the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 Aug 2024
Impacts of irrigation on ozone and fine particulate matter (PM2.5) air quality: Implications for emission control strategies for intensively irrigated regions in China
Abstract. Intensive irrigation is known to alleviate crop water stress and alter regional climate, which can in turn influence air quality, with ramifications for human health and food security. However, the interplay between irrigation, climate and air pollution in especially the simultaneously intensively irrigated and heavily polluted regions in China has rarely been studied. Here we incorporated a dynamic irrigation scheme into a regional climate-air quality coupled model to examine the potential impacts of irrigation on ozone (O3) and fine particulate matter (PM2.5) in China. Results show that irrigation increases the concentrations of primary air pollutants, but reduces O3 concentration by 3–4 ppb. PM2.5, nitrate and ammonium rise by 28 %, 70 % and 40 %, respectively, upon introducing irrigation, with secondary formation contributing to 5–10 %, ~60 %, and 10–30 %, respectively. High humidity and low temperature are the top two factors promoting the formation of ammonium nitrate aerosols. To mitigate these adverse effects on PM2.5 air quality, we found that a 20 % combined reduction in NH3 and NOx emissions is more effective compared with individual emission reductions, while the enhancement in O3 due to the NOx reduction can be completely offset by irrigation itself. Our study highlights the potential benefits of irrigation regarding O3 pollution but possible problems regarding PM2.5 pollution under currently prevalent irrigation modes and anthropogenic emission scenarios, emphasizing the need for an integrated approach to balance water conservation, air pollution, climate change mitigation and food security in the face of development needs.
- Preprint
(2718 KB) - Metadata XML
-
Supplement
(969 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-1557', Anonymous Referee #1, 29 Aug 2024
The authors incorporated a dynamic irrigation scheme into a regional climate-air quality coupled model to investigate the impacts of intensive irrigation on air quality in China. They found that irrigation increases the concentrations of PM2.5 but reduces O3 concentration. They further suggested a 20% combined reduction in NH3 and NOx emissions to mitigate the adverse effects of irrigation on PM2.5 air quality based on additional sensitivity experiments. The manuscript was generally well organized and well written. I only have some minor comments/questions.
- The physical processes of the irrigation scheme need some further clarifications. In this study, the model checked if irrigation can be triggered at each timestep. Did you mean irrigation will be turned off immediately when the relative soil moisture is above the management allowable deficit (i.e., 60%)? Note that the objective of irrigation is to make the relative soil moisture approach 100% (i.e., eq. 1). Another question: can the irrigation scheme be activated during both daytime and nighttime?
- The O3 concentrations in the PBL and in the free atmosphere were both reduced due to irrigation. The authors have well analyzed the reasons for the reduction in the PBL. However, it remains unclear why the O3 concentration was also reduced above the PBL. I suggest adding some explanations in this regard.
Citation: https://doi.org/10.5194/egusphere-2024-1557-RC1 -
RC2: 'Comment on egusphere-2024-1557', Anonymous Referee #2, 18 Oct 2024
General comments
This study performs a model experiment to estimate the impact of irrigation on meteorology in China, and subsequent effects on air pollutant concentrations. The authors implement a new dynamic irrigation scheme in the WRF-GC model, which they show substantially reduces LST biases in heavily irrigated regions of China. They then quantify the impacts of irrigation, which interestingly has a substantial impact on boundary layer meteorology, resulting in increased PM2.5 but decreased ozone. They then explore the impacts of reduced emissions in several scenarios, and discuss their trade-offs in the context of climate change and potential future increases in irrigation.
The paper is very well written, and the analysis is performed to a high standard. Most of my comments below are of a minor nature. In general, the methods are described adequately, and the results are comprehensively reported, although there are rather a lot of figures. I highlight below where I believe the paper could benefit from more comparison with measurements. I also believe that the discussion/conclusion could be more tightly written, so the reader could more easily find the key takeaways of the paper.
Specific comments
Some of the figures and much of the results section discusses the modelling results for two specific locations, Puyang and Chengdu. Since the results are reported in detail for these two cities, there should be some justification added for why they were chosen, and why not just report averages over your areas of interest (NCP and SCB)?
Would it be useful to calculate human/plant health-relevant metrics for O3 for the sensitivity scenarios? E.g. for Figure 12, it may be useful to add some discussion of how much the key ozone metrics such as MDA8, AOT40 etc. This might help to put your results in context with other studies looking at ozone reduction policies.
Compare your modelled soil moisture with observations? It would be very useful to know whether the irrigation scheme you implement in WRF-GC is able to achieve realistic soil moisture levels in the NCP and SCB. Since this paper introduces a dynamic irrigation scheme into this model for the first time, it should be evaluated.
The result that adding irrigation into WRF-GC improves comparison with MODIS LST substantially in heavily irrigated areas is very interesting and maybe deserves to be featured more prominently. It leaves me wondering whether IRR has any decreased biases of T2, windspeed, RH and other meteorological variables in the NCP and SCB. However, I recognise that you evaluate the nudged CTL run rather than the non-nudged IRR run. That being said, it would be useful to contextualise the changes in the aforementioned meteorological variables between the NOIRR and IRR run in the NCP+SCB with comparisons with meteorological observations, and comment on whether including irrigation improves any biases.
On L388, you give percentage change for PM2.5 and its components. I think similar relative change percentages should be given throughout the results section (where possible), to help contextualise the changes.
I think in Figure 12 it is confusing to compare the concentrations in the sensitivities to NOIRR rather than to IRR. In the paper, IRR is framed as the most realistic model run, as the point is well made that including a representation of irrigation is necessary for accurately representing meteorology (and therefore chemistry). Therefore, it seems to make more sense to me to compare the sensitivity scenarios with the most realistic representation of the current atmosphere, the IRR scenario.
Technical corrections
L35-6: I think I know what you mean, but referring to ozone as a primary pollutant here is confusing. Also, citing a study to show that PM/ozone is a major cause of avoidable mortality in China here would be useful, e.g. one of the Global Burden of Disease studies.
L37: 106 μg m–3 seems high and I can’t find it in An et al. (2019). In this paper, Figure 1 seems to show NCP PM2.5 reaching a maximum of an annual mean value ~83 μg m–3 during 2012.
L40: Use more specific language than ‘high PM2.5’. Does this refer to NAAQS or WHO AQGs or something else?
L41: Positive trends rather than upwards?
L51: Missing word “During the COVID-19 [pandemic/period/lockdowns/etc.]”
L52: Typo, should be rose not rosed
L56: Maybe [several/multiple/?] instead of “considerable”
L76: ‘lowers’ the PBLH instead of ‘thins’? And ‘a’ coupled model rather than ‘the’?
L128: Typo: biogenic not biomass
L208-210: I think the four sensitivity experiments could be more clearly described here
L243-244: Does the ‘default model’ refer to CTL? If so, maybe clearer to just use “CTL”
L288: would be useful to quantify the relative increase in soil moisture
L491: Discussion and Conclusions is maybe a better title for this section?
Fig2: The figure caption should make it clear these are seasonal averages
Fig6: The x axis ticklabels could be shown without decimal places to make them easier to read in a, b and f.
Fig13: Please repeat the figure legend from Figure 12. Figures should be comprehensible in isolation.
The authors could consider using colourblind friendly colormaps throughout. For example, red-green diverging and rainbow colourbars are not colourblind friendly.
Citation: https://doi.org/10.5194/egusphere-2024-1557-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 238 | 73 | 78 | 389 | 37 | 8 | 6 |
- HTML: 238
- PDF: 73
- XML: 78
- Total: 389
- Supplement: 37
- BibTeX: 8
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1