the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Analysis of GEFS-Aerosols annual budget to better understand the aerosol predictions simulated in the model
Partha S. Bhattacharjee
Li Zhang
Raffaele Montuoro
Barry Baker
Jeff McQueen
Georg A. Grell
Stuart A. McKeen
Shobha Kondragunta
Xiaoyang Zhang
Gregory J. Frost
Fanglin Yang
Ivanka Stajner
Abstract. In September 2020, a global aerosol forecasting model was implemented as an ensemble member of the National Oceanic and Atmospheric Administration (NOAA) National Centers for Environmental Prediction (NCEP) Global Ensemble Forecasting System (GEFS) v12.0.1 (hereafter referred to as “GEFS-Aerosols”). In this study, GEFS-Aerosols simulation results from September 1, 2019 to September 30, 2020 were evaluated using an aerosol budget analysis. These results were compared with results from other global models as well as reanalysis data. From this analysis, the global average lifetimes of black carbon (BC), organic carbon (OC), dust, sea salt, and sulfate are 4.06, 4.29, 4.59, 0.34 and 3.3 days, respectively, with the annual average loads of 0.135, 1.29, 4.52, 6.80 and 0.50 TG. Compared to National Aeronautics and Space Administration (NASA)’s Goddard Earth Observing System-Goddard Chemistry Aerosol and Radiation Transport Model (GEOS4-GOCART), the aerosols in GEFS-Aerosols have a relatively short lifetime because of the faster removal processes in GEFS-Aerosols. Meanwhile, in GEFS-Aerosols, aerosol emissions are the determining factor for the mass and composition of aerosols in the atmosphere. The size (bin) distribution of aerosol emissions is as important as its total emissions, especially in simulations of dust and sea salt. Also most importantly, the strong monthly and interannually variations in natural sources of aerosols in GEFS-Aerosols suggests that improving the accuracy of prognostic concentrations of aerosols is important for applying aerosol feedback to weather and climate predictions.
- Preprint
(2430 KB) - Metadata XML
- BibTeX
- EndNote
Li Pan et al.
Status: final response (author comments only)
-
RC1: 'Comment on gmd-2023-61', Anonymous Referee #1, 22 Jun 2023
This manuscript described a process-based budget analysis of the GEFS-Aerosols chemical transport model, including the processes of emissions, reactions and removal. This model budget analysis includes the comparison to the MERRA-2 and GEO4-GOCART, but has few verification with observations, making it hard to evaluate which process has big uncertainties.
Here are the detailed comments.
Section 2.1 and 2.2. The mass balance equation and associated processes did not mention the model’s advection, diffusion, and physical processes. How well has the aerosol mass been conserved in these processes? What’s this model’s top boundary treatment, and how does the model control the mass leakage through the domain top?
The gravitational settling of aerosols is usually applied to mass movement from upper layers to the lowest layer of the model, which won’t affect the total mass. The removal of aerosol mass from the model’s lowest layer to ground surface refers to dry deposition, which should include the gravitational sediment. Does the GEFS-Aerosol’s dry deposition scheme exclude the gravitational sedimentation? Please clarify.
Fig 2, and line 115, Are the sea salt AOD calculation method same in GEFS-Aerosol and MERRA-2? Besides AOD comparison, it is better to have mass concentration comparison with observations for sea salt near sea surface. How about the mass flux for other species?
Fig 3, 4 and line 117-120. Similar question, does the dry deposition of the lowest model layer exclude the sedimentation? It would better to show the net surface flux of sea salt in GEFS-Aerosol and MERRA-2, to show whether their emission and removal processes are balanced.
Fig 6 and corresponding discussion around line 138-145. The model error for dust mass is highly correlated with the dust emission, but these correlations do not exist for BC, OC and sea salt. Any further discussion about the difference among these species, which processes result in the difference?
Fig 10. What is the temporal and spatial extents of these profiles? Please clarify. It is better to separate the profiles over ocean and land, and have a deeper discussion for sea salt and land-source aerosols etc.
Line 282-290. It is more convincing if any direct comparison with observations can be included here, like CALIOP or in-situ measurements.
Table 2 and Section 3.11 It is better to make consistent by changing the units of wet/dry deposition and sediment from percent to Tg/Year, comparable to emission etc.
Line 350-365. Giving the big discrepancies between the models, is there any observation available to verify aerosol removal process or surface net mass flux?
Citation: https://doi.org/10.5194/gmd-2023-61-RC1 -
AC1: 'Reply on RC1', L. Pan, 28 Aug 2023
The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2023-61/gmd-2023-61-AC1-supplement.pdf
-
AC1: 'Reply on RC1', L. Pan, 28 Aug 2023
-
RC2: 'Comment on gmd-2023-61', Anonymous Referee #2, 01 Aug 2023
The authors conduct aerosol budget analysis based on 1-year GEFS-Aerosols simulations. The budget analysis for an operational model is an important work for the community. The approach undertaken is solid, but some discussions could be clarified. I recommend ‘the acceptance after revision’.
My major comments:
-The authors plainly describe what they see, not what they learn from the analysis. The manuscript could be further improved if the authors present the results with more scientific insight.
-It is very difficult to verify simulated lifetime and annual emission and removal due to lack of observational evidence. The authors only compare the GEFS-Aerosols results with GEOS4-GOCART from Colarco et al. 2010. How about the AeroCom consensus?
Here are some minor comments for the authors to consider:
-Line 36 ‘As a first step towards this goal, ‘GEFS-Aerosols was implemented to replace NGAC. The efforts to enable prognostic aerosol capability toward the goal started with the implementation of NGAC. It is not clear to me why the authors view the GEFS-Aerosols implementation as the "first step."
-Line 43-44: ‘because these processes occur before the model output and they are the determinants of aerosol concentration.” I agree that budget analysis is important to examine model’s fidelity/performance. However, the justification “these processes occur before the model output and they are the determinants of aerosol concentration” is very odd and weak. Budget analysis can reveal whether the model have the bulk emission and removal processes right. Whether these tendency diagnostics are model output is totally irrelevant.
-Line 40 ‘instead of focusing on aerosol concentration and aerosol optical depth (AOD) in a general aerosol evaluation’. Comparing GEFS-Aerosols model output with PM/AOD observations is needed to thoroughly assess the model performance and identify potential model deficit. It is certainly all right for the authors to focus on budget analysis in this manuscript. Since the model vs observation evaluation has been conducted and reported in other papers [Lines 109-111], the authors should briefly describe the efforts.
-The introduction is fairly short. The authors can easily extend it with more scientific motivation or technical descriptions.
-Line53 Eq1: Initial + Emissions + Reactions = Final + Removal
Based on the governing equation, I’ll probably present the equation as
Final = Initial + Emissions + Reactions – Removal
-Line 78: 2.3 GEFS-Aerosols. Consider presenting this sub-section first in Section 2.
-Line 83: ‘GOCART’ Please define the acronym
-Line 104: ‘Fire Radiant Power (FRP) ‘. Fire Radiative Power?
-Line 113: ‘These processes ultimately define the aerosol concentration and AOD output by the model.’ These processes ultimately determined 3-d aerosol distribution, which in term affect concentration and AOD. But this sentence is somehow odd.
-Line 115 ‘MERRA2’ MERRA-2 is also based on GOCART. Does sea salt emission and removal scheme in GEFS-Aerosols differ from those in MERRA2?
-Line 138 ‘Fig 6’ The principal behind the budget analysis is that aerosols net production is approximately equal to net loss when averaged over a long time (say multiple years). It is not clear whether the monthly residual (Left side of Eq1 – Right side of Eq 1) should be interpreted as ‘model error’.
-Line 155 “Therefore, the model errors for dust and sea salt are higher than those for BC and OC, while the model errors for dust are the highest.”. The text seems indicate that the model errors for dust and sea salt are caused by non-linearity in the emission/removal scheme. This is not necessarily true.
-Line 159 ‘Global Aerosol Mass’. It is insightful to specify when specific aerosol species reach max and min. For instance, dust loading peaks in June and reached min in Nov. This results are consistent with Africa dust activities. However, it seems unnecessary for the authors to specify the exact date.
-Line 172 ‘Annual trend’. How annual trend can be inferred from one-year simulation? Please clarify it.
-Line 181-186. The discussions about the partition can be presented in a table.
-Line 189 ‘Aerosol emissions are directly and indirectly related to their mass in the atmosphere’. Aerosol loading is certainly related to their emissions, and aerosol emissions certainly affect aerosol mass. However, the statement is very awkward.
-Line 224-225: ‘the size distribution of aerosol emissions becomes too important for the removal process in GEFS-Aerosols simulations when the aerosol particle size is not changed in the model’ Please clarify this sentence.
-Line 229 ‘as they do not undergo a size (bin) change.’ The GOCART is a bulk mass scheme. It’s not clear to me why the authors expect bin change.
-Line 265 ‘GOCART’ Presume it’s GOES4-GOCART. It does not hurt to make it clear.
-Line 263-269, The differences in lifetime between GEFS-Aerosols and GEO4-GOCART are attributed to model resolution and simulation period. Both model use GOCART scheme. The differences and similarities between the two GOCART schemes should be considered. The difference between the two host AGCMs should also be discussed. If identical emission and removal scheme are implemented in both GEFS and GEOS4, the emissions and removal fluxes from the two model will still be different. The model with more active moisture process may produce more wet removal. The model with more noisy wind field may produce higher dust emissions.
-Line 310 ‘interannual variations’ It is not clear why the authors attempt to analyze interannual variations with a 15-month data set.
-Line 332-333: ‘The study of monthly and interannual variations in aerosol mass is important because it determines whether it is appropriate to use aerosol climatology fields rather than aerosol prognostic fields in weather forecasting to save computational resources.’ I thought that the use of climatological, prescribed, or prognostic aerosols in the operational model is largely determined by the resource constraint. The study of monthly and interannual variations is important because it addresses many important aerosol-related scientific questions.
Citation: https://doi.org/10.5194/gmd-2023-61-RC2 -
AC2: 'Reply on RC2', L. Pan, 28 Aug 2023
The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2023-61/gmd-2023-61-AC2-supplement.pdf
-
AC2: 'Reply on RC2', L. Pan, 28 Aug 2023
Li Pan et al.
Li Pan et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 189 | 47 | 13 | 249 | 1 | 1 |
- HTML: 189
- PDF: 47
- XML: 13
- Total: 249
- BibTeX: 1
- EndNote: 1
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1