The Observation Path Problems and the Formation Conditions of the Elevated Layer of Black Carbon Aerosol

Studies on the detection of layers with elevated black carbon aerosol (BC) concentrations and the formation conditions of these layers help understand the vertical distribution of BC concentrations, which will provide a basis for the assessment of climate effects and early pollution warnings. By using the Weather Research and Forecasting with Chemistry (WRF-Chem) numerical model, we performed a numerical simulation analysis on the authenticity of strongly elevated BC concentration layers that were detected by an aircraft in the mixing layer over Harbin, China, which is a high-emission area, on a clear sunny afternoon in the early heating period of 2016. We then discuss possible problems and solutions when non-vertical paths are used to detect the vertical distribution of BC concentrations. Finally, we discuss the favorable conditions for the formation of elevated BC concentration layers by a weak vertical flow based on the simulation. The modeling results show that the horizontal variability of BC concentration in the mixing layer in the observation area in Harbin was sufficiently large during the measurement. This produced a false elevated layer, as detected by the aircraft during one round of spiral flight in the mixing layer. The root mean square of the horizontal distribution of BC concentration did not change with height in the mixing layer during the daytime, but it decreased with the thickness of the mixing layer and was higher in the mixing layer than in the free atmosphere. Therefore, the thinner the mixing layer, in which the vertical distribution of the BC concentration is detected in an inclined path, the stronger interference of the horizontal variability on the detected results. In the daytime, due to strong turbulence in the mixing layer, weak vertical uplift is not favorable for the occurrence of elevated BC concentration layers in the mixing layer. In the nighttime, if weak vertical uplift is well-matched with the BC concentration or its vertical gradient, elevated BC concentration layers can be formed in the atmosphere. Compared with upper layers far from the ground, nighttime elevated layers are easier to form in lower layers near the ground because high BC concentrations or large vertical gradients are more likely to occur in the lower layers. Both cases facilitate the occurrence of large vertical upward transport rates of BC.


Introduction
Black carbon aerosol particles (BCs) are strongly absorptive atmospheric aerosol particles.
BCs absorb solar radiation and heat the surrounding atmosphere, thereby affecting the ground temperature. The impact of BCs on the ground temperature varies with the height of the heated atmosphere [1]. In the Arctic, for example, if the BCs are close to snow and ice, they may increase the ground temperature; if the BCs are located in the free atmosphere, they may decrease the ground temperature and increase sea ice [2,3]. The Intergovernmental Panel on Climate Change Fifth Assessment Report (IPCC-AR5) points out that the uncertainty in the estimated BC direct radiative forcing contributes to a very large proportion of the uncertainty in the estimated aerosol direct radiative forcing [4]. One of the reasons for the uncertainty in the estimated BC direct radiative forcing is that the vertical distribution of BCs is insufficiently understood [5]. In recent years, it has also been shown that BCs can cause heating in the boundary layer atmosphere (especially in the upper boundary layer atmosphere), which is accompanied with a reduction in the ground heat flux. These effects will inhibit the development of the boundary layer and increase the number of extreme haze pollution events. This process is called the BC "dome effect" [6][7][8].
Studies have shown that the vertical distribution of BCs is very important. Observational studies have shown that the BC concentration sometimes peaks at a certain height, which indicates the 3 appearance of an elevated BC concentration layer. Elevated BC concentration layers have strong radiation and dome effects [9]. Therefore, studies on the detection and formation conditions of elevated BC concentration layers are helpful for understanding the vertical distribution of the BC concentration and can provide a basis for the assessment of the climate impact of BCs and early pollution warnings in pollutant emission regions.
During a strong foggy period in autumn and winter, Chilinski et al. (2016) observed BC concentrations along a 100-m uphill trail on a slope of the Wisłok Mountain valley in Poland using a manually carried micro-aethalometer AE-51. The results show that the BC concentration generally decreased significantly with altitude. However, multilayer structures were also observed during nighttime inversion conditions [15]. Chilinski et al. (2016) also used unmanned aerial vehicles to measure up to the top of the boundary layer, sometimes reaching the free atmosphere.
As a result, many multilayer structures of BC concentrations were observed [15]. Ferrero et al.  Figure 1) is located on the edge of a high-pressure and a low-pressure system, respectively, and the weather conditions were fine and partly cloudy.
The surface wind speed was not high during the two observations.

Observation scheme
The King Air aircraft was selected for this observation. The King Air has a long-range fuel tank on its wing and a flight distance of 4000 kilometers. In this observation, an SP2 manufactured by Droplet Measurement Technologies, Inc. was used to observe BCs. The SP2 is the only precise instrument in the world that can directly measure the mass of black carbon in a single particle. The detection diameter ranges from 45 to 470 nm. The mass of BCs outside the detection range is estimated by fitting the measured data using a lognormal distribution. The mass concentration of BCs is obtained by dividing the total mass of BCs by the volume of the air samples. The sample flow rate was controlled at 120 cm 3 /min, and the temporal resolution for data collection was 1 s.
The BC mass concentration value was processed under standard temperature (288.15 K) and     Chemical Tracers-4 (MOZART-4) simulation results.
The physical and chemical process parameterization schemes used in this study are shown in Table 1. The photolysis rate required for photochemical reaction processes is calculated by the Fast-J method [29]. In the calculation process, the scattering and absorption of solar radiation by the atmosphere are considered. The photolysis rate is updated once per hour for the gas-phase    (Figure 7), the distribution of the FNL horizontal wind speed at 14:00 shows no peak at approximately 500m. Given the low spatial resolution of the FNL analysis data and the lack of horizontal distribution of BC concentrations, we tried to simulate the peak at 500m ASL and analyzed its causes using the WRF-Chem model. Peak Figure 8 shows the location of the flight area and the six nearby ground grid points in the simulation. Figure 9 shows the distribution of the simulated BC concentration with altitude at 13:00 on October 25 at these points. The simulated maximum value in the mixing layer is close to the observed results, but the altitude of the peak value is about 800m, higher than that in the observation (500m) as shown in figure 5b, and the elevated layers are not very distinct. The results obtained by using other simulation schemes are similar to this result. To investigate the relationship between advection transport and the peak in the upper part of the mixing layer, the advection transport rate is calculated. The horizontal advection transport rate and vertical advection transport rate are calculated by the following equations:

Analysis of airborne detection results
In Eqs. (1) and (2) Figure 11 shows the profiles of the horizontal advection transport rate, the vertical advection transport rate, and the total advection transport rate (which is the sum of the first two) of the BC concentration at grid point (97, 76) at 13:00. It can be seen that the total advection transport rate was negative at altitudes above 850m and positive at altitudes below 850 m. Except for certain altitudes, the horizontal advection transport rate was greater than the vertical advection transport rate. The total advection transport rate shows a peak at approximately 650m and 800m, but the BC concentration only shows a peak at 800m. This result could be related to the fact that the former has greater turbulent energy than the latter (Figure 10b). The simulation results of this case show that compared with the upper layer of the mixing layer, the turbulent energy in the lower and middle layers is larger, and there is no particularly prominent peak advection transport. Therefore, the simulation results do not show the peak value of BC concentrations at approximately 500m.

Weak upward motion and the formation of elevated BC concentration layers
When we calculated the vertical advection rate above, we found that in the daytime respectively. Figure 13 also shows that the three elevated BC concentration layers, A, B, and C, all corresponded to a relatively low BC concentration on the ground. Layers A, B, and C were all accompanied with a weak vertical upward flow, but not all weak vertical upward flow areas corresponded to elevated BC concentration layers. Further, no elevated BC concentration layer appeared in downward flow areas.