Based upon your recommended changes in your to the previous question, how would you expect the data of Menzies et al. (2022) to impact predictions for the percentage of polyvinyl alcohol in post-treatment wastewater?

1. The data from Menzies et al. (2022) will provide a good reference to the predictions of the percentage of polyvinyl alcohol in post-treatment wastewater. Data of Menzies et al. (2022) is expected to provide more reliable data on the emission of PVA in wastewater and the degradation of PVA in wastewater. The data/findings should be able to determine if the PVA emission is a threat to the ecosystem due to the potential mobilization of heavy metals and other hydrophilic contaminants.

2. The published data by Menzies et al. (2022) will help Rolsky and Kelkar (2021) to re-evaluate their assumptions of PVA content (and the amount of degraded PVA) in post-treatment wastewater. On the other hand, the data from Menzies et al. (2022) will also help Rolsky and Kelkar(2021) to revise their calculated amount of nationwide PVA emissions from WWTPs to landfill and water bodies.

3. The findings from the study should be able to suggest the solution to reduce PVA presence in the environment. For example, future direction and recommendation for further study should be included.

I would expect a much higher removal capacity in a WWTP treating wastewater containing PVA. The respirometric assays can be useful to determine real PVA specific removal rates. If the kinetics can be determined it will help to use a more well-developed kinetic model with much better prediction capabilities.

In comparison with the Rolsky & Kelkar model, I would expect that an approach based primarily on the data of Menzies et al. to lead to much lower estimates of PVA entering the environment. With a properly functioning activated sludge return, the residual amounts in activated sludge would probably be at least one order of magnitude lower than expected based on the Rolsky & Kelkar model. This would then also significantly decrease the PVA input for the anaerobic digester, which might be considered the bottleneck in a model which is more in line with the suggestions made in the previous answers. Lower input in the anaerobic digester would lead to lower output in the absolute sense, but it is also likely that the removal efficiency might be significantly higher with a lower influx. Combining this with higher removal efficiencies in the aqueous portions of the model (supported by high degree of biodegradation observed in environmental river water samples at orders of magnitude higher PVA concentrations), the output parameters landfill/incineration/fertilizer/effluent would all contain lower amounts of PVA. This effect would likely be most significant in effluent (= post-treatment wastewater), since several stages are likely to have increased removal by biodegradation (aeration basin, secondary clarifier, sand filter and potentially even the disinfection basin). I would expect output from anaerobic digester may decrease with a factor 2-3, and output in effluent with at least a factor 5-10.

I would like to note that, in my opinion, it would not be acceptable to revise the entire Rolsky & Kelkar model based on just the data and conclusions from the Menzies et al. paper. However, the findings in the Menzies et al. paper are completely in line with most of the reasoning I used in answering the questions above. In my critical review of the Rolsky & Kelkar model, I came across several significant factors that would have impacted the outcome in a similar direction as using the Menzies et al. data. I have not evaluated all references used by Rolsky & Kelkar at the level of detail applied to the Menzies et al. paper. In my opinion, the Menzies et al. paper is a very solid support for the reasoning I applied to the Rolsky & Kelkar model, but some of the references might have been equally supportive for this alternative approach.

The study could have benefitted from spot sampling of effluents from wastewater facilities across the country, to verify or refine the predictions. Without that kind of verification, the results will always be subject to uncertainty. So, if the changes to online survey and estimates of PVA degradation are refined through uncertainty and sensitive analyses, and the model results are supplemented with real data through environmental sampling, the predictions of the percentages of PVA will be reported not as approximate values (i.e. ~61% of PVA ending up in the environment via the sludge route and ~15.7%
via the aqueous phase), but with ranges that capture scenarios which are very influential in determining the fate according to various geographic, technological, and demographic variables.

No major change as Menzies showed overall biodegradation potential in long-term studies, giving bacteria time to grow and adapt to the prevalent food source (PVA). The much shorter hydraulic residence time in WWTPs makes it unlikely to achieve much greater reductions in PVA; though more PVA is potentially degraded post treatment once released into the environment.

I think that in the long run, and provided that the key knowledge/data gaps identified in response to the previous, the data of Menzies et al. will certainly affect the prediction of levels of PVA in post-treatment wastewater. The key reason for this is the fact that I think that microbial adaptation to PVA is likely to happen in WWTPs, as opposed to the assumption of Rolsky and Kelkar in their model.
I am not convinced about the option of extrapolating test-results according to the OECD guidelines, to natural waters. Here more data are to be generated.

To my mind, the data from Menzies et al. (2022) do not significantly change the findings of Rolsky and Kelkar (2021). The lag periods observed before PVA degradation commences in their lab trials are of the same order or longer as the retention periods in ASP in WWTPs reported by Rolsky and Kelkar. This suggests that, if anything, the degradation values used by the Rolsky and Kelkar model are optimistic. And for PVA that leaves WWTPs (or fails to enter them in the first place), the data from Menzies et al. are not encouraging, with variable but generally low degradation rates observed from the river inoculum trials, despite their trials seemingly being conducted at favourable (but possibly unrepresentative) temperatures (22˚C).

Yes I would expect >80% removal in total by biodegradation in WWTPs for ¨ water-soluble¨ PVA. Predicting the removal of ´ ¨lower¨ water-soluble¨ is requires better data. Test periods of anaerobic tests and experiments should at least be 60 days (see for instance OECD 311).