the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Feb 2025
Implementation of a dry deposition module (DEPAC v3.11) in a large eddy simulation code (DALES v4.4)
Abstract. High-resolution data on reactive nitrogen deposition are needed to inform cost-effective policies. Large eddy simulation models coupled to a dry deposition module present a valuable tool for obtaining these high-resolution data. In this paper we describe the implementation of a dry deposition module (DEPAC v3.11) into a large eddy simulation code (DALES v4.4), and its first application in a real-world case study. With this coupled model, we are able to represent the turbulent surface-atmosphere exchange of passive and reactive tracers at the hectometer resolution. A land surface module was implemented to solve the surface energy budget and provide detailed information for the calculation of deposition fluxes per land use (LU) class. Both the land surface model and the dry deposition module are extensively described, as well as the inputs that are needed to run them.
To show the advantages of this new modeling approach, we present a case study for the city of Eindhoven in the Netherlands, focusing on the emission, dispersion and deposition of NOx and NH3. We find that DALES is able to reproduce the main features of the boundary layer development and the diurnal cycle of local meteorology well, with the exception of the evening transition. DALES calculates the dispersion and deposition of NOx and NH3 in great spatial detail, clearly showing the influence of local LU patterns on small-scale transport, removal efficiencies and mixing characteristics.
-
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
(2197 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2197 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-426', Ivo Suter, 01 Apr 2025
-
AC2: 'Reply on RC1', Ruud Janssen, 30 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-426/egusphere-2025-426-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Ruud Janssen, 30 May 2025
-
RC2: 'Comment on egusphere-2025-426', Anonymous Referee #2, 10 Apr 2025
This study implemented a deposition module into a large-eddy simulation code and presented a case study on the dispersion and deposition of NOx and NH₃. However, due to the lack of model sensitivity analysis, it remains unclear whether the implemented model accurately reproduces gas-phase deposition. Below are several major concerns:
Before applying the developed deposition module within DALES, a theoretical analysis should be conducted to examine the relationship between deposition velocity and the dry deposition parameters of each gas species in DEPAC. These relationships should then be compared with existing literature to validate the correctness and reliability of the DEPAC implementation.
The condensation and evaporation processes of HNO₃ and NH₃ can have effects comparable to their dry deposition, and different deposition schemes may even reverse the gas-particle partitioning (Lin et al., 2024). However, aerosol dynamics are not considered in this study, which introduces substantial uncertainty into the case study results. The exclusion of this key mechanism undermines confidence in the conclusions drawn from the simulations.
The spatial and temporal distributions of Ra, Rb, Rc, deposition velocity, and deposition flux during the simulation period should be provided and discussed. This would allow for a more comprehensive understanding of the deposition process. Furthermore, it is important to clarify which of the resistances (Ra, Rb, or Rc) is the dominant factor under the simulated conditions.
Minor comments:
Line 196:’ In addition, a sensitivity analysis in a related project pointed out that the effect of switching between dry and wet land on the deposition fluxes of NH3 is not very strong.’
Please provide a citation or additional evidence to support this statement. Without substantiation, the conclusion may appear speculative.
Reference
Lin, C., Ooka, R., Kikumoto, H., Kim, Y., Zhang, Y., Flageul, C., & Sartelet, K. (2024). Impact of gas dry deposition parameterization on secondary particle formation in an urban canyon. Atmospheric Environment, 333, 120633.
Citation: https://doi.org/10.5194/egusphere-2025-426-RC2 -
AC1: 'Reply on RC2', Ruud Janssen, 30 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-426/egusphere-2025-426-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Ruud Janssen, 30 May 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-426', Ivo Suter, 01 Apr 2025
The authors describe the extension of an atmospheric LES (DALES) with a dry deposition module (DEPAC), with the aim to better spatially resolve deposition of ammonia and NO2. The manuscript presents the development of a suitable tool to study deposition on smaller scale, relevant for very heterogeneous terrain and areas of particular interest, such as water conservation or nature preservation zones. As such, the presented work clearly is of sufficient interest to warrant publication in geoscientific model development.
However, while the methods are described theoretically, their actual implementation into the model remain somewhat unclear from the manuscript. The information can be obtained from the provided source code, of course. But this seems unnecessarily cumbersome. This is particularly true, since DEPAC is already described in the referenced literature. The authors should mention some aspects of implementation that are helpful to readers who might want to attempt a similar undertaking (e.g. are the deposition rates added to the tendencies? How is R=infinity realised? Are molecular or sub-grid diffusivities being used to calculate Rb).
The authors could not convince me that the modelled boundary layer height after sunset is not a problem. I encourage them to attempt to improve this (could a different sub-grid model or forcing help?). In the current state the inclusion of midnight (00:00 LT) into the analysis is somewhat moot.
There are several inaccuracies and inconsistencies about units, variable names etc. and several plots can be improved. These have been pointed out as individual comments.
I think the copyright statement in le_drydepos_gas_depac.f90 does not match GPLv3.
Lastly, a model validation against measurements would contribute significantly to the article. The authors make clear, that the model in the current state does not contain a chemistry module and is still under development, which unfortunately makes this task currently difficult. Maybe a weak comparison to NOx measurements is already possible?
I strongly encourage a proper validation as soon as possible.
Other comments:
P3L76 Why is only the OH marked as a radical?
P3L82 You can remove the “, the Netherlands,” after “Eindhoven
P3L84 Can you elaborate quickly why the water is a problem. Is it because of the double periodic boundary, so you assume opposite sides to be of similar nature? Or more specifically a problem to your case study with DEPAC?
P4L93 Missing reference
P5L145 Missing reference
P5L149/150 I think you should at least introduce the units of V, G and R once. Probably here. Units are emitted quite often throughout the manuscript. Please add them where useful (e.g. also when you introduce χ)
P6Fig1 Rext is called Rw throughout the manuscript. In the schematic it shows that Rinc depends on LAI, whereas in Eq. 12 it depends on SAI
P6L163 Can you explain why Rb is small when the quasi-laminar layer is small. It is not immediately clear from equation (3).
P7L1 Can you mention how these quantities are actually being calculated within DALES? Are you using the literature values of air etc or the sub-grid quantities from the LES?
P7L179 “Absorbed“ on the surface is a somewhat strange expression. Maybe use “canopy” instead of “surface” in this paragraph? Or do you mean “adsorbed on the surface”?
P7Eq5 Are the signs correct, the deposition flux is negative when deposition occurs? (xatm>xcomp; Vd is always ≥0)
P8L2 Can you give more information about this “related project”?
P8L207 line starts with “,”
P8L208 can you elaborate why phenology is negligible
P8L208 In this study it might not be relevant, but what about another study? Are the correction factors implemented but not used here, or are they not implemented?
P8L209 Please point at section 2.3.2 where you describe the seasonal dependence, i.e. how LAI affects Rinc,
P8Eq8 Suddenly you indicate that Gamma is a function of the surface temperature, but in other equations you don’t (e.g. Eq 7 or Eq 10). Please be consistent.
P8L216 what does “long term” mean?
P8L219 missing space between 2 and “s m-1”
P9L225 Which value is assumed for Rw
P9L226 I would rephrase and reference Eq. 7 “..calculated for external surfaces analogue to Eq 7 by substituting GammaW for GammaS”
P9L228 is the compensation point at 4m not “long term” anymore, what is it then?
P9Eq12 case 2: Can u* be negative?
P9L239 the formula is unclear, what are aSAI and bSAI, why is there a bracket around +bSAI? Please clarify.
P9L245 “Equation 7” should be “Eq. 7”
P9L246 I cannot find the compensation point or Gamma in Table A1
P10L255 Are these all from ERA5?
P10L255 what does “partly” mean, how are the other forcings implemented?
P10L256 The “surface energy balance calculation”
P10L265 It is not clear at this point that the 1 km x 1 km is the resolution of the emission inventory and 50 m x 50 m refers to the grid size in DALES
P10L266 move the sentence about the 12 fraction maps to line 279, after the fraction maps are actually described. “for several other source type (CBS, 2023). This results in a total of 12 fraction maps”
P10L280 “emission heights”
P10L281 how high is “ground level”?
P11L284 I don’t understand the argument about why road emissions are smaller than highway emissions? Wouldn’t we expect them to be smaller, what’s the ratio in the initial emission inventory? What is the problem with the downscaling? Does it affect other emission categories? Can you please elaborate on this point
P11L293 the large-scale wind direction was easterly? Because Figure 4 shows the local winds to be north/north-easterly
P11L297 The formula or numbers are wrong: 20 m*1.009^127 is only 62 m and not 95 m. Also, the cumulative sum is not 8.5 km.
P12L320 It seems ERA5 has the dimensions swapped? Should it be 28 km x 17 km?
P12L324 Remove the “since”, the STD is not solely a consequence of the rounding.
P12L325 replace “quantized” with “rounded”. “The wind speed is rounded down to the next integer value and the wind direction is rounded down to the next full degree.”
P12L319 at which height are these quantities in DALES and ERA5?
P12L325 “standard deviation of the measurement errors”. I don’t understand what exactly you mean by this. The measurement error? The error introduced by rounding? The standard deviation between the model and the measurement?
P12L327 °C or K instead of “degree”
P12L327 It’s the other way round, the model has a negative bias during the day
P12L327 Does DALES not offer a parameterised 2m-temperature? DALES follows quite closely the ERA5, how do they compare in terms of height/layer thickness?
P13Fig4 When comparing model to measurement, you should probably extract the value at the measurement location or interpolate and not use domain averages.
P13Fig4 WDIR How can there be a negative value, is the axis not from 0° to 360°?
P13Fig4 WDIR: It might help to change the y-axis to go from -180 to +180 degrees, then the artificial “peaks” would disappear. Since the range of wind directions is actually quite limited -90 to +90 would probably also suffice.
P14L345 Did you do the simulations at 25 m resolution? Is the set-up identical otherwise? I think a comparison to the 50 m run would be interesting (in terms of deposition). You could have tried a resolution of 10 m without using the emission downscaling to determine if that indeed is appropriate to simulate the evening transition. Why is the sub-grid model not able to compensate?
P14L347 Δθ/Δz ?
P14L349 You say the BLH of 0 m is due to the choice of criterion, but most other metrics also show 0 m. So is the BLH at night just wrong?
P14L357 Reference to the figure missing.
P14L357 Emissions at 06:00 LT
P14L365 Where are the time-profiles from? Reference
P14L368 “The resulting vertical profile of mean NOx concentration over the domain (Figure 7a) shows a trend similar to the boundary layer development (Figure 5d / Figure C1?).”
P14L369 The NOx peak near the surface is at midnight according to Fig 7a. The concentration at 06:00 are the lowest ones, even though one would expect the maximum at the morning rush hour? This does not match your description. How do you explain the huge concentrations at mid-night, is this realistic?
P15Fig5 Vertical profile of domain averaged …
P15Fig5 Please adopt the x-axis of (a) and (c) to better reflect the range of the data
P16Fig6 Are the time profiles domain averaged? You use “CEST” here but “LT” everywhere else.
P16L381 domain averaged concentrations at 10m
P16L382 What you mean is that during the entire day the emissions and the upward mixing are in a balance until sunset, leading to a relatively constant near-ground concentration? Only slightly elevated above the background of 1ppb. Is this realistic? Again, how do you explain the huge concentrations at night?
P19L1 Is the NOx-background the same on all vertical levels?
References: “doi” is sufficient and preferred over “url”. “Publisher” is not necessary, “_eprint”, some websites lack the date when they were accessed.
Citation: https://doi.org/10.5194/egusphere-2025-426-RC1 -
AC2: 'Reply on RC1', Ruud Janssen, 30 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-426/egusphere-2025-426-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Ruud Janssen, 30 May 2025
-
RC2: 'Comment on egusphere-2025-426', Anonymous Referee #2, 10 Apr 2025
This study implemented a deposition module into a large-eddy simulation code and presented a case study on the dispersion and deposition of NOx and NH₃. However, due to the lack of model sensitivity analysis, it remains unclear whether the implemented model accurately reproduces gas-phase deposition. Below are several major concerns:
Before applying the developed deposition module within DALES, a theoretical analysis should be conducted to examine the relationship between deposition velocity and the dry deposition parameters of each gas species in DEPAC. These relationships should then be compared with existing literature to validate the correctness and reliability of the DEPAC implementation.
The condensation and evaporation processes of HNO₃ and NH₃ can have effects comparable to their dry deposition, and different deposition schemes may even reverse the gas-particle partitioning (Lin et al., 2024). However, aerosol dynamics are not considered in this study, which introduces substantial uncertainty into the case study results. The exclusion of this key mechanism undermines confidence in the conclusions drawn from the simulations.
The spatial and temporal distributions of Ra, Rb, Rc, deposition velocity, and deposition flux during the simulation period should be provided and discussed. This would allow for a more comprehensive understanding of the deposition process. Furthermore, it is important to clarify which of the resistances (Ra, Rb, or Rc) is the dominant factor under the simulated conditions.
Minor comments:
Line 196:’ In addition, a sensitivity analysis in a related project pointed out that the effect of switching between dry and wet land on the deposition fluxes of NH3 is not very strong.’
Please provide a citation or additional evidence to support this statement. Without substantiation, the conclusion may appear speculative.
Reference
Lin, C., Ooka, R., Kikumoto, H., Kim, Y., Zhang, Y., Flageul, C., & Sartelet, K. (2024). Impact of gas dry deposition parameterization on secondary particle formation in an urban canyon. Atmospheric Environment, 333, 120633.
Citation: https://doi.org/10.5194/egusphere-2025-426-RC2 -
AC1: 'Reply on RC2', Ruud Janssen, 30 May 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-426/egusphere-2025-426-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Ruud Janssen, 30 May 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 729 | 66 | 26 | 821 | 37 | 48 |
- HTML: 729
- PDF: 66
- XML: 26
- Total: 821
- BibTeX: 37
- EndNote: 48
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Leon Geers
Gudrun Thorkelsdottir
Jordi Vilà-Guerau de Arellano
Martijn Schaap
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2197 KB) - Metadata XML
The authors describe the extension of an atmospheric LES (DALES) with a dry deposition module (DEPAC), with the aim to better spatially resolve deposition of ammonia and NO2. The manuscript presents the development of a suitable tool to study deposition on smaller scale, relevant for very heterogeneous terrain and areas of particular interest, such as water conservation or nature preservation zones. As such, the presented work clearly is of sufficient interest to warrant publication in geoscientific model development.
However, while the methods are described theoretically, their actual implementation into the model remain somewhat unclear from the manuscript. The information can be obtained from the provided source code, of course. But this seems unnecessarily cumbersome. This is particularly true, since DEPAC is already described in the referenced literature. The authors should mention some aspects of implementation that are helpful to readers who might want to attempt a similar undertaking (e.g. are the deposition rates added to the tendencies? How is R=infinity realised? Are molecular or sub-grid diffusivities being used to calculate Rb).
The authors could not convince me that the modelled boundary layer height after sunset is not a problem. I encourage them to attempt to improve this (could a different sub-grid model or forcing help?). In the current state the inclusion of midnight (00:00 LT) into the analysis is somewhat moot.
There are several inaccuracies and inconsistencies about units, variable names etc. and several plots can be improved. These have been pointed out as individual comments.
I think the copyright statement in le_drydepos_gas_depac.f90 does not match GPLv3.
Lastly, a model validation against measurements would contribute significantly to the article. The authors make clear, that the model in the current state does not contain a chemistry module and is still under development, which unfortunately makes this task currently difficult. Maybe a weak comparison to NOx measurements is already possible?
I strongly encourage a proper validation as soon as possible.
Other comments:
P3L76 Why is only the OH marked as a radical?
P3L82 You can remove the “, the Netherlands,” after “Eindhoven
P3L84 Can you elaborate quickly why the water is a problem. Is it because of the double periodic boundary, so you assume opposite sides to be of similar nature? Or more specifically a problem to your case study with DEPAC?
P4L93 Missing reference
P5L145 Missing reference
P5L149/150 I think you should at least introduce the units of V, G and R once. Probably here. Units are emitted quite often throughout the manuscript. Please add them where useful (e.g. also when you introduce χ)
P6Fig1 Rext is called Rw throughout the manuscript. In the schematic it shows that Rinc depends on LAI, whereas in Eq. 12 it depends on SAI
P6L163 Can you explain why Rb is small when the quasi-laminar layer is small. It is not immediately clear from equation (3).
P7L1 Can you mention how these quantities are actually being calculated within DALES? Are you using the literature values of air etc or the sub-grid quantities from the LES?
P7L179 “Absorbed“ on the surface is a somewhat strange expression. Maybe use “canopy” instead of “surface” in this paragraph? Or do you mean “adsorbed on the surface”?
P7Eq5 Are the signs correct, the deposition flux is negative when deposition occurs? (xatm>xcomp; Vd is always ≥0)
P8L2 Can you give more information about this “related project”?
P8L207 line starts with “,”
P8L208 can you elaborate why phenology is negligible
P8L208 In this study it might not be relevant, but what about another study? Are the correction factors implemented but not used here, or are they not implemented?
P8L209 Please point at section 2.3.2 where you describe the seasonal dependence, i.e. how LAI affects Rinc,
P8Eq8 Suddenly you indicate that Gamma is a function of the surface temperature, but in other equations you don’t (e.g. Eq 7 or Eq 10). Please be consistent.
P8L216 what does “long term” mean?
P8L219 missing space between 2 and “s m-1”
P9L225 Which value is assumed for Rw
P9L226 I would rephrase and reference Eq. 7 “..calculated for external surfaces analogue to Eq 7 by substituting GammaW for GammaS”
P9L228 is the compensation point at 4m not “long term” anymore, what is it then?
P9Eq12 case 2: Can u* be negative?
P9L239 the formula is unclear, what are aSAI and bSAI, why is there a bracket around +bSAI? Please clarify.
P9L245 “Equation 7” should be “Eq. 7”
P9L246 I cannot find the compensation point or Gamma in Table A1
P10L255 Are these all from ERA5?
P10L255 what does “partly” mean, how are the other forcings implemented?
P10L256 The “surface energy balance calculation”
P10L265 It is not clear at this point that the 1 km x 1 km is the resolution of the emission inventory and 50 m x 50 m refers to the grid size in DALES
P10L266 move the sentence about the 12 fraction maps to line 279, after the fraction maps are actually described. “for several other source type (CBS, 2023). This results in a total of 12 fraction maps”
P10L280 “emission heights”
P10L281 how high is “ground level”?
P11L284 I don’t understand the argument about why road emissions are smaller than highway emissions? Wouldn’t we expect them to be smaller, what’s the ratio in the initial emission inventory? What is the problem with the downscaling? Does it affect other emission categories? Can you please elaborate on this point
P11L293 the large-scale wind direction was easterly? Because Figure 4 shows the local winds to be north/north-easterly
P11L297 The formula or numbers are wrong: 20 m*1.009^127 is only 62 m and not 95 m. Also, the cumulative sum is not 8.5 km.
P12L320 It seems ERA5 has the dimensions swapped? Should it be 28 km x 17 km?
P12L324 Remove the “since”, the STD is not solely a consequence of the rounding.
P12L325 replace “quantized” with “rounded”. “The wind speed is rounded down to the next integer value and the wind direction is rounded down to the next full degree.”
P12L319 at which height are these quantities in DALES and ERA5?
P12L325 “standard deviation of the measurement errors”. I don’t understand what exactly you mean by this. The measurement error? The error introduced by rounding? The standard deviation between the model and the measurement?
P12L327 °C or K instead of “degree”
P12L327 It’s the other way round, the model has a negative bias during the day
P12L327 Does DALES not offer a parameterised 2m-temperature? DALES follows quite closely the ERA5, how do they compare in terms of height/layer thickness?
P13Fig4 When comparing model to measurement, you should probably extract the value at the measurement location or interpolate and not use domain averages.
P13Fig4 WDIR How can there be a negative value, is the axis not from 0° to 360°?
P13Fig4 WDIR: It might help to change the y-axis to go from -180 to +180 degrees, then the artificial “peaks” would disappear. Since the range of wind directions is actually quite limited -90 to +90 would probably also suffice.
P14L345 Did you do the simulations at 25 m resolution? Is the set-up identical otherwise? I think a comparison to the 50 m run would be interesting (in terms of deposition). You could have tried a resolution of 10 m without using the emission downscaling to determine if that indeed is appropriate to simulate the evening transition. Why is the sub-grid model not able to compensate?
P14L347 Δθ/Δz ?
P14L349 You say the BLH of 0 m is due to the choice of criterion, but most other metrics also show 0 m. So is the BLH at night just wrong?
P14L357 Reference to the figure missing.
P14L357 Emissions at 06:00 LT
P14L365 Where are the time-profiles from? Reference
P14L368 “The resulting vertical profile of mean NOx concentration over the domain (Figure 7a) shows a trend similar to the boundary layer development (Figure 5d / Figure C1?).”
P14L369 The NOx peak near the surface is at midnight according to Fig 7a. The concentration at 06:00 are the lowest ones, even though one would expect the maximum at the morning rush hour? This does not match your description. How do you explain the huge concentrations at mid-night, is this realistic?
P15Fig5 Vertical profile of domain averaged …
P15Fig5 Please adopt the x-axis of (a) and (c) to better reflect the range of the data
P16Fig6 Are the time profiles domain averaged? You use “CEST” here but “LT” everywhere else.
P16L381 domain averaged concentrations at 10m
P16L382 What you mean is that during the entire day the emissions and the upward mixing are in a balance until sunset, leading to a relatively constant near-ground concentration? Only slightly elevated above the background of 1ppb. Is this realistic? Again, how do you explain the huge concentrations at night?
P19L1 Is the NOx-background the same on all vertical levels?
References: “doi” is sufficient and preferred over “url”. “Publisher” is not necessary, “_eprint”, some websites lack the date when they were accessed.