the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Jan 2024
Gridded dataset of nitrogen and phosphorus point sources from wastewater in Germany (1950–2019)
Abstract. Knowledge about the long history of the anthropogenic inputs of Nitrogen (N) and Phosphorus (P) is crucial to capture long-term N and P processes (legacies) and to investigate water quality and ecosystem health. These inputs include N and P point sources, that originate mainly from wastewater, and that are directly discharged into surface waters, thus having an immediate impact on ecosystem functioning. However, N and P point sources are challenging to estimate, due to the scarcity and uncertainty in observational data. Here, we contribute towards improved characterisation of N and P point sources from wastewater, by providing a long-term (1950–2019), high resolution (0.015625° ≈ 1.4 km on average) dataset for Germany. The dataset includes both the domestic and industrial emissions treated in wastewater treatment plants and untreated domestic emissions that are collected in the sewer system. We adopt a modelling approach that relies on a large range of data collected from different sources. Importantly, we account for the uncertainties arising from different modelling choices (i.e., coefficients and downscaling approach). We provide 200 gridded N and P point sources realisations, which are constrained and evaluated using available (recent) observations of wastewater treatment plants outgoing loads. We discuss the uncertainties in our reconstructed dataset over a large sample of river basins in Germany, and provide guidance for future uses. Overall, by capturing the long-term spatial and temporal variations in N and P point sources and accounting for uncertainties, our dataset can facilitate long-term and large-scale robust water quality studies. The dataset is available at https://doi.org/10.5281/zenodo.10500535 (Sarrazin et al., 2024).
- Preprint
(2095 KB) - Metadata XML
-
Supplement
(9922 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on essd-2023-474', Maryna Strokal, 12 Mar 2024
The authors did a good job in developing and providing the gridded datasets for point source emissions of N and P. the manuscript is well written with good structure. Complements to the authors on balancing well between details and general descriptions. It is easy to follow the methodology. I appreciate that the authors analyzed the uncertainties in such detailed datasets. I have a few points that could improve the relevance and importance of this work:
1. The data sources were mainly at the NUTS-1 scale. The emissions of point sources of N and P are downscaled to grids. Uncertainties are analyzed at the river basin scale. It is an interesting choice for the scales. It would be good to reflect on this choice, especially on the choice for the gridded emissions, but uncertainties in those emissions are analyzed at the river basin. Why was that choice made? How can this basin scale uncertainty analysis build trust in modeled gridded emissions considering the data sources at the NUTS-1 level?
2. The role of rural point source emissions is important. Germany is a country with a lot of sewage systems in urban and rural areas. It would make the paper stronger if more discussion is provided on how rural emissions are considered, and the role of rural sanitation in those emissions. This might be interesting for other countries. Some countries do not have a lot of rural sewage systems. In this case: how can the proposed methods be still useful?
3. This point is a bit also related to the previous: the manuscript would benefit from a discussion the applicability of the proposed methods for other regions and countries. countries differ in their urban and rural waste management. On top of this, not all countries have such detailed datasets at the NUTS-1 levels. This is especially true for developing countries. Can the proposed methods be used for those developing countries, if yes, what needs to be adjusted? if not, why? what would be alternatives?
4. Some detailed comments:
- Please clarify the forms of N and p that are modeled. Please also justify the choice for those forms.
- Table 1 has lower and upper bound. it would also good to add mean or median
- Why point sources? Diffuse sources are as important as point sources and are more difficult to control.
- Why Germany? Can we learn from this exercise for other countries?
Citation: https://doi.org/10.5194/essd-2023-474-RC1 - AC1: 'Reply on RC1', Fanny Sarrazin, 19 May 2024
-
RC2: 'Comment on essd-2023-474', Olga Vigiak, 10 Apr 2024
Dear Authors,
first of all many apologies for being late and delaying the review process. The manuscript describes an excellent dataset for nutrient emissions (N and P) to waters in Germany from 1950 to 2019. The authors compile a very thorough dataset that can be very useful.
i have only minor comments for improving the manuscript.
1) I like very much the uncertainty analysis, but I miss in the results a reporting of the final uncertainty of the parameters. I think this could be very interesting as it can affect similar analysis elsewhere. can you give revised intervals for the parameters and comment on the most sensitive ones? Referring only to a supplementary material part is too little in my opinion.
2) the paper is very long, and at times verbose, and should be summarized more, particularly for the Material and methods part - we get to results only at page 27. I understand the need for details, and appreciate the work behind collecting all the data, but i think maybe section 3 on data can be shortened. Also, some sentences can be simplified: for example, authors could more simply refer to the many supplementary figures by mentioning them in brackets instead of saying 'as reported in...' etc.
I have a question with regards to section 3.5.1 (maybe I missed it): the EU dataset reports for some UWWTPs incoming and outgoing N and P content, but usually this is only a smaller subsample of all UWWTPs reported in the database. are these the values that were used? what is the % of WWTPs that reported these data? Further, the EU dataset now reports also 2018 (and 2020), these data were not used (why)?
other minor points:
line87: i would remove 'Yes,'
dashed lines in figure 2 and figure 7 are not dashed in printing, maybe use of a different color could help.
lines 403-404: the sentence starting with 'The assumption is supported.... ' is should be shortened and simplified
Paragraph starting at line 630 has some very long sentences that are difficult to read.
anyway, many congratulations for your excellent work!
Citation: https://doi.org/10.5194/essd-2023-474-RC2 - AC2: 'Reply on RC2', Fanny Sarrazin, 19 May 2024
Status: closed
-
RC1: 'Comment on essd-2023-474', Maryna Strokal, 12 Mar 2024
The authors did a good job in developing and providing the gridded datasets for point source emissions of N and P. the manuscript is well written with good structure. Complements to the authors on balancing well between details and general descriptions. It is easy to follow the methodology. I appreciate that the authors analyzed the uncertainties in such detailed datasets. I have a few points that could improve the relevance and importance of this work:
1. The data sources were mainly at the NUTS-1 scale. The emissions of point sources of N and P are downscaled to grids. Uncertainties are analyzed at the river basin scale. It is an interesting choice for the scales. It would be good to reflect on this choice, especially on the choice for the gridded emissions, but uncertainties in those emissions are analyzed at the river basin. Why was that choice made? How can this basin scale uncertainty analysis build trust in modeled gridded emissions considering the data sources at the NUTS-1 level?
2. The role of rural point source emissions is important. Germany is a country with a lot of sewage systems in urban and rural areas. It would make the paper stronger if more discussion is provided on how rural emissions are considered, and the role of rural sanitation in those emissions. This might be interesting for other countries. Some countries do not have a lot of rural sewage systems. In this case: how can the proposed methods be still useful?
3. This point is a bit also related to the previous: the manuscript would benefit from a discussion the applicability of the proposed methods for other regions and countries. countries differ in their urban and rural waste management. On top of this, not all countries have such detailed datasets at the NUTS-1 levels. This is especially true for developing countries. Can the proposed methods be used for those developing countries, if yes, what needs to be adjusted? if not, why? what would be alternatives?
4. Some detailed comments:
- Please clarify the forms of N and p that are modeled. Please also justify the choice for those forms.
- Table 1 has lower and upper bound. it would also good to add mean or median
- Why point sources? Diffuse sources are as important as point sources and are more difficult to control.
- Why Germany? Can we learn from this exercise for other countries?
Citation: https://doi.org/10.5194/essd-2023-474-RC1 - AC1: 'Reply on RC1', Fanny Sarrazin, 19 May 2024
-
RC2: 'Comment on essd-2023-474', Olga Vigiak, 10 Apr 2024
Dear Authors,
first of all many apologies for being late and delaying the review process. The manuscript describes an excellent dataset for nutrient emissions (N and P) to waters in Germany from 1950 to 2019. The authors compile a very thorough dataset that can be very useful.
i have only minor comments for improving the manuscript.
1) I like very much the uncertainty analysis, but I miss in the results a reporting of the final uncertainty of the parameters. I think this could be very interesting as it can affect similar analysis elsewhere. can you give revised intervals for the parameters and comment on the most sensitive ones? Referring only to a supplementary material part is too little in my opinion.
2) the paper is very long, and at times verbose, and should be summarized more, particularly for the Material and methods part - we get to results only at page 27. I understand the need for details, and appreciate the work behind collecting all the data, but i think maybe section 3 on data can be shortened. Also, some sentences can be simplified: for example, authors could more simply refer to the many supplementary figures by mentioning them in brackets instead of saying 'as reported in...' etc.
I have a question with regards to section 3.5.1 (maybe I missed it): the EU dataset reports for some UWWTPs incoming and outgoing N and P content, but usually this is only a smaller subsample of all UWWTPs reported in the database. are these the values that were used? what is the % of WWTPs that reported these data? Further, the EU dataset now reports also 2018 (and 2020), these data were not used (why)?
other minor points:
line87: i would remove 'Yes,'
dashed lines in figure 2 and figure 7 are not dashed in printing, maybe use of a different color could help.
lines 403-404: the sentence starting with 'The assumption is supported.... ' is should be shortened and simplified
Paragraph starting at line 630 has some very long sentences that are difficult to read.
anyway, many congratulations for your excellent work!
Citation: https://doi.org/10.5194/essd-2023-474-RC2 - AC2: 'Reply on RC2', Fanny Sarrazin, 19 May 2024
Data sets
Gridded dataset of nitrogen and phosphorus point sources from wastewater in Germany (1950-2019) Fanny J. Sarrazin, Sabine Attinger, and Rohini Kumar https://doi.org/10.5281/zenodo.10500535
Model code and software
NP_point_sources_model: v1.0 Fanny J. Sarrazin and Rohini Kumar https://doi.org/10.5281/zenodo.10501238
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 516 | 120 | 65 | 701 | 78 | 46 | 38 |
- HTML: 516
- PDF: 120
- XML: 65
- Total: 701
- Supplement: 78
- BibTeX: 46
- EndNote: 38
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1