the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Sep 2025
Development of CAS-ESM_MMF: Improving East Asian Summer Precipitation Simulation with a Multiscale Modeling Framework
Abstract. Traditional global climate models (GCMs) exhibit substantial biases in simulating precipitation over East Asia, largely due to uncertainties in convection parameterizations. To address this issue, we implement a Multiscale Modeling Framework (MMF), which explicitly resolves convection in a cloud resolving model, into the atmospheric component of the Chinese Academy of Sciences Earth System Model (CAS-ESM). Simulations using CAS-ESM with and without MMF reveal that the MMF implementation significantly reduces the wet bias around the Tibetan Plateau and the dry bias over South China and Southeast Asia. The intensity–frequency characteristics of precipitation are more realistically represented in the MMF version. In addition, the CAS-ESM with MMF better captures the monthly evolution of precipitation and simulates a more realistic seasonal migration of the East Asian rainband, albeit with a somewhat step-wise progression. Further enhancement is achieved by incorporating a convective momentum transport (CMT) parameterization, typically neglected in previous MMF implementations. This inclusion leads to a smoother northward migration of the rainband, more consistent with observations. Comparison with ERA5 reanalysis suggests that this improvement is associated with a more accurate simulation of the western Pacific subtropical high. These results demonstrate that MMF, especially when combined with CMT, substantially improves the simulation of East Asian precipitation. This modeling advancement offers a promising approach for evaluating regional precipitation responses to future climate change.
-
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
(2412 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2412 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-4311', Anonymous Referee #1, 21 Oct 2025
- AC1: 'Reply on RC1', Wei Liao, 30 Nov 2025
-
RC2: 'Comment on egusphere-2025-4311', Anonymous Referee #2, 22 Oct 2025
In the manuscript, the authors developed a Multiscale Modeling Framework (MMF) by implementing a 2-D cloud-resolving model SAM into the Chinese CAS Earth system model (CAS-ESM) and compared its performance over China in the simulation of East Asian summer monsoon with the conventional version of the CAS-ESM. They further included convective momentum transport (CMT) in the MMF as a further improvement of the model’s capability. Their results showed that the MMF improved the annual variation of precipitation in several selected regions over China. The precipitation intensity pdf is also better simulated. In the monsoon region (south China and east China) the northward progression of monsoon precipitation is captured in the MMF, especially the one including CMT. Overall, the improvements appear modest, but the capacity building itself is worth documenting. The paper is well written, and the topic fits the scope of GMD. I recommend minor revisions before acceptance.
Major comments
One of the main features of summer precipitation over China besides the north-south progression of the monsoon rain is the eastward propagation of the nocturnal precipitation systems in the foothills of the Tibetan Plateau. Is it captured in the CAS-ESM-MMF? A composite diurnal cycle using time-longitude hovmuller plot for region b) should be able to show this.
The inclusion of convective momentum transport (CMT) shows some interesting improvements in the monsoon simulation. What are the physical or dynamical mechanisms that led to such improvements? Fig. 11 shows the 500 hPa geopotential height and circulation changes. How are they related to CMT? Can the authors look at the tendencies of winds due to CMV and whether they can explain the circulation differences in the monsoon region?
Minor comments
Fig. 2 and several other figures. The color scheme for the difference plots is opposite to what is typically used in the literature. While it’s more of a personal preference, I’d suggest the authors use blue color for positive precipitation biases and red for negative biases for ease of reading.
L159. The SST is a climatological mean over 1995-2005. Can the simulation be called 2001-2006 simulation? I understand that Jan. 1, 2000 atmospheric conditions were used for model initialization, but the initial atmospheric state information should be lost quickly.
Fig. 6. Is the GPM average also over 2001-2006? As mentioned above, since SST is a climatological mean, comparing model simulations with 2001-2006 observations may not be a consistent comparison.
L170-171. “Overall, the MMF improves the simulation of convective precipitation on a global scale, though some systematic errors remain.” From the global correlation coefficient and the RMSE, MMF appears to have degraded the precipitation simulation instead. So, I wouldn’t make this statement.
L94. Delete “that” after “while”.
L98. Delete “situated”.
L145. “Analytic”? Do you mean “Analysis”?
L219. “contribute”
Please check for grammatic errors in general.
Citation: https://doi.org/10.5194/egusphere-2025-4311-RC2 - AC1: 'Reply on RC1', Wei Liao, 30 Nov 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-4311', Anonymous Referee #1, 21 Oct 2025
This paper, “Development of CAS-ESM_MMF: Improving East Asian Summer Precipitation Simulation with a Multiscale Modeling Framework,” presents the development and evaluation of a Multiscale Modeling Framework (MMF) version of the CAS-ESM model. By embedding a cloud-resolving model into the atmospheric component of CAS-ESM, the study demonstrates substantial improvements in simulating East Asian summer precipitation—particularly in alleviating wet and dry biases and in better capturing the seasonal evolution and northward migration of the East Asian rainband. The authors further enhance the framework by incorporating a convective momentum transport (CMT) parameterization, which leads to more realistic large-scale circulation and rainfall migration characteristics.
Under the current background of reduced maintenance and uncertain continuity of the E3SM-MMF codebase, this work represents a timely and valuable step toward advancing operational MMF-based climate simulations. The paper is well-structured and methodologically sound. I think the paper will be a valuable contribution to the MMF modeling community.
Specific Comments
-
Clarification of CMT Implementation (Section 3.5):
Please clarify whether, in the CAS-ESM_MMF_MF experiment, the convective momentum transport on the GCM side (Richter & Rasch 2008) remains active or is fully replaced by the ESMT parameterization. This distinction is essential to interpret whether the CMT effects arise solely from the embedded CRM or from combined GCM- and MMF-level treatments. - Line 118:
Remove the extra space before the comma in “momentum feedback , namely.” - Line 170:
The sentence “Overall, the MMF improves the simulation of convective precipitation on a global scale, though some systematic errors remain.” seems overstated, since the CORR and RMSE values in Figure 2e appear less favorable than those in Figure 2c. Please rephrase to acknowledge the slight degradation in global statistics despite regional improvements. - Line 258:
Since “eastern China” is newly introduced as a defined analysis region, please mark this region’s location on the maps in Figure 6 for clarity. - Lines 325–330:
The statement that “After introducing the CMT parameterization … convective intensity … is notably reduced” is supported by Figure 10; however, the new simulation appears drier than observations in several regions. Please discuss this residual dry bias and, if feasible, whether modest tuning could better balance the overall magnitude of precipitation. - Figures 4d & 5d:
Revise the title and references to “the middle and lower reaches of the Yangtze River Basin” for consistency with Figure 1 and the regional definitions used throughout. - Figure 6d:
The northward propagation of the rainband is not as visually clear as in panels (a–c). Consider adding a dashed line or similar visual guide to indicate its seasonal progression. - Figure 9:
The northward propagation of the rainband in the CAS-ESM_MMF_MF run is less clear than in the observational panels and the original MMF run. Please consider adding a visual guide (e.g., a dashed ridgeline of maximum zonal-mean precipitation or a shaded latitude band for the top decile).
In addition, CAS-ESM_MMF_MF shows local monthly peaks from May to September that are not evident in Figure 7d. Could the authors discuss whether these intra-month oscillations might be related to the inclusion of CMT/ESMT?
Citation: https://doi.org/10.5194/egusphere-2025-4311-RC1 - AC1: 'Reply on RC1', Wei Liao, 30 Nov 2025
-
-
RC2: 'Comment on egusphere-2025-4311', Anonymous Referee #2, 22 Oct 2025
In the manuscript, the authors developed a Multiscale Modeling Framework (MMF) by implementing a 2-D cloud-resolving model SAM into the Chinese CAS Earth system model (CAS-ESM) and compared its performance over China in the simulation of East Asian summer monsoon with the conventional version of the CAS-ESM. They further included convective momentum transport (CMT) in the MMF as a further improvement of the model’s capability. Their results showed that the MMF improved the annual variation of precipitation in several selected regions over China. The precipitation intensity pdf is also better simulated. In the monsoon region (south China and east China) the northward progression of monsoon precipitation is captured in the MMF, especially the one including CMT. Overall, the improvements appear modest, but the capacity building itself is worth documenting. The paper is well written, and the topic fits the scope of GMD. I recommend minor revisions before acceptance.
Major comments
One of the main features of summer precipitation over China besides the north-south progression of the monsoon rain is the eastward propagation of the nocturnal precipitation systems in the foothills of the Tibetan Plateau. Is it captured in the CAS-ESM-MMF? A composite diurnal cycle using time-longitude hovmuller plot for region b) should be able to show this.
The inclusion of convective momentum transport (CMT) shows some interesting improvements in the monsoon simulation. What are the physical or dynamical mechanisms that led to such improvements? Fig. 11 shows the 500 hPa geopotential height and circulation changes. How are they related to CMT? Can the authors look at the tendencies of winds due to CMV and whether they can explain the circulation differences in the monsoon region?
Minor comments
Fig. 2 and several other figures. The color scheme for the difference plots is opposite to what is typically used in the literature. While it’s more of a personal preference, I’d suggest the authors use blue color for positive precipitation biases and red for negative biases for ease of reading.
L159. The SST is a climatological mean over 1995-2005. Can the simulation be called 2001-2006 simulation? I understand that Jan. 1, 2000 atmospheric conditions were used for model initialization, but the initial atmospheric state information should be lost quickly.
Fig. 6. Is the GPM average also over 2001-2006? As mentioned above, since SST is a climatological mean, comparing model simulations with 2001-2006 observations may not be a consistent comparison.
L170-171. “Overall, the MMF improves the simulation of convective precipitation on a global scale, though some systematic errors remain.” From the global correlation coefficient and the RMSE, MMF appears to have degraded the precipitation simulation instead. So, I wouldn’t make this statement.
L94. Delete “that” after “while”.
L98. Delete “situated”.
L145. “Analytic”? Do you mean “Analysis”?
L219. “contribute”
Please check for grammatic errors in general.
Citation: https://doi.org/10.5194/egusphere-2025-4311-RC2 - AC1: 'Reply on RC1', Wei Liao, 30 Nov 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,119 | 133 | 25 | 1,277 | 23 | 23 |
- HTML: 1,119
- PDF: 133
- XML: 25
- Total: 1,277
- BibTeX: 23
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Guangxing Lin
Zhaohui Lin
He Zhang
Wenbin Kou
Xiaojie Guo
Zhenghui Xie
Chenglai Wu
Minghua Zhang
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2412 KB) - Metadata XML
This paper, “Development of CAS-ESM_MMF: Improving East Asian Summer Precipitation Simulation with a Multiscale Modeling Framework,” presents the development and evaluation of a Multiscale Modeling Framework (MMF) version of the CAS-ESM model. By embedding a cloud-resolving model into the atmospheric component of CAS-ESM, the study demonstrates substantial improvements in simulating East Asian summer precipitation—particularly in alleviating wet and dry biases and in better capturing the seasonal evolution and northward migration of the East Asian rainband. The authors further enhance the framework by incorporating a convective momentum transport (CMT) parameterization, which leads to more realistic large-scale circulation and rainfall migration characteristics.
Under the current background of reduced maintenance and uncertain continuity of the E3SM-MMF codebase, this work represents a timely and valuable step toward advancing operational MMF-based climate simulations. The paper is well-structured and methodologically sound. I think the paper will be a valuable contribution to the MMF modeling community.
Specific Comments
Clarification of CMT Implementation (Section 3.5):
Please clarify whether, in the CAS-ESM_MMF_MF experiment, the convective momentum transport on the GCM side (Richter & Rasch 2008) remains active or is fully replaced by the ESMT parameterization. This distinction is essential to interpret whether the CMT effects arise solely from the embedded CRM or from combined GCM- and MMF-level treatments.
Remove the extra space before the comma in “momentum feedback , namely.”
The sentence “Overall, the MMF improves the simulation of convective precipitation on a global scale, though some systematic errors remain.” seems overstated, since the CORR and RMSE values in Figure 2e appear less favorable than those in Figure 2c. Please rephrase to acknowledge the slight degradation in global statistics despite regional improvements.
Since “eastern China” is newly introduced as a defined analysis region, please mark this region’s location on the maps in Figure 6 for clarity.
The statement that “After introducing the CMT parameterization … convective intensity … is notably reduced” is supported by Figure 10; however, the new simulation appears drier than observations in several regions. Please discuss this residual dry bias and, if feasible, whether modest tuning could better balance the overall magnitude of precipitation.
Revise the title and references to “the middle and lower reaches of the Yangtze River Basin” for consistency with Figure 1 and the regional definitions used throughout.
The northward propagation of the rainband is not as visually clear as in panels (a–c). Consider adding a dashed line or similar visual guide to indicate its seasonal progression.
The northward propagation of the rainband in the CAS-ESM_MMF_MF run is less clear than in the observational panels and the original MMF run. Please consider adding a visual guide (e.g., a dashed ridgeline of maximum zonal-mean precipitation or a shaded latitude band for the top decile).
In addition, CAS-ESM_MMF_MF shows local monthly peaks from May to September that are not evident in Figure 7d. Could the authors discuss whether these intra-month oscillations might be related to the inclusion of CMT/ESMT?