flow constructed wetlands, promoting a more oxidizing environment for nitrification processes (Wang, Zhang, Dong, & Tan, 2017). Another study of winery wastewater treatment used similarly designed columns as a part of their treatment system and also assumed the columns to aerobic (Arienzo, Christen, Quayle, & Di Stefano, 2009). These references have been added to Line 175.

Additionally, our lab partnered in a study that monitored a full-scale land application treatment system treating food processing wastewater (Dong, Safferman, Miller, Hruby, & Bratt, 2017). After 8 years of extensive monitoring (including hydraulic and organic loading, and daily averages of water content, oxygen levels, and temperature measurements taken every 5 minutes), we found that the conditions 1, 2, and 3 feet below the surface remained aerobic throughout the year. The instantaneous hydraulic loading rate used in that study (2.85 L/m²/loading) was lower than in this study (1.7175 L/m²/loading) and the bed media was loamy sand which has a significantly higher water holding capacity compared to the pea gravel in our study. Based on these conditions and results, we can assume that our system also maintained aerobic conditions. This reference has also been added to Line 175.

Arienzo, M., Christen, E. W., Quayle, W., & Di Stefano, N. (2009). Development of a Low-Cost Wastewater Treatment System for Small-Scale Wineries. Water Environment Research, 81(3), 233-241. doi:10.2175/106143008x274356

Dong, Y., Safferman, S., Miller, S., Hruby, J., & Bratt, D. (2017). Effectiveness of food processing wastewater irrigation. Paper presented at the 90th Annual Water Environment Federation Exhibition Conf.

Tchobanoglous, G., Stensel, H. D., Tsuchihashi, R., Burton, F. L., Abu-Orf, M., Bowden, G., & Pfrang, W. (2014). Wastewater Engineering: Treatment and Resource Recovery (5th ed.): McGraw-Hill Education.

Wang, M., Zhang, D. Q., Dong, J. W., & Tan, S. K. (2017). Constructed wetlands for wastewater treatment in cold climate — A review. Journal of Environmental Sciences, 57, 293-311. doi:https://doi.org/10.1016/j.jes.2016.12.019

Unfortunately, we do not have the funds to perform another experiment as you suggest. Additionally, the set-up as proposed would disrupt the flow pattern of the wastewater in the column and not accurately represent the dynamic environment of the previously tested columns.

However, we are confident that oxygen was present within the columns that were assumed to be aerobic. This is due to the basic principle of bulk density where the entire volume consists of the media, water, and air. When water is not loaded into the column, the voids of the media are filled with air. When water is loaded into the column, it flows through the porous media and absorbs oxygen from the air (Arienzo et al., 2009). Please see the response to Comment 1 for the study that confirmed that aerobic conditions are present when wastewater is applied top-down in a porous medium. Similar to Comment 1, we have updated Lines 181-182 to clarify that the aerobic conditions were assumed.

The control system in this research is not abiotic but rather was designed to compare performance with and without the nitrogen sorbing materials (Clinoptiloite, tire chips, and oyster shells). This is stated in line 189, Table 3. A control can serve many purposes and is not always designed to prevent microbial processes. The lack of a statistical difference in COD removal between the treatment systems is not un-expected as the amendments to the SVFCW were for nitrogen removal only. This has been clarified on Line 314 as part of the COD discussion. Thank you for your comments in helping us to clarify this point.

Thank you for highlighting this point as this is an important area of study for the use of constructed wetlands in cold-weather climates and is one that is controversial (Wang et al., 2017). We certainly agree that biological kinetics decrease as temperature decreases. However, multiple studies on cold-weather constructed wetlands have actually found that BOD removal efficiencies are not impacted (Campbell & Safferman, 2015; Chang, Wu, Dai, Liang, & Wu, 2012; R. Kadlec & Knight, 1996; R. H. Kadlec & Reddy, 2001; Rozema, Rozema, & Zheng, 2016). The exact reasons for this phenomenon are not well understood and it is the authors’ understanding that this is an on-going area of research in wastewater treatment using constructed wetlands.

Additionally, on Lines 234-237, we state: “Wastewater was distributed at an approximate surface loading rate of 5.18 kg COD/m²/d (1.06e-2 lb COD/ft²/d); this loading rate was previously determined to be optimum for a cold-weather SVFCW treating milking facility wastewater [45] but had not been tested for winery wastewater”. As such, even if the kinetics of the biological processes were affected, the treatment system has enough capacity to accommodate this reduction. As stated on Lines 333-336, it is unlikely for complete removal of COD due to recalcitrant soluble fractions of COD in winery wastewater. Therefore, it is not unexpected that performance in the warmer temperatures was not higher than the colder temperatures for which it was designed.

We updated Lines 363-365 to clarify that the treatment system was originally sized for cold-weather operation.

Campbell, E. L., & Safferman, S. I. (2015). Design Criteria for the Treatment of Milking Facility Wastewater in a Cold Weather Vertical Flow Wetland. Transactions of the ASABE, 58(6), 1509-1519. doi:10.13031/trans.58.11068

Chang, J.-j., Wu, S.-q., Dai, Y.-r., Liang, W., & Wu, Z.-b. (2012). Treatment performance of integrated vertical-flow constructed wetland plots for domestic wastewater. Ecological Engineering, 44, 152-159.

Kadlec, R., & Knight, R. (1996). Treatment wetlands. Lewis Publ., Boca Raton. Treatment wetlands. Lewis Publ., Boca Raton., -.

Kadlec, R. H., & Reddy, K. R. (2001). Temperature effects in treatment wetlands. Water Environ Res, 73(5), 543-557. doi:10.2175/106143001x139614

Rozema, E. R., Rozema, L. R., & Zheng, Y. (2016). A vertical flow constructed wetland for the treatment of winery process water and domestic sewage in Ontario, Canada: Six years of performance data. Ecological Engineering, 86, 262-268.

5. Section 3.1.2.: What were the mechanisms of ammonia and nitrate removal in the control system that did not contain any clinoptilolite?

As described in the manuscript, the C:N:P ratio of the wastewater (Lines 440-450), the COD removal start-up time (Lines 337-350), the low adsorption capacity of gravel and the observed ammonia from the second column (Lines 404-413), and the high removal efficiencies of ammonia and nitrate indicate that classic nitrification and denitrification were occurring within the columns.

Please see Lines 347-348, in addition the response to Comment 5. Malandra, Wolfaardt, Zietsman, and Viljoen-Bloom (2003) did an exhaustive microbial analysis of winery wastewater and found that the microbial population naturally present within winery wastewater (a result of bacteria and fungi from grapes) is capable of colonizing a treatment system. The results observed in this study support this hypothesis. We added to Line 348 to make this connection clearer.

Malandra, L., Wolfaardt, G., Zietsman, A., & Viljoen-Bloom, M. (2003). Microbiology of a biological contactor for winery wastewater treatment. Water Res, 37(17), 4125-4134. doi:10.1016/S0043-1354(03)00339-7

Round 3

Reviewer 4 Report

The authors have answered my comments and questions. The manuscript can be accepted for publication.