the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 May 2024
Spatial-temporal patterns of anthropogenic and biomass burning contributions on air pollution and mortality burden changes in India from 1995 to 2014
Abstract. Anthropogenic and biomass burning emissions are the major sources of ambient air pollution. India has experienced a dramatic deterioration in air quality over the past few decades, but no systematic assessment has been made to investigate the individual contributions of anthropogenic and biomass burning emissions. In this study, we conducted a pioneering comprehensive analysis of the long-term trends of particulate matter with aerodynamic diameters < 2.5 μm (PM2.5) and ozone (O3) in India and their mortality burden changes from 1995 to 2014, using a state-of-the-art high-resolution global chemical transport model (CAM-chem). Our simulations revealed a substantial nationwide increase in annual mean PM2.5 (6.71 μg m-3 decade-1) and O3 (7.08 ppbv decade-1), with the Indo-Gangetic Plain (IGP) and eastern central India as hotspots for PM2.5 and O3 trend changes individually. Noteworthy substantial O3 decreases were observed in the northern IGP which were potentially linked to NO titration due to a surge in NOx emissions. Sensitivity analyses highlighted anthropogenic emissions as primary contributors to rising PM2.5 and O3, while biomass burning played a prominent role in winter and spring. In years with high biomass burning activity, the contributions from BB on both PM2.5 and O3 changes were comparable with or even exceeding anthropogenic emissions in specific areas. The elevated air pollutants were associated with increased premature mortality attributable to PM2.5 and O3, leading to 97.83 K and 73.91 K per decade. Despite a per capita decrease in the IGP region, the increased population offset its effectiveness.
- Preprint
(5924 KB) - Metadata XML
-
Supplement
(10054 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-974', Anonymous Referee #1, 30 May 2024
The authors have studied the contribution of anthropogenic and biogenic emission change between 1995-2014 to PM2.5 and O3 concentration in India. I have the following comments:
- The maps of the Indian subcontinent don’t seem to be right. If a region is disputed the authors can use dotted lines to represent it.
- In equation 1 of Theil-Sen estimator, the authors should clarify that xi and xj represent points from either PM2.5 or O3 or premature mortality. The sentence used now creates confusion that i and j might represent concentrations/premature mortality from different parameters.
- The authors indicate that they use integrated exposure response function to estimate risk due to PM2.5 exposure, however they don’t mention the contrafactual concentration used to estimate the risk. Does the contrafactual concentration used change over the years? Since the PM2.5 and O3 concentration change over the years in India, the contrafactual concentration used to estimate the risk should also change over the years else the risk estimated might be over or underestimated.
- In line 200, how can ANTHRO contribute to above 100% increase in PM2.5 and O3 concentration?
- As per figure 3, while the annual biomass burning contribution to total PM2.5 and O3 concentration remain lower I have 2 observations:
a) PM2.5 concentration due to burning should at least increase during the burning season i.e. March-May and Sep-Nov, the plot 3b doesn’t capture it.
b) O3 concentration due to burning in plot 3e have large increases in some years whereas the increase doesn’t seem much on other years. What’s the reason for this yearly variability?
- What does the dots in figure 4 in the O3 plot indicate?
- How are the seasonal differences in PM2.5 and O3 concentration in Figure 4 &5 estimated? Are they estimated as the average of the difference over the years from 1995-2014 with respect to base year 1995? Were there any years with notable increase in anthropogenic or biogenic emissions?
- The authors need to check for grammatical errors throughout the manuscript.
Citation: https://doi.org/10.5194/egusphere-2024-974-RC1 - AC1: 'Reply on RC1', Yuqiang Zhang, 21 Oct 2024
-
RC2: 'Comment on egusphere-2024-974', Anonymous Referee #2, 19 Sep 2024
Review for “Spatial-temporal patterns of anthropogenic and biomass burning contributions on air pollution and mortality burden changes in India from 1995 to 2014” submitted to ACP
Summary
The authors conducted a systematic study on the sources and health impacts of air pollutants PM2.5 and O3 in India, one of the most polluted countries. It would be very beneficial for air pollution control if done properly after addressing the major comments below covering the research scope, methods, and result interpretations.
Major comments
- Research scope.
The authors claim that “...but no systematic assessment has been made to investigate the individual contributions of anthropogenic and biomass burning emissions.” However, after a quick search, many are published already such as, https://www.science.org/doi/epdf/10.1126/sciadv.abm4435 and
https://www.sciencedirect.com/science/article/pii/S1352231016304630 . The authors should do a more thorough literature review in India and report both observation-based and model-based results and refine their claim, possibly highlighting the extension to non urban areas. Particularly, explain upfront what is GBD2019 and what have they found and what is not in GBD2019 that is first studied here.
Also, the two decades model simulation seems not fully exploited given the efforts to run the models in the first place. In the current work, only two sources are separated and their magnitude is very different. From emission data only, the results are sort of expected. What sectors are included in the current anth? Is it possible to expand the model or just analysis into specific anthropogenic sources such as industry, indoor fire use for cooking or heating, transport, etc, to provide more practical guidance for future policies and actions? Maybe use certain gas tracers from different sources or the best case scenario, they are separated in the emission data.
2. Methods.
- What are the assumptions for the Mann-Kendall test? Are they compatible with the trend method used?
- Modeled seasonality is the only component evaluated with observations; how about the long term trend, which is more critical for this study? For the seasonality evaluation, how are the stations selected? Have the authors fully investigated the availability of all observations? Also, please show locations for all sites used.
- The magnitudes of y0 and RR seem very important as the authors explain the difference compared with GBD2019 etc in Line 260 etc, changing from generally underestimated pollutants to a highly overestimated health risk. Have any sensitivity studies been done to quantify the uncertainty? Why and how are the current factors chosen? Y0 is country-dependent and age dependent only? RR is disease dependent, shouldn’t it be higher in highly polluted regions such as India and IGP particularly? Explain more about these factors in addition to just showing the number.
- Result Interpretations.
- Basic presentation. Please specify what statistical metrics are used for PM2.5 and ozone in all figures and for all numbers. Annual mean or other? Not just PM2.5 and Ozone. Also, the units are not even consistent. E.g. trend for Ozone and concentration for PM in Figure 4. Please check throughout the manuscript. Lastly, the colorbar/scale for FIgure 2 etc could use some improvement as no spatial info can be seen for regions with highest numbers.
- Interpretations in Section 3.3 need a lot of improvements to be readable. Explain in detail: what you are what you want to tackle; what you are showing in the figures including how you calculated the results; what did you see and what are you comparing certain things with or based on. It seems the authors are comparing different seasons rather than explaining trends (main focus) in Page 9. For Section 3.3.2, What’s the point given Section 3.3.1 and Figure 4? why focus on change from 1995 to 2014 only? BB “contributed significantly”? It seems the analysis is not fair and conclusion driven given the large variation in the nature of BB emission.
- The conclusions have not been reviewed yet given above mentioned problems.
Minor comments
- BB is used in the abstract before its full form is introduced. Check all other abbreviations as well throughout the manuscript.
- Some citations are problematic. Check and correct the whole manuscript for all vague and mismatched citations. Some examples below.
Line 31: What is WHO database? A link?
Line 35: Is Murray et al (2020) the proper citation for GBD2019? I suspect not. Use proper citation for GBD2019 as in the ref list. Also Murray is not in the list if it is really a proper citation here.
Line 40: IQAir is not in the reference list and with no link.
Line 118: is Stanaway… the proper citation for GBD2017?
- “Meanwhile, the faster chemical reaction rates in India due to the strong convection, sunlight, and warm temperatures, making it a hot spot for accumulating major air pollutants compared with other regions and easily affecting the air quality in downwind regions (Zhang et al., 2016, 2021a). “ doesn’t make sense and please rewrite, maybe breakdown for clarity. E.g., warm temperatures seem too conservative for high ozone; strong convection contradicts accumulating.
- Figure S1. Please include the model domain (if not global) and grid cells used for each state in Figure S1 and move it to Figure 1 as the majority of readers would want to see that. Incorporate IGP in the figure caption/figure as it is referred to a lot. Also, as topography plays an important role as well for the transport of air pollutants etc, please include topography and surrounding regions as well. Also, Section 2.3 seems too late to introduce your study region.
- How was the grid size chosen as most of the listed references/model input have 0.5 degree resolution? Why is there a large difference between lat and lon resolution?
- A more detailed description of the emission data used. What temporal and spatial resolutions? Is there any interpolation involved? Are they evaluated? How well do they perform? Were other emission sources used in addition to Anth and BB? Line 157, mention this earlier at your model description.
- Line 84. Is year-varying even a word?
- Are all “trend/trends” and “rates” in the manuscript derived using the same method in Section 2.3? Are the ”change/changes” only referring to change from Year a to Year b? Sometimes they are a bit confusing with the “change/changes”. Please clarify. Also how did you decide where to use trends or changes.
- Why is there only one data point each year for the 6 mon MDA8 in figure 1b?
- Line 154: Again, please show your model domain. Is the performance difference because of boundary conditions?
- Use proper color scale for Figure S4c.
- FIgure S6c is misleading. Should not use stacked bars.
- Line 170, 285: Please show a fire and anthro spatial map somewhere.
- Why not use the emission trend in S7? Also, S7c does not show negative as suggested Line 170, if it is comparing the change with other regions in the first place. Consider changing the colorbar if so.
- Line 175 “Unlike”. They seem similar to me.
- How large is a grid in Figure S9?
- Line 200. What is area-weight? Spatial average?
- FIgure 6. What is the 95% CI? Spatial variability by grid?
- Rewrite Line 286-287 and add base year of 1995. Also, please explain why these years are chosen.
- What is the satellite-derived PM in acknowledgement??? Where was it used?
Citation: https://doi.org/10.5194/egusphere-2024-974-RC2 - AC2: 'Reply on RC2', Yuqiang Zhang, 21 Oct 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 488 | 148 | 195 | 831 | 52 | 31 | 34 |
- HTML: 488
- PDF: 148
- XML: 195
- Total: 831
- Supplement: 52
- BibTeX: 31
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1