Impacts of reducing scattering and absorbing aerosols on the temporal extent and intensity of South and East Asian summer monsoon
Abstract. The vast majority of reductions in aerosol emissions are projected to take place in the near future; however, associated impacts on the large-scale circulation over the populated Asian monsoon region remain uncertain. Using the state-of-the-art UK Earth System Model version 1 (UKESM1), this study examines the response of the South Asian and East Asian summer monsoon (SASM and EASM) to idealized reductions in anthropogenic emissions of carbonaceous aerosols and SO2. The analysis focuses on changes in the monsoon temporal extent and intensity of precipitation following decreases in either scattering (SCT), absorbing (ABS) aerosols, or decreases in both. For SCT, the combination of the early transition of land-sea thermal contrast and sea level pressure gradient during the pre-monsoon season together with the late transition in the post-monsoon season associated with the tropospheric warming, advances the monsoon onset but delays its withdrawal, which leads to an extension of the summer rainy season across South and East Asia. The northward shift of the upper-tropospheric Asian jet forced by the SCT reduction causes the anomalous convergence of tropospheric moisture and low-level ascent over northern India and eastern China. The intensification of the South Asian High (SAH) due to the warming over land also contributes to the dynamic instability over Asia. These changes enhance the rainy season of these regions in boreal summer. Reductions in absorbing aerosol act in the opposite sense, making the Asia's rainy season shorter and weaker due to the opposite impacts on land-sea contrast, Asian jet displacement and SAH intensity. With reductions in both SCT and ABS aerosol together the monsoon systems intensify, as the overall impact is dominated by aerosol scattering effects and results in the strengthening of monsoon precipitation and 850-hPa circulation. Although aerosol scattering and absorption play quite different roles in the radiation budget, their effects on the monsoon precipitation seem to add almost linearly. Specifically, the patterns of monsoon-related large-scale responses from reducing both SCT and ABS together are similar to the linear summation of separate effect of reducing SCT or ABS alone, despite of the inherent nonlinearity of the atmospheric systems. Our findings suggest that emission controls that target e.g. emissions of black carbon that warm the climate would have a different response to those that target overall aerosol emissions.
Chenwei Fang et al.
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2023-407', Anonymous Referee #1, 02 May 2023
- RC2: 'Comment on egusphere-2023-407', Anonymous Referee #2, 03 May 2023
Chenwei Fang et al.
Climate Prediction Center (CPC) unified gauge-based daily observations https://psl.noaa.gov/data/gridded/data.cpc.globalprecip.html
Global Precipitation Climatology Project (GPCP) rain gauge-satellite combined precipitation dataset https://www.ncei.noaa.gov/data/global-precipitation-climatology-project-gpcp-daily/access/
ECMWF's (European Center for Medium-Range Weather Forecast) Fifth-generation Reanalysis (ERA5) https://cds.climate.copernicus.eu/cdsapp#!/search?type=dataset&text=ERA5
CMIP6 historical simulations from the UK Earth System Model version 1 (UKESM1) http://esgf-node.llnl.gov/search/cmip6/
Chenwei Fang et al.
Viewed (geographical distribution)
Based on UKESM1 simulations, the authors investigated the different roles of scattering (SCT) and absorbing (ABS) aerosols in the South Asian and East Asian summer monsoons. The authors demonstrated that SCT leads to an extension of monsoon seasons, while ABS acts in the opposite way. In addition, the authors also performed analyses on the linearity between SCT and ABS, and it turns out that the effects of ABS and SCT on SASM and EASM can be added linearly in most cases.
The research question included in this study is interesting and could be beneficial to the field. Overall, the model experiments are well-designed, and the major parts of the analyses are solid and convincing. With that being said, some of the discussions on statistical results (primarily section 3.2) are still confusing and may need more modifications and/or clarifications from my point of view. Therefore, I would recommend a major revision of the current manuscript.
Line 33 - the sentence starting with “Our findings suggest that…” is hard to read and unclear. Please rephrase it.
Line 140: The authors mentioned that SSP3-7.0 represents a high baseline climate with strong pollution, which is obviously true. However, since the simulations included in this study stops in 2024, I guess the emission levels between SSP3-7.0 and the other SSP scenarios are largely the same, which may be worth noting.
Line 196: In addition to the positive difference between the model and ERA5 north of 40N, the negative difference at around 30N (jet core) should also be emphasized. Effectively, the model-simulated jet is wider but less intensive compared to observation, especially during the pre-monsoon season, if I read Fig2g correctly.
Line 199 - “Southerly ind prevailing over East Asia is slightly underestimated”: this is unclear to me: which region is mentioned here? Please clarify.
Paragraph starting at line 218: The discussion on the box plots needs more clarification from my perspective. The authors argued that W2009 and N2016 share similar statistical results, but I feel less confident about this. For example, the SCT-driven extension of the SASM duration is less obvious based on W2009 (Fig3a). I would recommend adding more quantitative descriptions, and including more comparisons between the two methods as well as explaining the possible reasons.
Fig4 and other contour figures: How are the precipitation and wind response calculated? Are they the difference between control runs and aerosol-cut runs? Please clarify this in either the caption or method section.
Fig4c & g: the wind responses over the Indian Ocean (10N-20N) are quite different between W2009 and N2016, which also significantly affect the patterns in Fig4b&f. Can you explain the possible reasons and guess which method is potentially better representing the general structure of SASM? I suggest mentioning this issue in the relative paragraph and adding some discussions.
Line 240: according to Fig5d&h, The decreased precipitation induced by ABS seems to be insignificant. Please double check and clarify the descriptions.
Line 242: The SCT seems to only dominate the precipitation over the north-eastern regions in Fig4, while the decrease in precipitation at around 20N seems to be related to ABS, are nonlinear effect between ABS and SCT. Please double check and clarify.
Paragraph starting at line 244: Similar to my previous concerns about the box plot. W2009 seems to show very small differences between control runs and aerosol-cut runs (e.g., the SASM duration in the SCT run). Do the differences mentioned in this paragraph pass the significance test? Please clarify.
Line 273 - “lowers SLP anomaly over Asian continent compared with that over Indian and western Pacific oceans”: maybe worth noting the opposite changes over the Indian Ocean and western Pacific in Fig8g.
Line 277: The reduced land-sea pressure contrast is not shown in Fig8h, as both land and sea show increases in SLP.
Line 282: How do you get the conclusion that Fig8a & i show patterns with combined effects of SCT and ABS? Do you calculate the map correlation or any other regression methods? It seems to me that the pre and post-monsoon patterns are more complicated. For example, Fig8a is very similar to Fig8c but shows an insignificant pattern over the Indian Ocean. I would suggest a more careful statement here; otherwise, more analyses are necessary.
Fig8: Are these values over land surface pressure instead of sea level pressure? Also, the color bar for panels e-h should be extended since it is hard to see more detailed patterns in panels e and g.