the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Sep 2026
Sensitivity of NO2 photolysis frequency to aerosol optical properties
Abstract. The photolysis frequency of nitrogen dioxide (NO2), J(NO2), links aerosol radiative effects with near-surface photochemistry and ozone formation. Its response to aerosol optical properties is difficult to quantify because scattering, absorption, hygroscopic growth, black carbon coating properties, aerosol vertical distribution, and surface reflection are coupled. We developed a framework combining Mie calculations, the Constrained Parameter (CP) method, and the Tropospheric Ultraviolet and Visible (TUV) model to quantify sensitivities and interactions controlling J(NO2) using inputs representative of northern China. Two pathways were examined: direct perturbation of aerosol optical depth (AOD), single-scattering albedo (SSA), the asymmetry factor (g), and surface albedo in TUV, and a microphysical pathway in which AOD, SSA, and g were derived from aerosol microphysical parameters using Mie calculations. Among direct radiative inputs, SSA and surface albedo showed the largest CP sensitivities, approximately 12.0×10-3 and 5.3×10-3 dJ/(dx/x), respectively, indicating stronger control by scattering–absorption partitioning and the lower radiative boundary than by column extinction alone. In the microphysical analysis, coating thickness, black carbon density, black carbon absorption enhancement (BCAE), the imaginary part of the black carbon refractive index, hygroscopicity, and surface albedo were key factors. Interaction and fixed-factor analyses showed that sensitivity rankings were state dependent. The strongest interaction occurred between coating thickness and BCAE. Better constraints on SSA, surface albedo, coating thickness, BCAE, hygroscopicity, and aerosol vertical distribution are needed to improve photolysis-frequency calculations and assessments of coupled fine particulate matter and ozone pollution.
- Preprint
(7714 KB) - Metadata XML
-
Supplement
(3811 KB) - BibTeX
- EndNote
Status: open (until 02 Nov 2026)
- RC1: 'Comment on egusphere-2026-4877', Anonymous Referee #1, 28 Sep 2026 reply
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 270 | 51 | 35 | 356 | 31 | 37 | 38 |
- HTML: 270
- PDF: 51
- XML: 35
- Total: 356
- Supplement: 31
- BibTeX: 37
- EndNote: 38
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript explores the sensitivity of calculated JNO2 to aerosol microphysical parameters that used to compute aerosol optical properties, which are then put into the TUV model to compute JNO2. While a lot of work went into the study, novel or useful results are not yet apparent. A major revision could address some of the issues listed below.
The calculation has two complex steps:
(A) microphysical properties (size distr, CT, BCAE, K, RH, ...) ---> optical properties (AOD, ssa, g)
(B) optical properties (AOD, ssa, g) ---> JNO2
Calculating an overall sensitivity for the combined processes (A + B) is problematic because (A) is fundamentally local, based on properties of an air parcel at the point of interest, while (B) JNO2 is mostly non-local, since it depends on the optical properties of the atmosphere above and below the point of interest. To combine these two conceptually different sensitivities can be misleading and even wrong. Several issues arise, with SSA being a good illustration:
(1) The response of JNO2 to SSA can be a strong function of altitude (the non-local effect), even within a well mixed layer. See for example Figure 9f of Palancar et al. 2013 (doi:10.5194/acp-13-1011-2013): For ssa=0.95, JNO2 is enhanced in the upper part of the PBL and reduced in the lower part. In other words, the sensitivity of JNO2 to AOD is changing sign as a function of altitude, something not captured by the present manuscript. Similarly, O3 concentrations averaged through the PBL could have entirely different responses to AOD than JNO2 at the ground, maybe even in the opposite direction.
(2) The definition of sensitivity coefficient used here, dJ/(dx/x), is strange. Although units are not specified, as written they would have to be 1/sec, rather than dimensionless as is more conventional, e.g. (dJ/J)/(dx/x). Also, the numbers don't make sense. For example, Fig. 1 suggest that JNO2 has a constant SSA sensitivity, dJ/(dx/x) = 9e-3 (presumably units of 1/sec). Therefore a 10% variation in SSA (dx/x = 0.1) would result in a constant dJ = 9e-4/sec. The true sensitivity of JNO2 to SSA is far from constant, and ranges from zero (no aerosol effect) when AOD=0, to 100% reduction for large AOD. The constant values given in Fig. 1 do not seem to have a physical explanation.
(3) A practical consideration is that the intermediate optical properties (AOD, SSA, g) are usually better known than the microphysical properties. AOD and SSA at several visible and UV wavelengths are routinely measured by CIMEL radiometers of the AERONET network, with about 20 sites in China. This offers an opportunity to evaluate the calculation of process (A).
If the authors wish to consider a major revision, a possibility would be a cleaner separation between the local and non-local problem (maybe two separate papers):
1. A detailed description of Process A, to derive the "intermediate" optical properties SSA, g, AOD/m (the latter is per meter, local). Importantly, comparisons should be made with AERONET AOD and SSA observational data.
2. A detailed study of process B, how JNO2 depends on input values of AOD, SSA, g, possibly using the uncertainty ranges estimated in the study of process A. This should be done as a function of height within and above the aerosol layer, solar zenith angle, different ground albedo, etc. Many such studies have already been published, but there may be room for more detail and regional specifics. Sensitivities should be defined more carefully.