the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Sep 2025
Enhancing Volcanic Eruption Simulations with the WRF-Chem v4.8
Abstract. Volcanic eruptions are one of the major natural hazards, exerting profound effects on the environment and climate. The emissions associated with such eruptions pose substantial risks to terrestrial systems and public health, particularly through the induction of acid rain and air pollution. Volcanic ash influences populations at distances reaching several thousand kilometers from the erupted volcano. Also, accurate forecasting of volcanic clouds is crucial for air traffic control. This study introduces enhancements in the simulation of volcanic eruptions and the transport of volcanic material using the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) v4.8 model. Improvements include the addition of wet and dry deposition of ash and sulfate, improved SO2 chemical transformation mechanisms, and corrections of the gravitational deposition of ash. Ash, sulfate, and SO2 mass budget analyses were conducted. Furthermore, we included the direct radiative effect of ash and sulfate aerosols. Additionally, we developed an open-source Python-written emission preprocessor called PrepEmisSources to facilitate and streamline the preparation of volcanic emissions. Accordingly, the model code was extended to simulate complex volcanic emission scenarios using the emissions file prepared using the PrepEmisSources tool. The results suggest that the enhanced WRF-Chem v4.8 code provides an accurate representation of volcanic ash, SO2, and sulfate dispersion, deposition, and chemical transformation. These improvements will aid in volcanic debris forecasting and will allow for the use of the model for assessments of volcanic aerosols on climate and for geoengineering problems, including modeling of stratospheric aerosol injection.
- Preprint
(5130 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 19 Nov 2025)
- RC1: 'Comment on egusphere-2025-4124', Mingzhao Liu, 09 Oct 2025 reply
-
RC2: 'Comment on egusphere-2025-4124', Anonymous Referee #2, 03 Nov 2025
reply
Review of GMD paper “Enhancing Volcanic Eruption Simulations with the WRF-Chem v4.8”
The paper “Enhancing Volcanic Eruption Simulations with the WRF-Chem v4.8” describes a few updates in WRF-chem to improve the volcanic ash and sulfur simulations. The author well described the changes they made, showing figures of parameters improvement after the changes, and apply the real example of Mt. Pinatubo as a test case. The paper is pretty well-rewritten and I recommend the publication after the author addresses the major comments I listed below. This is because as a GMD paper, I expect a little more clear and thorough description the model itself.
Major:
- Between section 1 previous work and section 2 code modification
I highly recommend to add a section here for model description. You described so many previous works using WRF-chem, but apparently not all the modifications are used in your current version. It is quite confusing not knowing your start point. So, I recommend you to add and additional section with the following three aspects:
- Only describe the modification cannot comprehensively for the audience to know the whole structure. You need to at least briefly describe the sulfur chemistry and microphysical processes in WRF-chem for how the sulfate forms, how does sulfate interact with dust (or there's no interactions, then you should quote some of these interactions may be important for future development for initial plume dispersion), how does the nucleation and coagulation work to convert H2SO4 to sulfate, how does the aerosols interact with the cloud.
- It's hard to know what processes you consider besides the processes you modified. For example, how are you consider the formation of ash aggregates (Brown et al., 2012), sulfate nucleation (what sulfate nucleation scheme you use?) and coagulation (is Van De Var force considered? is sulfate coagulate with ash?), removal of sulfate due to adsorption on ash particle (Zhu et al., 2020). These are all very important processes for calculating removal from atmosphere, fall velocity and sedimentation. If they are previously implemented, you need to describe them. If they are not considered in your current version, you should mention some processes are not currently considered and acknowledge their importance.
- A schematic diagram will be really helpful to show the workflow of the model, writing down what is the code before and use a different color to show what has been implemented by this work.
2. Section 2.5 The description of sulfate size distribution implementation is not easy to follow. Also, some implication is questionable:
- Section 2.5: Now, you are saying you use a size distribution for aerosol. could GOCART track the size distribution? It should just track the number and total burden, right? since it is a two-moment scheme.
- Line 168: “0.2 um” radius or diameter? Are you talking about background sulfate or volcanic sulfate aerosol?
- Line 169: “bigger”, The stratospheric background sulfate usually not bigger than 0.2 um radius. but volcanic sulfate is bigger. So again, it is a little confusing what you are talking about.
- Line 173-174: is the size distribution for volcanic sulfate too or just background sulfate? The size distribution between them is very different.
- Eqn 12: my impression is that kappa values are suitable for tropospheric aerosols but too low for the stratospheric sulfate. You should make a plot of H2SO4 weight percent inside the H2SO4-H2O binary solution as a function of altitude. The stratospheric H2SO4 weight percent generally ranges from 60-90%.
- Eqn 30: You previously listed the hygroscopicity of sulfate with kappa of 1.19 in Eqn 12 but here you list sulfate kappa of 0.5.
Minor:
Line 14: and chemical transformation-> and SO2 chemical transformation
Line 37: “In contrast, water vapor and sulfate are important in further stages of volcanic cloud dispersion.”->”In contrast, sulfate, and sometimes water vapor, are important in further stages of volcanic cloud dispersion.”
Line 50: “In the model, ash is gravitationally settled, and SO2 undergoes oxidation to sulfate using the prescribed OH vertical distribution.” After reading the whole paper, I am still not sure if you use the same prescribed OH as they are. I thought you changed the OH to be interactive, but from you description later on line 106-109, it seems to be not. So please specify if you use the same method as they are, or if you improved, what’s the difference between yours and theirs.
Line 106-109: If they are not interactive, how could you deplete OH? Are you having an OH fiend vary with time because the sun rise and sun set? But SO2 doesn't deplete OH?
Line 53-55, Line 61-64: these sentences just describe what the previous work did, but haven’t list what they found. You other descriptions in this section are good.
Line 71-73: I am curious why you use “MOSAIC” instead of “GOCART” for this research, because MOSAIC is generally for tropospheric chemistry especially near the surface, rather than upper atmosphere.
Line 87-89: This is a following question for line 71-73. Stenchikov et al., 2025 is using MOSAIC in the Hunga case, right? But you are using GOCART. Do you need to apply some extra work to add the water vapor and sulfate in since you are using different aerosol schemes?
Line 138: “large scale scavenging” Is it stratiform clouds scavenging? you should add a little explanation.
Line 186: you should describe the theories you use to calculate the fall velocity of sulfate and ash. Cite the paper.
Line 205 and 211: µu -> µm?
Line 234-235: “where air density is less by two orders of magnitude with respect to the surface” only one order of magnitude if you talk about UTLS where most volcano injected into.
Line 273-275 and the whole document where you say “MOSAIC bin”: It's really confusing to mention MOSAIC here since you are not using MOSAIC (you are using GOCART), but just have identical bin width as MOSAIC. Any model can define a bin width. I recommend to change “MOSAIC bins” to a different name.
Question: These bins don’t cover Aitken mode and large ash particles. Any shortcomes?
Line 301: “10 ash species” do you mean 10 sulfate species?
Line 317: “We emitted 65 Mt of ash, and 15 Mt of SO2.” What these numbers based on? Citation needed.
Line 318: “The ash fractions for all 10 ash bins were set at 0.1” What does this mean?
Figure1: move the legend outside the panel b
Figure 3: b and c are not easy to understand. I thought column loading should be the highest line because it only contains production term. Then, other line applies one or several sink terms. So why the grey line is the highest? where's the green line? Why dry deposition and gravitational settling are two different lines (they are the same physical process, just one in the air and one on the surface, right)?
Also, the purple line and gray line are too close in color. I cannot tell which one is which.
Whole paper: CTRL and PRTB. What’re these abbreviations mean? Your test is not control and perturbed; they are Radiative on and off. So, I would suggest to change to different names that easier for audiences to follow.
Line 373: how do you distribute the sulfate and ash amount vertically? uniformly from surface to 1 hPa?
Line 378: “75 Mt of water vapor” what is this number based on? Citation. Also, you haven’t talked about any water progression or radiative impact later on, why do you inject them?
Line 397: explain aerosol index
Line 401: “Ukhov et al. (2023) a better agreement between these fields was achieved” Why don't you use what you have before?
Figure 6: what’s the difference between gravitational settling and dry deposition? I thought they are the same process.
Line 448: “ This number confirms that nearly all of the injected ash (54.17 out of 66.53 Mt)is removed from the atmosphere within 3 months.” many papers presented observed ash particles a year after the Pinatubo eruption, seems conflict with your simulation here. Please comment on that.
Citation: https://doi.org/10.5194/egusphere-2025-4124-RC2 - Between section 1 previous work and section 2 code modification
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,214 | 37 | 13 | 1,264 | 13 | 16 |
- HTML: 1,214
- PDF: 37
- XML: 13
- Total: 1,264
- BibTeX: 13
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
This paper introduces several key enhancements to the WRF-Chem v4.8 model aimed at improving the simulation of volcanic eruptions, including the implementation of wet and dry deposition for ash and sulfate, SO2 oxidation mechanisms, gravitational settling corrections, and the direct radiative effects of volcanic aerosols. The authors also developed the calculation of ash and aeresol radiation, which has the feedback effects to the meteorology. Using the 1991 Mt. Pinatubo eruption as a case study, the authors evaluate the model’s performance through both short-term and long-term experiments, demonstrating clear improvements in mass conservation and a better agreement with satellite observations, particularly when radiative feedback is activated.
Overall, the paper presents a thorough and valuable contribution to the field of volcanic plume modeling. The enhancements address important shorts in WRF-Chem’s capabilities. Here recommend minor revisions before publication.
Main comments:
1. Fig. 4 and 5 show significant improvements in aerosol and SO₂ transport when radiative feedback is included. To further strengthen the model–satellite comparison, the authors should consider applying satellite-specific Averaging Kernels to the model output. This would account for the vertical sensitivity of the satellite retrievals and enable a more rigorous and physically consistent validation.
2. Figure 5 illustrates how radiative feedback alters the spatial pattern and magnitude of the SO₂ plume. The manuscript would benefit from a more detailed explanation of the underlying physical mechanism. Specifically, how does the absorption of solar radiation by ash influencing SO₂ transport and dispersion? A brief discussion linking the radiative heating (e.g., as shown in Fig. 10) to the dynamical response (e.g., enhanced lofting or altered wind patterns) would strengthen the scientific insight of the paper.
3. Some abbreviations are not explicitly defined upon first use, such as LW/SW/PRTB/CTRL.
4. In conclusion section, it is claimed that an open-source preprocessor called PrepEmisSources is developed. However, there is no detail introduction to this tool. Please expand it for more details.
Technical corrections/suggestions:
L. 85: "fixed and error" -> "fixed an error"
L. 119: "The updated SO2 concentration(mol mol−1) is calculated":The rate coefficient k is given in units of cm^3 molecule^-1 s^-1. Please verify and ensure unit consistency throughout the calculation