the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Jul 2024
Beyond Total Impervious Area: A New Lumped Descriptor of Basin-Wide Hydrologic Connectivity for Characterizing Urban Watersheds
Abstract. Urbanization impacts on hydrologic response are typically indexed as a function of the fraction of total impervious area (TIA), i.e., the proportion of impervious areas in a basin. This implicitly assumes that changes in flood characteristics are somehow proportional to the extents of land-development, without considering that such impacts may vary widely depending on the location of the developed areas with respect to each other, the less-developed land patches, the stream network, and the basin outlet. In other words, TIA is blind to the spatial arrangement of the different types of land patches within a basin, and to the nuanced ways in which runoff volumes are differentially generated over them and then subsequently retained or detained, as they are routed towards the stream network and then the outlet. To overcome such limitations, we propose a new lumped index that measures the impacts of urbanization on basin response in terms of the emerging hydrologic connectivity, the distributed, directional basin property driven by topographically induced runoff pathways and locally affected by the different land-use/land-cover types present in a watershed. This alternative, hydrologic-connectivity-based index of urbanization (HCIU) displays sensitivity to the spatial arrangement of both fully developed as well as less developed or undeveloped patches, each with different degrees of imperviousness, roughness, and other characteristics affecting their abilities to either generate or else retain/detain runoff, reflecting their distinct localized effects on hydrologic connectivity. The proposed HCIU can be readily obtained in a GIS environment from easily available raster geospatial data. We found that HCIU improves the predictive power of regional equations for peak flow in three large case-study homogeneous regions, when used in place of the traditional TIA.
- Preprint
(10117 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-1956', Anonymous Referee #1, 10 Oct 2024
Beyond Total Impervious Area: A New Lumped Descriptor of Basin-Wide Hydrologic Connectivity for Characterizing Urban Watersheds
General Comments:
The authors present a novel approach to characterize hydrologic connectivity of urban watersheds using the hydrologic-connectivity based index (HCIU), moving beyond typical total impervious metrics. The HCIU is a lumped index computed in a GIS environment from raster data. The connectivity of runoff producing patches are assigned varying degrees of imperviousness, and may be weighted topographically and based on roughness (Manning’s surface roughness coefficient) or the Curve Number. Evaluation of each HCIU index was done against total impervious area (TIA) and done using predictive peak-flow equations using three case study areas with several catchments with varying levels of imperviousness. The authors found that the HCIU out preforms traditional TIA metrics.
The paper is technically sound, scientifically significant and well written. I think the authors do a good job explaining and testing their approach, however I think integrating the limitations of the approach into the introduction would be beneficial for readers considering implementing a similar methodology.
Specific comments:
The number of studies relying solely on TIA as an indicator of hydrologic response is beginning to decrease, although I agree with the authors that historically it has been one of the most widely used methodologies. I think it would be beneficial if the authors made it more clear why they opted to compare to their approach to TIA compared to other more robust metrics. For example, the authors mention the use of the effective impervious area (EIA) and the directly connected impervious area (DCIA) however do not compare to these approaches. I would recommend adding some additional justification as to why the authors compared the HCIU index to TIA only.
Although the paper is technically sound, I think that it would be better if the authors did not wait until the end of the paper to address the limitations of the proposed methodology. I think it is very important to address that this approach does not currently account for the urban drainage system. The urban drainage systems in heavily urbanized watersheds are a major runoff routing component and are important for accurately evaluating hydrologic connectivity. I think it is important to address that for heavily urbanized watersheds differences in connectivity may be controlled by the presence and capacity of storm and combined sewers. This component is missing from the methodology based on the limitations addressed in the discussion. However, based on the methodological approach could be added relatively easily, which is excellent. I agree with the authors that it may be difficult to acquire sewer attribute information, however this is not this case in all Countries. In Canada for example much of the storm sewer attribute information is now being made publicly accessible via OpenData government portals. I think making this clear in the introduction may make the authors approach more accessible to individuals who may have access to this information.
The proposed HCIU index is presently heavily dependent on topography, in some heavily urbanized regions (impervious cover above 80%) stormwater pumping across topographic gradients, and stormwater detention tanks may impact outcomes. In the case study component, the authors have a few catchments with high levels of imperviousness but these catchments seem to be relatively small and were only part of the VA case-study. I would suggest addressing this as a limitation in the paper.
I found the explanation and use of figure 1b, for creating a totally impervious copy to be complicated. I think it would be nice if the authors could simplify this text or provide a different graphic to support this process.
I am curious to why the authors evaluate the predictive power of their approach (HCIU) using peak flow only. Why did the authors not consider other important metrics of hydrologic response?
With regards to the case studies could you provide what type of flow data was used to generate the flow statistics? Daily, or sub-daily?
The authors mention in the acknowledgments the need for computing power, is this a limitation of the approach? It would be nice to know what technical equipment they had access to for completing the analysis. Is this approach feasible for local governments or conservation groups to do?
I think it could be beneficial for the authors to discuss the role of major vs minor system with regards to event size and hydrologic connectivity, especially for the peak flows represented by the extreme flood values.
Do you think the poor overall performance of the HCIU(CN) could be due to the quality of the raster product you are using?
Line 560: This is interesting. Please elaborate on this point.
Technical Corrections:
Line 159: I would like to see some examples of LULC types.
Line 124: The references are repeated multiple times.
Figure 1: Suggest expanding on your figure caption to help explain the process in more detail.
Figure 2: What Ecoregion is this?
Line 252-254: Suggest adding a reference to Appendix Table A1.
Figure 4: add legend item for shaded blue bars as you have one for the basin averages.
Line 329-330: Suggest defining the LULC ranges for the different percentages in the text as you have done in the figure.
Figure 6: with all the reference lines I found this confusing. Perhaps you could just color code the points you are highlighting based on the outline of the box color.
Citation: https://doi.org/10.5194/egusphere-2024-1956-RC1 -
AC1: 'Reply on RC1', Francesco Dell'Aira, 20 Oct 2024
We thank the Reviewer for their in-depth analysis of our work, which is clear from the comprehensiveness of their review. We are grateful for the constructive comments.
We attach a PDF file with replies to the comments, explaining how we plan to modify the manuscript accordingly.
Thank you.
-
AC3: 'Additional reply to RC1', Francesco Dell'Aira, 25 Oct 2024
Dear Reviewer 1,
We include this additional reply to please ask for some more information as to how to find stormwater sewer network data for Canadian urban areas. As we mentioned in our reply to your comments, we are highly interested in examining these data, to possibly use them within our ongoing research. We are specifically interested in finding the web links to any storm sewer dataset that you would know of, including vectorized data for the pipe network, inlet and outlet nodes, as well as information about pipe diameter sizes and materials. We were able to track down the open data repository at the link https://open.canada.ca/en but, from a preliminary search, it seems that it does not include information on the stormwater pipe network layout for any municipality. We are unsure whether you were referring to that specific repository (and if you were indeed, whether we should try to dig deeper into it). We would like to please ask you for further help in locating publicly accessible GIS vectorized data for any sewer network infrastructure in urbanized areas in Canada, including network links as well as inlet and outlet nodes. We thank you in advance for any information you can provide us regarding these data.
Citation: https://doi.org/10.5194/egusphere-2024-1956-AC3 -
RC3: 'Reply on AC3', Anonymous Referee #1, 25 Oct 2024
Great question. No, I was not referring to that repository, but interesting. You need fairly decent knowledge of the municipalities to find the data as it can be tricky. Below are some examples of data that is available, although the repositories may not have all the attribute information you want, it could be worth taking a look.
City of Mississauga, Ontario Canada.
Nodes: https://hub-mississauga.opendata.arcgis.com/datasets/mississauga::storm-sewer-network-storm-node/about
Pipes: https://data.mississauga.ca/datasets/8aff41843ec44a74a309148d28e1b989_0/explore?location=43.593771%2C-79.681590%2C13.51
City of Hamilton, Ontario, Canada (although it does not seem to be working at the moment)
https://spatialsolutions.maps.arcgis.com/apps/webappviewer/index.html?id=52396c951e9c49a39ba64c674c99cd46
City of Barrie, Ontario, Canada
https://opendata.barrie.ca/datasets/barrie::storm-linear/explore?location=44.362729%2C-79.688183%2C19.12
City of Kingston, Ontario Canada
https://opendatakingston.cityofkingston.ca/explore/dataset/storm-pipe/map/?location=10,44.24749,-76.16718&basemap=72525b
Good luck with your searches, you may want to consider taking a look at the municipal breakdown to help you find data.
For Ontario only: https://data.ontario.ca/dataset/municipal-boundaries
Citation: https://doi.org/10.5194/egusphere-2024-1956-RC3 -
AC4: 'Reply on RC3', Francesco Dell'Aira, 25 Oct 2024
Thank you so much for pointing out these resources. We truly appreciate your help!
Best regards,
FD
Citation: https://doi.org/10.5194/egusphere-2024-1956-AC4
-
AC4: 'Reply on RC3', Francesco Dell'Aira, 25 Oct 2024
-
RC3: 'Reply on AC3', Anonymous Referee #1, 25 Oct 2024
-
AC1: 'Reply on RC1', Francesco Dell'Aira, 20 Oct 2024
-
RC2: 'Comment on egusphere-2024-1956', Anonymous Referee #2, 11 Oct 2024
The paper’s theme is relevant and actual, as the techniques and models used show a new use for the index. Adapting the connectivity index to understand urban areas shows a new use for the index.
The paper shows excellent Scientific Significance, Scientific Quality and Presentation Quality. There is only one main question that needs to be clarified:
Lines 217-230—It is unclear how implementing an “along-the-stream network” differs from the well-known IC_outlet approach from Cavalli et al. (2013) and several other researchers/papers. It is necessary to explain why to chose this new approach over IC_outlet.
Other minor issues:
Figure 4: What do the blue bars represent?
No comment exists about how the urban drainage structure could affect urban hydrology.
Citation: https://doi.org/10.5194/egusphere-2024-1956-RC2 -
AC2: 'Reply on RC2', Francesco Dell'Aira, 20 Oct 2024
We thank the Reviewer for their valuable comments, which help better place the proposed methodology in the context of the existing literature on the connectivity index.
We attach here a PDF file with replies to the comments, also explaining how we plan to address them in a revised version of the manuscript.
Thank you.
-
AC2: 'Reply on RC2', Francesco Dell'Aira, 20 Oct 2024
Status: closed
-
RC1: 'Comment on egusphere-2024-1956', Anonymous Referee #1, 10 Oct 2024
Beyond Total Impervious Area: A New Lumped Descriptor of Basin-Wide Hydrologic Connectivity for Characterizing Urban Watersheds
General Comments:
The authors present a novel approach to characterize hydrologic connectivity of urban watersheds using the hydrologic-connectivity based index (HCIU), moving beyond typical total impervious metrics. The HCIU is a lumped index computed in a GIS environment from raster data. The connectivity of runoff producing patches are assigned varying degrees of imperviousness, and may be weighted topographically and based on roughness (Manning’s surface roughness coefficient) or the Curve Number. Evaluation of each HCIU index was done against total impervious area (TIA) and done using predictive peak-flow equations using three case study areas with several catchments with varying levels of imperviousness. The authors found that the HCIU out preforms traditional TIA metrics.
The paper is technically sound, scientifically significant and well written. I think the authors do a good job explaining and testing their approach, however I think integrating the limitations of the approach into the introduction would be beneficial for readers considering implementing a similar methodology.
Specific comments:
The number of studies relying solely on TIA as an indicator of hydrologic response is beginning to decrease, although I agree with the authors that historically it has been one of the most widely used methodologies. I think it would be beneficial if the authors made it more clear why they opted to compare to their approach to TIA compared to other more robust metrics. For example, the authors mention the use of the effective impervious area (EIA) and the directly connected impervious area (DCIA) however do not compare to these approaches. I would recommend adding some additional justification as to why the authors compared the HCIU index to TIA only.
Although the paper is technically sound, I think that it would be better if the authors did not wait until the end of the paper to address the limitations of the proposed methodology. I think it is very important to address that this approach does not currently account for the urban drainage system. The urban drainage systems in heavily urbanized watersheds are a major runoff routing component and are important for accurately evaluating hydrologic connectivity. I think it is important to address that for heavily urbanized watersheds differences in connectivity may be controlled by the presence and capacity of storm and combined sewers. This component is missing from the methodology based on the limitations addressed in the discussion. However, based on the methodological approach could be added relatively easily, which is excellent. I agree with the authors that it may be difficult to acquire sewer attribute information, however this is not this case in all Countries. In Canada for example much of the storm sewer attribute information is now being made publicly accessible via OpenData government portals. I think making this clear in the introduction may make the authors approach more accessible to individuals who may have access to this information.
The proposed HCIU index is presently heavily dependent on topography, in some heavily urbanized regions (impervious cover above 80%) stormwater pumping across topographic gradients, and stormwater detention tanks may impact outcomes. In the case study component, the authors have a few catchments with high levels of imperviousness but these catchments seem to be relatively small and were only part of the VA case-study. I would suggest addressing this as a limitation in the paper.
I found the explanation and use of figure 1b, for creating a totally impervious copy to be complicated. I think it would be nice if the authors could simplify this text or provide a different graphic to support this process.
I am curious to why the authors evaluate the predictive power of their approach (HCIU) using peak flow only. Why did the authors not consider other important metrics of hydrologic response?
With regards to the case studies could you provide what type of flow data was used to generate the flow statistics? Daily, or sub-daily?
The authors mention in the acknowledgments the need for computing power, is this a limitation of the approach? It would be nice to know what technical equipment they had access to for completing the analysis. Is this approach feasible for local governments or conservation groups to do?
I think it could be beneficial for the authors to discuss the role of major vs minor system with regards to event size and hydrologic connectivity, especially for the peak flows represented by the extreme flood values.
Do you think the poor overall performance of the HCIU(CN) could be due to the quality of the raster product you are using?
Line 560: This is interesting. Please elaborate on this point.
Technical Corrections:
Line 159: I would like to see some examples of LULC types.
Line 124: The references are repeated multiple times.
Figure 1: Suggest expanding on your figure caption to help explain the process in more detail.
Figure 2: What Ecoregion is this?
Line 252-254: Suggest adding a reference to Appendix Table A1.
Figure 4: add legend item for shaded blue bars as you have one for the basin averages.
Line 329-330: Suggest defining the LULC ranges for the different percentages in the text as you have done in the figure.
Figure 6: with all the reference lines I found this confusing. Perhaps you could just color code the points you are highlighting based on the outline of the box color.
Citation: https://doi.org/10.5194/egusphere-2024-1956-RC1 -
AC1: 'Reply on RC1', Francesco Dell'Aira, 20 Oct 2024
We thank the Reviewer for their in-depth analysis of our work, which is clear from the comprehensiveness of their review. We are grateful for the constructive comments.
We attach a PDF file with replies to the comments, explaining how we plan to modify the manuscript accordingly.
Thank you.
-
AC3: 'Additional reply to RC1', Francesco Dell'Aira, 25 Oct 2024
Dear Reviewer 1,
We include this additional reply to please ask for some more information as to how to find stormwater sewer network data for Canadian urban areas. As we mentioned in our reply to your comments, we are highly interested in examining these data, to possibly use them within our ongoing research. We are specifically interested in finding the web links to any storm sewer dataset that you would know of, including vectorized data for the pipe network, inlet and outlet nodes, as well as information about pipe diameter sizes and materials. We were able to track down the open data repository at the link https://open.canada.ca/en but, from a preliminary search, it seems that it does not include information on the stormwater pipe network layout for any municipality. We are unsure whether you were referring to that specific repository (and if you were indeed, whether we should try to dig deeper into it). We would like to please ask you for further help in locating publicly accessible GIS vectorized data for any sewer network infrastructure in urbanized areas in Canada, including network links as well as inlet and outlet nodes. We thank you in advance for any information you can provide us regarding these data.
Citation: https://doi.org/10.5194/egusphere-2024-1956-AC3 -
RC3: 'Reply on AC3', Anonymous Referee #1, 25 Oct 2024
Great question. No, I was not referring to that repository, but interesting. You need fairly decent knowledge of the municipalities to find the data as it can be tricky. Below are some examples of data that is available, although the repositories may not have all the attribute information you want, it could be worth taking a look.
City of Mississauga, Ontario Canada.
Nodes: https://hub-mississauga.opendata.arcgis.com/datasets/mississauga::storm-sewer-network-storm-node/about
Pipes: https://data.mississauga.ca/datasets/8aff41843ec44a74a309148d28e1b989_0/explore?location=43.593771%2C-79.681590%2C13.51
City of Hamilton, Ontario, Canada (although it does not seem to be working at the moment)
https://spatialsolutions.maps.arcgis.com/apps/webappviewer/index.html?id=52396c951e9c49a39ba64c674c99cd46
City of Barrie, Ontario, Canada
https://opendata.barrie.ca/datasets/barrie::storm-linear/explore?location=44.362729%2C-79.688183%2C19.12
City of Kingston, Ontario Canada
https://opendatakingston.cityofkingston.ca/explore/dataset/storm-pipe/map/?location=10,44.24749,-76.16718&basemap=72525b
Good luck with your searches, you may want to consider taking a look at the municipal breakdown to help you find data.
For Ontario only: https://data.ontario.ca/dataset/municipal-boundaries
Citation: https://doi.org/10.5194/egusphere-2024-1956-RC3 -
AC4: 'Reply on RC3', Francesco Dell'Aira, 25 Oct 2024
Thank you so much for pointing out these resources. We truly appreciate your help!
Best regards,
FD
Citation: https://doi.org/10.5194/egusphere-2024-1956-AC4
-
AC4: 'Reply on RC3', Francesco Dell'Aira, 25 Oct 2024
-
RC3: 'Reply on AC3', Anonymous Referee #1, 25 Oct 2024
-
AC1: 'Reply on RC1', Francesco Dell'Aira, 20 Oct 2024
-
RC2: 'Comment on egusphere-2024-1956', Anonymous Referee #2, 11 Oct 2024
The paper’s theme is relevant and actual, as the techniques and models used show a new use for the index. Adapting the connectivity index to understand urban areas shows a new use for the index.
The paper shows excellent Scientific Significance, Scientific Quality and Presentation Quality. There is only one main question that needs to be clarified:
Lines 217-230—It is unclear how implementing an “along-the-stream network” differs from the well-known IC_outlet approach from Cavalli et al. (2013) and several other researchers/papers. It is necessary to explain why to chose this new approach over IC_outlet.
Other minor issues:
Figure 4: What do the blue bars represent?
No comment exists about how the urban drainage structure could affect urban hydrology.
Citation: https://doi.org/10.5194/egusphere-2024-1956-RC2 -
AC2: 'Reply on RC2', Francesco Dell'Aira, 20 Oct 2024
We thank the Reviewer for their valuable comments, which help better place the proposed methodology in the context of the existing literature on the connectivity index.
We attach here a PDF file with replies to the comments, also explaining how we plan to address them in a revised version of the manuscript.
Thank you.
-
AC2: 'Reply on RC2', Francesco Dell'Aira, 20 Oct 2024
Model code and software
HCIU urbanization metric Francesco Dell'Aira https://github.com/dllaira/HCIU-urbanization-metric
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 208 | 96 | 94 | 398 | 6 | 8 |
- HTML: 208
- PDF: 96
- XML: 94
- Total: 398
- BibTeX: 6
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1