the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Mar 2026
Can Aerosols improve Urban Flood Prediction? A case study of 2015 Chennai extreme event
Abstract. In December 2015, Chennai, a coastal megacity in India, faced an extreme precipitation-flooding event (EPF) that triggered a devastating 1-in-100-year flood. Several previous attempts failed to accurately simulate the spatiotemporal variability of this EPF at the urban basin scale. Even though incorporating aerosols into operational weather models can improve the accuracy of EPF simulations, it is often ignored due to its computational cost. To address this, we conducted ensemble experiments to highlight the significance of aerosol-cloud interactions in simulating this EPF. In that regard, we use a computationally intensive, high-resolution WRF model configured in large-eddy simulation (LES) mode to represent the interactions in the complex urban microphysics. The results indicate that explicit aerosol representation significantly influenced the microphysics-dynamics interaction during the 2015 EPF and produced rainfall patterns in closer agreement with satellite and rain gauge observations, with basin-scale improvement of ~22 %. Further, employing simulated rainfall in a coupled hydrologic-hydraulic modeling framework increased inundation accuracy by ~50 %. Thus, this study suggests that explicit aerosol representation can improve the space-time simulation of rainfall and flooding for the EPF in Chennai and potentially for similar coastal megacities.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1648 KB)
-
Supplement
(5750 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1648 KB) - Metadata XML
-
Supplement
(5750 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2026-1218', Mirela-Adriana Anghelache, 21 Jul 2026
-
AC1: 'Reply on RC1', Oscar Paul, 01 Aug 2026
We sincerely thank the Reviewer for the thorough evaluation of our manuscript and for the constructive comments and valuable suggestions. Our point-by-point responses are provided below.
1. Cold pools can indeed strengthen with more precipitation—but you should explain the mechanism more clearly. For example: Enhanced warm-rain production increased precipitation reaching the lower troposphere, where evaporation of falling raindrops strengthened cold pools ... just in order the explanations shouldn't be too abrupt.
Reply: We thank the reviewer for the helpful suggestion. We have revised the text to clearly explain the physical sequence linking warm-rain processes and cold pool intensification. Specifically, we now state that enhanced warm-rain production due to low CCN increases precipitation where evaporation of falling raindrops strengthens cold pools, which subsequently enhance low-level moisture convergence and secondary convection.
The revised section in the manuscript is as follows: “Lower CCN concentrations favoured warm-rain processes and increased the amount of precipitation. The subsequent evaporation of falling raindrops enhanced evaporative cooling, thereby strengthening cold pool activity near the basin (Mallinson and Lasher-Trapp, 2019). These persistent cold pools intensified low-level moisture convergence through thermodynamic forcing, establishing favourable conditions for secondary convection (Schlemmer and Hohenegger, 2014; Schlemmer and Hohenegger, 2016), thereby simulating the bimodal rainfall peaks observed in the basin.”
2. Line 12: LES experiments further supported these findings instead of LES experiments further demonstrated; demonstrated is quite strong.
Reply: We agree with the reviewer's suggestion and have revised the text accordingly by replacing " further demonstrated" with "further supported these findings."
3. Line 13: ....rainfall magnitude and timing, what about their metrics?
Reply: Rainfall magnitude and timing were evaluated by comparing the simulated cumulative rainfall with the rain gauge observations within the basin.
This has now been clarified in the manuscript. The revised line reads: "LES experiments further supported these findings that, under low-CCN conditions, cumulative rainfall magnitude and timing showed better agreement with rain gauge observations within the basin, inflow biases were reduced, and inundation extent estimates were enhanced."
4. Line 29: "... establishes quantitative aerosol microphysical plausibility ..."
"Plausibility" is not something typically quantified.
Instead:
- provides quantitative evidence for aerosol microphysical impacts, or
- provides quantitative support for aerosol microphysical effects, or
- quantifies the aerosol microphysical response
which sound much more natural.
Reply: We have revised the text accordingly by replacing “establishes quantitative aerosol microphysical plausibility” with “..provides quantitative evidence for aerosol microphysical impacts on urban-scale flooding”
5. Line 32: in EPF instead on EPF
Reply: We have revised the text accordingly as suggested by the reviewer
6. I suggest replacing "prediction" with "forecasting" in the title. "Forecasting" is the more commonly used term in meteorological and hydrological applications and better reflects the operational implications discussed in the manuscript, particularly regarding improved rainfall forecasts and urban flood management. As the study is based on a single hindcast of the 2015 Chennai flood event, future work should extend the analysis to other major Indian flood events (e.g., the 2005 Mumbai flood) to assess the robustness and generalizability of the proposed forecasting framework. This phrase can be included in the Conclusions, as well.
Reply: We thank the reviewer for the valuable suggestion. Accordingly, "Prediction" has been replaced with "Forecasting" to better reflect the operational focus of the study.
The Discussion and Implications section has been replaced with a dedicated Conclusions section (Section 4), which emphasizes the key findings and their relevance for urban flood modeling and reservoir management. As suggested, the Conclusions also highlight the need to evaluate the proposed framework across other major Indian flood events to assess its robustness.
Citation: https://doi.org/10.5194/egusphere-2026-1218-AC1
-
AC1: 'Reply on RC1', Oscar Paul, 01 Aug 2026
-
RC2: 'Comment on egusphere-2026-1218', Anonymous Referee #2, 21 Jul 2026
This paper is about a single extreme precipitation event in a small area and its sensitivity to aerosols being represented
in the microphysics. It makes a convincing case that a justified reduction over climatological
aerosols provides a better simulation. Different reduction amounts and an ensemble of initial
conditions seem to provide some robustness to the results which are interpreted physically too.They were also able to demonstrate improvement to inputs for a hydrological model using LES resolutions of 200 m.
The paper is easily understood and close to acceptable as is. I will list some minor points below.
Minor Points
1. p3, line 26. Which of the WRF LES options was used?
2. p7, line 13. Maybe be 3a should be 2g?
3. p10, line 9. Define cold pool intensity.
4. p10, line 12. I think references to Figures 3j and 3k are reversed. Please check.
5. p10, lines 32-33. Looking at Figure 4b-c, I don't know how these percentages are defined. In 4b
the lines are very close to the dashed line, so how is it a 14% error?Citation: https://doi.org/10.5194/egusphere-2026-1218-RC2 -
AC2: 'Reply on RC2', Oscar Paul, 01 Aug 2026
We thank the Reviewer for the positive assessment of our work and for the helpful comments and suggestions. Our point-by-point responses are provided below.
1. p3, line 26. Which of the WRF LES options was used?
Reply: We appreciate the reviewer’s suggestion. The WRF-LES configuration for the innermost domain has been clarified in the revised manuscript. Additional details regarding the treatment of subgrid-scale turbulence and numerical diffusion have been included to improve the description of the LES setup.
The revised section is as follows: "However, the innermost domain of 0.2 km is configured in the large-eddy simulation (LES) mode, without any PBL parameterization scheme. In this LES configuration, subgrid-scale turbulent motions are represented using the 1.5-order turbulent kinetic energy (TKE) closure, while sixth-order monotonic horizontal diffusion is applied to suppress grid-scale numerical noise".
2. p7, line 13. Maybe 3a should be 2g?
Reply: Figure 2g and Figure 3a serve different purposes. Figure 2g compares simulated rainfall from all three experiments with rain gauge observations to evaluate the temporal rainfall pattern within the Adyar basin. In contrast, Figure 3a illustrates the latitude-time rainfall pattern based on IMERG data over and around the Adyar basin, which is compared with the simulated rainfall to examine aerosol-induced changes in storm morphology and cloud microphysical characteristics. Although both figures exhibit bimodal rainfall patterns, they address different aspects of the analysis. Therefore, Figure 3a has been retained.
3. p10, line 9. Define cold pool intensity.
Reply: We have revised the text to clarify that cold pool intensity is represented by the colored shading, which denotes the cold-pool regions in Figure 3h.
The revised text is as follows: "At 12 UTC, the spatial distribution of cold pool intensity (colored shading indicates cold pool regions; Fig. 3h) shows a domain-wide band of strong cold pool signatures between 79.6-79.8°E in LCCN100-Ens, whereas CNTRL-Ens concentrates stronger cold pools farther north".
4. p10, line 12. I think references to Figures 3j and 3k are reversed. Please check.
Reply: We thank the reviewer for pointing this out. The references to Figures 3j and 3k were inadvertently reversed and have been corrected in the revised manuscript.
5. p10, lines 32-33. Looking at Figure 4b-c, I don't know how these percentages are defined. In 4b the lines are very close to the dashed line, so how is it a 14% error?
Reply: We thank the reviewer for identifying this error. We rechecked and corrected the cumulative inflow error calculations. The corrected errors for the cb reservoir are 0.18% and 2.40% for the LCCN10-LES and LCCN100-LES simulations, respectively. The manuscript has been revised, and all values have been verified and corrected where necessary.
Citation: https://doi.org/10.5194/egusphere-2026-1218-AC2
-
AC2: 'Reply on RC2', Oscar Paul, 01 Aug 2026
Peer review completion
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2026-1218', Mirela-Adriana Anghelache, 21 Jul 2026
This study investigates how aerosol concentrations, represented by cloud condensation nuclei (CCN), influence the simulation of an extreme precipitation and flood event over Chennai on 1 December 2015. Using coupled atmospheric, hydrologic, and hydraulic models with ensemble and large-eddy simulations, it shows that reduced CCN concentrations enhance warm-rain processes, improve rainfall timing and magnitude, and lead to more accurate flood predictions. The results highlight the importance of representing aerosol–cloud interactions in extreme precipitation forecasting and their potential value for improving urban flood management and reservoir operation decisions. Overall, the manuscript presents a valuable contribution to the field and is recommended for publication after several revisions to improve the clarity of some arguments and moderate a few claims. The manuscript should definitely include a Conclusions section.
At Chapter 4 - Discussion and Implications:
Cold pools can indeed strengthen with more precipitation—but you should explain the mechanism more clearly. For example: Enhanced warm-rain production increased precipitation reaching the lower tropospehere, where evaporatio of falling raindrops strengthebed cold pools ... just in order the explanations shouldn't be too abrupt.
Luine 12: LES experiments further supported these findings instead of LES experiments further demonstrated; demonstrated is quite strong.
Line 13: ....rainfall magnitude and timing, what abut their metrics?
Line 29: "... establishes quantitative aerosol microphysical plausibility ..."
"Plausibility" is not something typically quantified.
Instead:
- provides quantitative evidence for aerosol microphysical impacts, or
- provides quantitative support for aerosol microphysical efects, or
- quantifies the aerosol microphysical response
which sound much more natural.
Line 32: in EPF instead on EPF
I suggest replacing "prediction" with "forecasting" in the title. "Forecasting" is the more commonly used term in meteorological and hydrological applications and better reflects the operational implications discussed in the manuscript, particularly regarding improved rainfall forecasts and urban flood management.
As the study is based on a single hindcast of the 2015 Chennai flood event, future work should extend the analysis to other major Indian flood events (e.g., the 2005 Mumbai flood) to assess the robustness and generalizability of the proposed forecasting framework. This phrase can be included in the Conclusions, as well.
Citation: https://doi.org/10.5194/egusphere-2026-1218-RC1 -
AC1: 'Reply on RC1', Oscar Paul, 01 Aug 2026
We sincerely thank the Reviewer for the thorough evaluation of our manuscript and for the constructive comments and valuable suggestions. Our point-by-point responses are provided below.
1. Cold pools can indeed strengthen with more precipitation—but you should explain the mechanism more clearly. For example: Enhanced warm-rain production increased precipitation reaching the lower troposphere, where evaporation of falling raindrops strengthened cold pools ... just in order the explanations shouldn't be too abrupt.
Reply: We thank the reviewer for the helpful suggestion. We have revised the text to clearly explain the physical sequence linking warm-rain processes and cold pool intensification. Specifically, we now state that enhanced warm-rain production due to low CCN increases precipitation where evaporation of falling raindrops strengthens cold pools, which subsequently enhance low-level moisture convergence and secondary convection.
The revised section in the manuscript is as follows: “Lower CCN concentrations favoured warm-rain processes and increased the amount of precipitation. The subsequent evaporation of falling raindrops enhanced evaporative cooling, thereby strengthening cold pool activity near the basin (Mallinson and Lasher-Trapp, 2019). These persistent cold pools intensified low-level moisture convergence through thermodynamic forcing, establishing favourable conditions for secondary convection (Schlemmer and Hohenegger, 2014; Schlemmer and Hohenegger, 2016), thereby simulating the bimodal rainfall peaks observed in the basin.”
2. Line 12: LES experiments further supported these findings instead of LES experiments further demonstrated; demonstrated is quite strong.
Reply: We agree with the reviewer's suggestion and have revised the text accordingly by replacing " further demonstrated" with "further supported these findings."
3. Line 13: ....rainfall magnitude and timing, what about their metrics?
Reply: Rainfall magnitude and timing were evaluated by comparing the simulated cumulative rainfall with the rain gauge observations within the basin.
This has now been clarified in the manuscript. The revised line reads: "LES experiments further supported these findings that, under low-CCN conditions, cumulative rainfall magnitude and timing showed better agreement with rain gauge observations within the basin, inflow biases were reduced, and inundation extent estimates were enhanced."
4. Line 29: "... establishes quantitative aerosol microphysical plausibility ..."
"Plausibility" is not something typically quantified.
Instead:
- provides quantitative evidence for aerosol microphysical impacts, or
- provides quantitative support for aerosol microphysical effects, or
- quantifies the aerosol microphysical response
which sound much more natural.
Reply: We have revised the text accordingly by replacing “establishes quantitative aerosol microphysical plausibility” with “..provides quantitative evidence for aerosol microphysical impacts on urban-scale flooding”
5. Line 32: in EPF instead on EPF
Reply: We have revised the text accordingly as suggested by the reviewer
6. I suggest replacing "prediction" with "forecasting" in the title. "Forecasting" is the more commonly used term in meteorological and hydrological applications and better reflects the operational implications discussed in the manuscript, particularly regarding improved rainfall forecasts and urban flood management. As the study is based on a single hindcast of the 2015 Chennai flood event, future work should extend the analysis to other major Indian flood events (e.g., the 2005 Mumbai flood) to assess the robustness and generalizability of the proposed forecasting framework. This phrase can be included in the Conclusions, as well.
Reply: We thank the reviewer for the valuable suggestion. Accordingly, "Prediction" has been replaced with "Forecasting" to better reflect the operational focus of the study.
The Discussion and Implications section has been replaced with a dedicated Conclusions section (Section 4), which emphasizes the key findings and their relevance for urban flood modeling and reservoir management. As suggested, the Conclusions also highlight the need to evaluate the proposed framework across other major Indian flood events to assess its robustness.
Citation: https://doi.org/10.5194/egusphere-2026-1218-AC1
-
AC1: 'Reply on RC1', Oscar Paul, 01 Aug 2026
-
RC2: 'Comment on egusphere-2026-1218', Anonymous Referee #2, 21 Jul 2026
This paper is about a single extreme precipitation event in a small area and its sensitivity to aerosols being represented
in the microphysics. It makes a convincing case that a justified reduction over climatological
aerosols provides a better simulation. Different reduction amounts and an ensemble of initial
conditions seem to provide some robustness to the results which are interpreted physically too.They were also able to demonstrate improvement to inputs for a hydrological model using LES resolutions of 200 m.
The paper is easily understood and close to acceptable as is. I will list some minor points below.
Minor Points
1. p3, line 26. Which of the WRF LES options was used?
2. p7, line 13. Maybe be 3a should be 2g?
3. p10, line 9. Define cold pool intensity.
4. p10, line 12. I think references to Figures 3j and 3k are reversed. Please check.
5. p10, lines 32-33. Looking at Figure 4b-c, I don't know how these percentages are defined. In 4b
the lines are very close to the dashed line, so how is it a 14% error?Citation: https://doi.org/10.5194/egusphere-2026-1218-RC2 -
AC2: 'Reply on RC2', Oscar Paul, 01 Aug 2026
We thank the Reviewer for the positive assessment of our work and for the helpful comments and suggestions. Our point-by-point responses are provided below.
1. p3, line 26. Which of the WRF LES options was used?
Reply: We appreciate the reviewer’s suggestion. The WRF-LES configuration for the innermost domain has been clarified in the revised manuscript. Additional details regarding the treatment of subgrid-scale turbulence and numerical diffusion have been included to improve the description of the LES setup.
The revised section is as follows: "However, the innermost domain of 0.2 km is configured in the large-eddy simulation (LES) mode, without any PBL parameterization scheme. In this LES configuration, subgrid-scale turbulent motions are represented using the 1.5-order turbulent kinetic energy (TKE) closure, while sixth-order monotonic horizontal diffusion is applied to suppress grid-scale numerical noise".
2. p7, line 13. Maybe 3a should be 2g?
Reply: Figure 2g and Figure 3a serve different purposes. Figure 2g compares simulated rainfall from all three experiments with rain gauge observations to evaluate the temporal rainfall pattern within the Adyar basin. In contrast, Figure 3a illustrates the latitude-time rainfall pattern based on IMERG data over and around the Adyar basin, which is compared with the simulated rainfall to examine aerosol-induced changes in storm morphology and cloud microphysical characteristics. Although both figures exhibit bimodal rainfall patterns, they address different aspects of the analysis. Therefore, Figure 3a has been retained.
3. p10, line 9. Define cold pool intensity.
Reply: We have revised the text to clarify that cold pool intensity is represented by the colored shading, which denotes the cold-pool regions in Figure 3h.
The revised text is as follows: "At 12 UTC, the spatial distribution of cold pool intensity (colored shading indicates cold pool regions; Fig. 3h) shows a domain-wide band of strong cold pool signatures between 79.6-79.8°E in LCCN100-Ens, whereas CNTRL-Ens concentrates stronger cold pools farther north".
4. p10, line 12. I think references to Figures 3j and 3k are reversed. Please check.
Reply: We thank the reviewer for pointing this out. The references to Figures 3j and 3k were inadvertently reversed and have been corrected in the revised manuscript.
5. p10, lines 32-33. Looking at Figure 4b-c, I don't know how these percentages are defined. In 4b the lines are very close to the dashed line, so how is it a 14% error?
Reply: We thank the reviewer for identifying this error. We rechecked and corrected the cumulative inflow error calculations. The corrected errors for the cb reservoir are 0.18% and 2.40% for the LCCN10-LES and LCCN100-LES simulations, respectively. The manuscript has been revised, and all values have been verified and corrected where necessary.
Citation: https://doi.org/10.5194/egusphere-2026-1218-AC2
-
AC2: 'Reply on RC2', Oscar Paul, 01 Aug 2026
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 462 | 116 | 35 | 613 | 83 | 27 | 35 |
- HTML: 462
- PDF: 116
- XML: 35
- Total: 613
- Supplement: 83
- BibTeX: 27
- EndNote: 35
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Oscar Paul
N. Nithila Devi
Rakesh Teja Konduru
Soumendra Nath Kuiry
Kundan Lal Shrestha
Chandan Sarangi
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1648 KB) - Metadata XML
-
Supplement
(5750 KB) - BibTeX
- EndNote
- Final revised paper
This study investigates how aerosol concentrations, represented by cloud condensation nuclei (CCN), influence the simulation of an extreme precipitation and flood event over Chennai on 1 December 2015. Using coupled atmospheric, hydrologic, and hydraulic models with ensemble and large-eddy simulations, it shows that reduced CCN concentrations enhance warm-rain processes, improve rainfall timing and magnitude, and lead to more accurate flood predictions. The results highlight the importance of representing aerosol–cloud interactions in extreme precipitation forecasting and their potential value for improving urban flood management and reservoir operation decisions. Overall, the manuscript presents a valuable contribution to the field and is recommended for publication after several revisions to improve the clarity of some arguments and moderate a few claims. The manuscript should definitely include a Conclusions section.
At Chapter 4 - Discussion and Implications:
Cold pools can indeed strengthen with more precipitation—but you should explain the mechanism more clearly. For example: Enhanced warm-rain production increased precipitation reaching the lower tropospehere, where evaporatio of falling raindrops strengthebed cold pools ... just in order the explanations shouldn't be too abrupt.
Luine 12: LES experiments further supported these findings instead of LES experiments further demonstrated; demonstrated is quite strong.
Line 13: ....rainfall magnitude and timing, what abut their metrics?
Line 29: "... establishes quantitative aerosol microphysical plausibility ..."
"Plausibility" is not something typically quantified.
Instead:
- provides quantitative evidence for aerosol microphysical impacts, or
- provides quantitative support for aerosol microphysical efects, or
- quantifies the aerosol microphysical response
which sound much more natural.
Line 32: in EPF instead on EPF
I suggest replacing "prediction" with "forecasting" in the title. "Forecasting" is the more commonly used term in meteorological and hydrological applications and better reflects the operational implications discussed in the manuscript, particularly regarding improved rainfall forecasts and urban flood management.
As the study is based on a single hindcast of the 2015 Chennai flood event, future work should extend the analysis to other major Indian flood events (e.g., the 2005 Mumbai flood) to assess the robustness and generalizability of the proposed forecasting framework. This phrase can be included in the Conclusions, as well.