A Comprehensive Research of Atmospheric Haze by Optical Remote Sensing in Central China Region (CCR)

pollution gas on the measuring route can be acquired by combing the information from GPS on longitude and latitude. affected by the downdraft in leeward slope terrain, leading to the convergence of the regional weak wind field pollutants, and finally to heavy pollution.


Introduction
With the changing pollution scenario, the task of improving air quality is now faced with increasingly serious challenges. Aerosols and gaseous pollutants have great impact on the radiative balance of the Earth, and thus on climatic changes. Furthermore, the radiative forcing of natural and anthropogenic aerosols exhibits a strong seasonal and geographical variability, which can affect the diffusion of atmospheric pollution locally [1]. The increase of aerosols or gaseous pollutants results in impairment of visibility. Besides, high aerosol concentration is deleterious to human health. The Chinese Academy on Environmental Planning considers that air pollution is responsible for 411,000 premature deaths in China in 2003, primarily from respiratory problems and heart diseases [2]. Therefore, obtaining information on the vertical distribution of aerosols and gaseous pollutants in the pollution haze is crucial to understanding their optical characteristics as well as their impact on air quality and on human health.
During the past 10 years, with its dense population, rapid economic growth, and dramatic rate of urbanization, China has been experiencing elevated levels of air pollution [3][4][5]. This is particularly so in the case of the Pearl River Delta (PRD), Yangtze River Delta (YRD), and Central China Region (CCR). Among these, the CCR region is where the frequency of occurrence of haze has been increasing rapidly from 2012. In CCR, NOx, SO 2 and VOCs discharged by pollution sources are the significant primary pollutants, which greatly influence the formation of particulate matter and ozone because of photochemical reaction. Besides, the prevalence of mild and humid climate throughout the year in CCR leads to transformation and accumulation of particulate matter and gaseous pollutants. In addition, regional transportation contributes to combined pollution problems of atmosphere, manifested by high-concentration fine particulate matter, low visibility, highconcentration NO 2 column, etc.
Thus far, extensive studies using different methods have been conducted in China on atmospheric pollution [6][7][8][9]. For a better understanding of aerosol properties and their effect on environment, two major aerosol experiments were conducted in mainland China since 2004: (i) East Asian Study of Tropospheric Aerosols -An International Regional Experiment (EAST-AIRE) [10] and (ii) East Asian Study of Tropospheric Aerosols and Impact on Regional Climate (EAST-AIRC) [11]. Information on the properties of aerosols and their temporal and spatial distribution was obtained from studies using both short-term intensive field experiments and long-term operations, as well as modeling studies. Besides these filed experiments, lidar (LIght Ranging and Detection) and DOAS (Differential Absorption Spectroscopy) studies were conducted at several sites in China [12][13][14][15][16][17][18]. However, knowledge on the properties of aerosols and trace gases, particularly in the vertical dimension, in CCR China is rather scanty. Also, because of lack of real-time data on the quantum of pollutants discharged by major industrial parks, it is hard to resolve the task of pollution prevention. The present research was designed to take advantage of advanced environmental monitoring technical equipment (including Lidar [19,20], multi-axis DOAS [21,22], vehicle-mounted DOAS /infrared SOF [23][24][25], and meteorological parameter analyzer, etc.) to establish a stereoscopic observing system for key sources of regional pollution in CCR, China. The campaign was carried out from Dec 2013 to April 2014 in three key cities, namely Anyang (AY), Jiaozuo (JZ) and Sanmenxia (SMX). Continuous and automatic stereoscopic monitoring of primary pollution sources, transportation of pollutants, emission characteristics of pollution from major industrial parks, as well as the transport conditions in the surrounding regions of pollution under specific meteorological conditions was carried out to obtain data that can help in air quality improvement and in development of forecast warning model, besides making up for the deficiency in current ground monitoring system.
The scope of research resolves mainly into two parts: 1. Real-time monitoring of spatio-temporal distribution of regional haze pollution: Representative monitoring points were identified in selected cities and surrounding areas along the direction of pollution transportation to measure spatial and temporal distribution characteristics of SO 2 , NO 2 and aerosol particulate matter with lidar, MAX-DOAS, etc.

2.
Real-time monitoring of disorganized pollutant discharge in key industrial park sources: VOC is an important precursor to the formation of photochemical smog. The type and volume of VOC discharge, as also of SO 2 and NO 2, play a significant role in pollution. Aiming at the key source areas of elevated discharge, vehicle-mounted DOAS and vehicle-mounted SOF-FTIR were utilized for mobile scanning measurement of discharged smoke plumes. Using the absorption characteristics of pollution gas molecules in infrared and ultraviolet spectrum, the concentration of exhaust gas was qualitatively and quantitatively determined through SOF-FTIR and DOAS spectrum analytical methods.

Lidar
The Lidar system, used for the study of spatial and temporal distribution of atmospheric particles, is shown in Fig 1; it adopts 532m Mie scattering and 532 polarization channels. Laser transmitter launches linear polarization lasers at 532nm by shooting vertically into atmosphere through beam expander, inversed mirror and quartz skylight. The laser pulse is scattered through air molecules, atmospheric aerosols or cloud, on its transmission route. The backwardscattered light firstly passes through skylight glass and then as parallel light through the diaphragm hole and convergent lens of the telescope. The converged parallel light is split into two beams by dichroic mirror: one is at 532nm, received by photomultiplier C after passing through optical filter; the other is at 355nm, whose vertical and parallel components are received respectively by A and B photomultipliers, after decomposition by an analyzer prism. After the conversion of optical signal into electrical signal by photomultiplier, the signal will be delivered to data acquisition system, where its ASCII code data file will be recorded and saved by main control computer.

MAX-DOAS
MAX-DOAS is a special type of DOAS technology, which uses characteristic absorption structure of trace gases (such as NO 2 , SO 2 , O 3 ) in visible ultraviolet band for qualitative and quantitative analysis of gas. As a real-time and online measuring method, it is eminently suitable for remote sensing measurement of air pollution. Besides, the data on vertical column density of pollution gases acquired from this method can help in verifying satellite results [26]. MAX-DOAS receives scattered sunlight from telescopes at different elevations, as shown in Fig 2. Based on the characteristic absorption spectrums of pollution gas, it inverses the vertical column density of pollution gas and distribution profile of density. Besides, it acquires data on spatial distribution of pollution gas in atmosphere by measuring atmospheric absorption spectrums in zenith direction and several off-axis directions. During the monitoring process, the reflector orients itself to the north and is then driven by stepper motor to turn to 1°, 2°, 3°, 4°... 90° to finish one measurement cycle, thus realizing measurement of different angles. The schematic diagram of MAX-DOAS is presented below.

Vehicle-mounted DOAS
With zenith-sky scattered light as the light source, vehicle-mounted DOAS technology is a kind of spectrum detection technology, which uses "fingerprint" absorption characteristics of optical radiation by gas molecule for qualitative and quantitative measurements of gas. The system includes mainly a spectrograph, a GPS system and a telescope [27]. As shown in Fig  3, the whole system is put on mobile platform and the telescope gathers solar spectrum along the measurement route. During operation, the band of spectrograph is fixed at 290nm~420nm (spectral region 130nm) as this band has strong SO 2 and NO 2 absorption structures. After the spectrograph delivers the gathered spectrum to computer for data processing, DOAS method is used to calculate the column density of pollution gas. Finally, density distribution of pollution gas on the measuring route can be acquired by combing the information from GPS on longitude and latitude.
4 spectrum detection technology, which uses "fingerprint" absorption characteristics of optical radiat by gas molecule for qualitative and quantitative measurements of gas. The system includes mainl spectrograph, a GPS system and a telescope [27] . As shown in Fig 3, the whole system is put on mob platform and the telescope gathers solar spectrum along the measurement route. During operation, band of spectrograph is fixed at 290nm~420nm (spectral region 130nm) as this band has strong S and NO 2 absorption structures. After the spectrograph delivers the gathered spectrum to computer data processing, DOAS method is used to calculate the column density of pollution gas. Fina density distribution of pollution gas on the measuring route can be acquired by combing information from GPS on longitude and latitude.

Satellite
The spatial resolution of the OMI detector mounted on AURA satellite is 13 km×24 km [28] . N VCD of troposphere, below the satellite orbit, can be obtained through DOAS inversion of measu spectrum. After differential and meshing treatment, NO 2 VCD of each acquired orbit is integrated a then NO 2 density distribution over the region obtained. Acquisition of regional density distribut data is significant for investigation of the processes of generation and transportation of pollution. the present study, OMI Level 2 NO 2 VCD data was adopted. From the weekly average between N and Dec, it was found that CCR pollution reached its peak in the third week.

Stereoscopic monitoring of regular pollution gas
(1) Comparison of areal distribution at the three sites The average of the values obtained during measurement period at the three observation sites (AY, and SMX) was taken as the areal distribution level of NO 2 and SO 2 in CCR (Fig 4). It can observed that the average level of NO 2 density at AY is 63.4*10 15 molec./cm 2 , which is higher than values at JZ and SMX. Density of NO 2 in SMX is the lowest at its average level 35.3*10 15 molec./cm 2 . SO 2 , density at JZ is the highest at 149*10 15 molec./cm 2 , which is higher th that at AY and SMX.

Satellite
The spatial resolution of the OMI detector mounted on AURA satellite is 13 km×24 km [28]. NO 2 VCD of troposphere, below the satellite orbit, can be obtained through DOAS inversion of measured spectrum. After differential and meshing treatment, NO 2 VCD of each acquired orbit is integrated and then NO 2 density distribution over the region is obtained. Acquisition of regional density distribution data is significant for investigation of the processes of generation and transportation of pollution. For the present study, OMI Level 2 NO 2 VCD data was adopted. From the weekly average between Nov and Dec, it was found that CCR pollution reached its peak in the third week.

Comparison of areal distribution at the three sites
The average of the values obtained during measurement period at the three observation sites (AY, JZ and SMX) were taken as the areal distribution level of NO 2 and SO 2 in CCR (Fig 4). It can be observed that the average level of NO 2 density at AY is 63.4*10 15 molec./cm 2 , which is higher than the values at JZ and SMX. Density of NO 2 in SMX is the lowest at its average level of 35.3*10 15 molec./cm 2 . SO 2 , density at JZ is the highest at 149*10 15 molec./cm 2 , which is higher than that at AY and SMX. (2) Results of monthly average variation The results obtained during the measurement period were used to calculate the averages of different months and to analyze variation trends of pollutant concentration. It can be observed from the calculation results (see Fig 5) that NO 2 VCD (vertical column density) reached its peak in Dec and Jan, decreased from Feb onward, and then remained low between Mar and June. Monthly average of SO 2 VCD at each station is shown in the figure below. It is obvious from the calculation results that, just as NO 2 , VCD of SO 2 is high in winter; it reached its peak in Dec at AY, in Jan at SMX, and in Feb at JZ. It obviously decreased from Mar and thereafter remained low up to June. From an overview of the monthly averages discussed above, it can be concluded that the accumulation of pollutants had been rapid and heavy because of high discharge and unfavorable

Results of monthly average variation
The results obtained during the measurement period were used to calculate the averages of different months and to analyze variation trends of pollutant concentration. It can be observed from the calculation results (see Fig 5) that NO 2 VCD (vertical column density) reached its peak in Dec and Jan, decreased from Feb onward, and then remained low between Mar and June. Monthly average of SO 2 VCD at each station is shown in the figure below. It is obvious from VCD of SO 2 is high in winter just as NO 2 ; it reached its peak in Dec at AY, in Jan at SMX, and in Feb at JZ. It obviously decreased from Mar and thereafter remained low up to June. (2) Results of monthly average variation The results obtained during the measurement period were used to calculate the averages of different months and to analyze variation trends of pollutant concentration. It can be observed from the calculation results (see Fig 5) that NO 2 VCD (vertical column density) reached its peak in Dec and Jan, decreased from Feb onward, and then remained low between Mar and June. Monthly average of SO 2 VCD at each station is shown in the figure below. It is obvious from the calculation results that, just as NO 2 , VCD of SO 2 is high in winter; it reached its peak in Dec at AY, in Jan at SMX, and in Feb at JZ. It obviously decreased from Mar and thereafter remained low up to June. From an overview of the monthly averages discussed above, it can be concluded that the accumulation of pollutants had been rapid and heavy because of high discharge and unfavorable From an overview of the monthly averages discussed above, it can be concluded that the accumulation of pollutants had been accumulated rapidly and heavily because of high discharge and unfavorable weather conditions in winter. With the end of hot season after

Current Air Quality Issues
February, change of wind direction and increase in rainfall would start favoring diffusion and sedimentation of pollutants, and consequently reduction of pollutant concentration.

Stereoscopic monitoring of aerosols
Monitoring of aerosols through lidar, from Nov 2013 to April 2014, (Fig 6) revealed that haze pollution was severe mainly from Nov. 2013 to Jan. 2014. Particularly from Dec 12 to Dec 20, and Jan 4 to Jan 17, the pollution lasted for almost half a month at AY and JZ. Overall air quality improved continuously, from Feb to June 2014, at AY, JZ and SMX. However, during wheat harvesting in June, because of straw burning, pollution levels increased at AY, JZ and SMX sites. The highest pollution levels were confined to the middle of June (14 th -15 th ); the peak of aerosol extinction coefficients at AY, JZ and SMX sites are respectively 0.83km -1 , 0.91km -1 and 0.66km -1 .
6 weather conditions in winter. With the end of hot season after February, change of wind direction and increase in rainfall would start favoring diffusion and sedimentation of pollutants, and consequently reduction of pollutant concentration. ) reveal that near-ground pollution at these three sites came mainly from local pollution (industrial emission, motor vehicle exhaust and coal combustion) and fire coal transportation to some northern cities in the southeast direction. By the end of Jan 2014, local pollution aggravated because of discharge from firecrackers during spring festival. The pollution at higher levels in the air above 1km came mainly from the northwest.

Stereoscopic monitoring of aerosols
In conclusion, it emerges that particulate pollution occurred almost simultaneously at the three sites AY, JZ and SMX. Local pollution (local source) is typical of AY, and near-ground pollution of JZ, where higher pollution levels because of industrial emission are also common. Local pollution was In conclusion, it emerges that particulate pollution occurred almost simultaneously at the three sites AY, JZ and SMX. Local pollution (local source) is typical in AY, and near-ground pollution in JZ is higher due to the local industrial emission. Local pollution was relatively slighter in SMX, but it can be easily aggravated by northwest sand dust transportation. Sand dust pollution at AY manifests as superposed locally raised dust over northwest sand dust. In CCR, the ground weather conditions of uniform-pressure were conducive to pollutant diffusion. This weather had lower ground pressure field, lower ground and low-altitude wind speeds, and even calm wind sometimes. These were often accompanied by stronger radiation inversion or low-altitude temperature inversion that is difficult to eliminate. The structure of lower atmosphere layer was stable and might contribute to higher pollution.

Vehicle-mounted monitoring of pollutant discharge in some representative cities and key industrial parks
Quantitative modeling was done in some cities to evaluate the influence of key industries and industrial parks on ambient air quality, especially PM10. Vehicle-mounted DOAS system was used for monitoring industrial parks in both winter (Dec 2013) and summer (May-June, 2014) by way of acquiring data on spatial distribution of urban VOD of SO 2 and NO 2 , emission characteristics of SO 2 and NO 2 in key industrial sources, as well as pollution transportation parameters under specific meteorological conditions.
At AY, the results obtained in winter and summer is similar. The concentration of SO 2 and NO 2 was caused mainly by the discharge from steelworks factories. However, observations during southeast/south winds in summer show that the concentration of SO 2, in the southwestnorthwest section of the measurement route, was high. This indicates that sporadic small-scale industrial discharge occurred beyond southwest-northwest section (south and southwest of AY), which was not found during observations in winter. Judging from the relative geographic positions of steelworks and urban area at AY, as well as forward track model of wind field (Fig 7), it can be concluded that during northwest winds, discharge of AY steelworks had great influence on the air quality of urban area in AY.
Mean values per hour (MV/H), obtained from ground monitoring of SO 2 at monitoring point of environmental protection agency in AY from May 18 to May 20 are shown in Fig 8. From these results, it can be observed that ground concentration of SO 2 in urban area increased during northwest winds. At the same time, it is also found that during southeast winds, ground concentration of SO 2 in urban area was high at about 11a.m. on May 18, when the observation time was just the same as that of third auxiliary chart in Figure 9. It is concluded that, under southeast wind field, sporadic pollution in the southwest and south, beyond measurement area, had great influence on the air quality of main urban area in AY.

Analysis of the typical haze process
Haze episode (Dec 12, 2013-Dec, 16,2013) From Dec 12, 2013 to Dec 16, 2013, a typical haze phenomenon was observed at AY, JZ and SMX, when the pollution was mainly local. Under the influence of regional weak wind field, the distribution range of local pollution was large and the duration long, but intermittent. Besides, the extinction coefficient of aerosol and concentration of NO 2 reflect a completely accumulative, rising, sustaining and stable process. Based on the distribution range and strength, two main phases were identified: generation phase of local pollution and pollution stabilization phase.

Aerosols
AY city: Since the afternoon of Dec 12, 2013, the pollutants gradually rose to 1.9 km and then began descending on Dec 13 (Fig 10). In the afternoon of Dec 14, it again started rising, but this time the height was lower than that on Dec 13. During Dec 15-16, near-ground pollution intensified and lasted for a long time, with the maximum value of extinction coefficient reaching 0.87km-1 (see Fig 11).
a. Formative period of haze (Dec 12-14) b. Stable formation of pollution (Dec 15-16) JZ city: The height of pollutants started rising from the afternoon of Dec 12 and the average of extinction coefficient was higher than that at AY (Fig 12). The height of daytime pollutants dropped on Dec 13, but it began rising again in the morning of Dec 14; near-ground pollution continued on Dec 15-16 (Fig 13). SMX city: Local pollution was noticed from the afternoon of Dec 12; the height of the pollutants rose gradually from the evening of Dec 13 and the average of extinction coefficient was 0.39km -1 (Fig 14). High concentration pollution was confined to the period from the evening of Dec 13 to evening of Dec 14. No obvious change of extinction coefficient profile is observed on Dec 15-16; the height of the pollutants was below 1.2km (Fig 15).

Meteorological conditions
On Dec 15, the average wind speed was 4.1 m/s and the temperature was1.8°C at the height of 1000m. The prevailing wind was northwest wind (Fig 18). The high-altitude west wind was affected by the downdraft in leeward slope terrain, leading to the convergence of the regional weak wind field pollutants, and finally to heavy pollution.
owndraft in leeward slope terrain, leading to the convergence of the regional weak wind fie ollutants, and finally to heavy pollution.

Figure 18
Meteorological conditions and wind rose diagram From the HYSPLIT backward trajectory model in Fig 19 [ 29], the pollution at the three observation points was found to be mainly from the northwest. The main pollution source could be from burning of fossil fuels or bio-fuels, aided by the continental airstream. By noon on Nov 25, there was evident increase in pollutants at both AY and JZ. Subsequently, until the evening on Dec 26, at both the observation points, the pollutants decreased and then again increased and stabilized. The overall aerosol extinction coefficient at AY was higher than that at JZ, and the peal values are 0.87km -1 and 0.51km -1 , respectively. The pollution layer was below 1.9 km, corresponding to the pollutant transmission layer in a low level jet (LLJ) [30] of stable or neutral boundary layer. For two days, it was observed that the height of the mixing layer grew linearly over time. When the height of the mixing layer grows to the height of the pollutant transmission belt in the nocturnal boundary layer, the vertical mixing effect would result in rapid concentration of pollutants on the ground. Therefore, there was increase in aerosol values, as well as in local pollutants discharge during the days; the downward transmission from the higher layers was also an important factor. Compared with AY and JZ, the air at SMX was cleaner and the height of pollutants was lower at 1.2 km below, but by the evening of June13 the pollution became serious.

Satellite
From the satellite results in Fig 20, it can be observed that high NO 2 VCD values existed during the haze period. From the daily variations in NO 2 VCD values between Nov and Dec at AY, JZ and SMX (seen in Fig 21), it can be seen that the results at AY clearly indicate a relation between NO 2 VCD values and haze. For example, at AY, the daily average value of NO 2 VCD during Nov 1-20 was 3.32×10 16 molec./cm 2 . For the first haze period during Nov 21-23, the daily average value was 8.37×10 16 molecule/cm 2 , which was 2.5 times more than the normal average value. For the second haze period on Dec 8, it was 5.8×10 17 molecule/cm 2 , which was the highest and 18 times more than the normal average value. For the third haze period during Dec 16-17, it was 1.57×10 17 molecule/cm 2 , which was 4.7 times more than the average value. From the values of NO 2 during these three haze periods, which were multiple times more than the average value, it clearly follows that there exists certain relation between NO 2 and haze, and that high NO 2 VCD values contribute to hazy weather.  Fig 21), it can be seen that the results at AY clearly indicate a relation between NO 2 VCD values and haze. For example, at AY, the daily average value of NO 2 VCD during Nov 1-20 was 3.32×10 16 molec./cm 2 . For the first haze period during Nov 21-23, the daily average value was 8.37×10 16 molecule/cm 2 , which was 2.5 times more than the normal average value. For the second haze period on Dec 8, it was 5.8×10 17 molecule/cm 2 , which was the highest and 18 times more than the normal average value. For the third haze period during Dec 16-17, it was 1.57×10 17 molecule/cm 2 , which was 4.7 times more than the average value. From the values of NO 2 during these three haze periods, which were multiple times more than the average value, it clearly follows that there exists certain relation between NO 2 and haze, and that high NO 2 VCD values contribute to hazy weather.

Typical straw burning process analysis (June 12-16, 2014) (1) Aerosols
Wheat harvesting from June 12 to June 16 in early summer of 2014, straw burning contributed to local pollution at all the three observation points in AY, JZ and SMX (Figs 22 & 23). At all these points, pollution was at its peak during June 14-15. That the burning of biomass, such as straw, releases particulate matter of small grain size is confirmed by the depolarization ratio values obtained at the three measuring points.     Besides particulates, usually straw burning also releases various gaseous components such as CO 2 , CO, NOx, VOC, sulfuric acid gas, hydrochloric acid gas, and alkaline gas ammonia. The retention time of ammonia in the air is short, because it quickly reacts with acidic gases or acidic particulate matter, and the resulting components can be precipitated on the particle surfaces of the particulate matter leading to the generation of new condensation nucleus. The aging process will increase the amount of particulate matter, which can lead to the increase in particulate emission factors. The burning of straw near CCR, together with surrounding pollution transmission, aging of particulate matter during long distance transmission, and the emission factors such as diameter aggravated local pollution.

(3) Fire spots and backward trajectory
To identify the sources of pollution transmission, local information was taken as the reference, but during the wheat harvesting season, the burning of straw on the surrounding farmlands was believed to be the main cause for high values of pollution. Therefore, the analysis was made by referring to the satellite fire spots diagram of Ministry of Environmental Protection, which is shown in Fig 25.

Fire spots and backward trajectory
To identify the sources of pollution transmission, local information was taken as the reference. But during the wheat harvesting season, the burning of straw on the surrounding farmlands was believed to be the main cause for high values of pollution. Therefore, the analysis was made by referring to the satellite fire spots diagram of Ministry of Environmental Protection, which is shown in Fig 25. From these fire spots [31], it was inferred that multiple fire points existed around AY and JZ since June. However, by the middle of June they increased and many more fresh points appeared in the agricultural area of south CCR in the southeast. The backward trajectories, shown in Fig 26, indicate that concentrated burning of wheat straw brought about peripheral transmission.
From the foregoing results of aerosols and pollution gas, it emerges that the transportation brought by wheat straw burning manifested mainly as rapid increase of PM10, which is related to moisture absorption and expansion of particulate matter during transportation process. NO 2 and SO 2 were at their peaks on June 13 and 14 respectively; so, the amplification was not Current Air Quality Issues 268

(3) Fire spots and backward trajectory
To identify the sources of pollution transmission, local information was taken as the reference, but during the wheat harvesting season, the burning of straw on the surrounding farmlands was believed to be the main cause for high values of pollution. Therefore, the analysis was made by referring to the satellite fire spots diagram of Ministry of Environmental Protection, which is shown in

Conclusions
CCR is an important industrial site in the mid-west region of China. Historical review of satellite data shows that the cities with relatively serious air pollution are confined mainly to the north because of their geographic location, weather conditions and industrial activity. Besides, the "accumulation zone" of pollutants exists in the region east of Tai-hang Mountains. Analysis of the data obtained from stereoscopic monitoring of the selected sites shows that the prevalence of pollution at the three sites, AY, JZ and SMX, was almost simultaneous. However, AY was characterized by local pollution (local source) and JZ by near-ground pollution, where the pollution concentration was also higher due to higher emission of industrial gases. Though local pollution was milder at SMX, it still can be easily aggravated by northwest sand dust transportation. Through statistical analysis of satellite and ground monitoring data, pollution was more serious during the 1st and 4th quarters (winter and spring), followed by the 2nd quarter and the 3rd quarter. During the heating period, the air pollution intensified obviously, because the urban energy consumption structure was still coal-oriented. During late May to mid-June, straw burning in the south CCR and surrounding provinces, such as Anhui and Jiangsu, contributed to the increase of aerosol, NO 2 and SO 2, in some cities of CCR.
As for the reasons for the complex process of haze formation, it was found that local pollution (local sources) was the main cause, besides the effect of superposition of pollution transportation in some time frames. Near-ground pollution at the three sites was caused mainly by local pollution (industrial emission, motor vehicle exhaust, coal combustion, etc.) and pollution transportation of some northern cities in the southeast direction. AY was more affected by local transportation and emissions from some industrial parks, and JZ by near-ground local pollution. Moreover, pollution concentration was higher at JZ than that at AY because of more industrial emissions. Local pollution at SMX was comparatively less obvious, but was lasting longer due to the unfavorable diffusion condition. In CCR, formation of ground weather situations of uniform-pressure field and North China low pressure, which are bad for pollutant diffusion, is easy. Such weather has lower ground pressure field, and smaller speeds of ground and low-altitude winds, which are sometimes calm. It is often accompanied with stronger radiation inversion or low-altitude temperature inversion that is difficult to eliminate. The structure of lower atmosphere layer is stable and might contribute to higher pollution.
So far many efforts were made from ground based and space borne observations to study aerosol distribution and properties, along with model simulations, but such integrated studies are still limited over the East Asian, especially in CCR region in China. Satellite observations provide time constrained observations of aerosol vertical distributions but ground based lidar can give information on the evolution of the distribution over a location. Thus the optical properties of aerosols or gaseous pollutants obtained from ground-based remote sensing technology, especially lidar and DOAS is being studied to determine the aerosol optical characteristics and source origin. In the future, the more integrated combination of lidar, satellite data, in-situ instruments, and sunphotometer data will help us to improve the characterization of a variety of aerosol events and get deeper insight into the mechanism of long-range transportation of pollution in China. Furthermore, the information of vertical distribution of haze layer during long-range transportation could permit us to make better assessment of the effect of aerosols on the radiative forcing and climate change.