Based on the data of AQI and wind profiler for the whole year of 2015, the daily variation characteristics of boundary layer under severe polluted weather conditions in Beijing were analyzed. The results show that: (1) Wind speed is very small at severe pollution weather, especially near-surface wind speed is nearly 0 m/s. Vertical wind velocity is evenly distributed. Layer is gentle and boundary layer is stable and almost undulating. (2) The existence of high concentrations of pollutants blocked the transmission of solar radiation on the ground, while the surface of the long-wave radiation is difficult to spread, it will be accumulated to the development when upward to a certain extent. The lack of thermal power makes the boundary layer more stable. The stable boundary layer is not conducive to the spread of pollutants, and the formation of a stable boundary layer is conducive to pollution weather maintenance. This process forms a vicious circle of mutual promotion between pollutants and stable boundary layers. (3) The dynamical effect of the wind field is insufficient, the thermal turbulence plays the leading role, and the turbulence intensity is much smaller than that of the clear sky.

Raindrop size distribution (DSD) plays an important role in microphysics of clouds and precipitation. A case study about a heavy rainfall with squall line characteristic passing through Beijing is presented by analyzing the DSD observation data. The variability of DSD and induced rainfall parameters in different rain regimes, namely initial convective rain, transition and trailing stratiform, is investigated. Compared with the transition and trailing stratiform region, leading convective line is characterized by large raindrop concentrations(>4600 counts/m3), large mass-weighted mean diameter(2.69 mm), and maximum drop size(6.3 mm). Convective and transition regime have the similar DSD with standard deviations(>0.3 mm), wider than stratiform region(>0.24 mm). The empirical Z-R relationships are conspicuously different between different rain phases.

Compared with the tropospheric NO2 column concentration measured by the Monitoring Instrument Ozone(OMI) detector mounted on the AURA EOS satellite, the mobile passive differential optical absorption spectroscopy (DOAS) has higher spatial resolution. It can better reflect the distribution of NO2 pollution in a region. Therefore the distribution of NO2 column concentration over Shijiazhuang was reconstructed using data from OMI database(June 2013), combined with the wind data, the variation of spatial and temporal patterns of the distribution of NO2 concentration under southwest wind in Shijiazhuang was obtained. At the same time, the mobile DOAS system was used to observe the distribution of NO2 concentration on the southwest transport channel (Shijiazhuang-Baoding-Beijing). Finally, the exponentially-modified Gaussian (EMG) method was used to estimate the NOx emission rate over Shijiazhuang in June 2013. The results showed that, the emission fluxes of OMI NOx and emission fluxes modified by mobile DOAS was 195.8 mol/s and 160.6 mol/s, respectively. The reason why the latter is less than the former may be the low spatial resolution of the satellite.

Global warming, climate change and other issues caused by greenhouse gases have been the hotspot of international concern. CO2 is one of the important greenhouse gas, monitoring and controling of CO2 have been the focus of all countries. Based on the spectral absorption structure of CO2 in the 1.6 μm, the retrieval algorithm applied to obtain the CO2 column information from spectroscopic measurements was researched by using the weighted function modified difference absorption spectroscopy (WFM-DOAS) method. Based on atmospheric radiative transfer model, effects of the parameters on the sensitivity of the weighted function (WF) calculation were studied and simulated. The influence of different parameters on CO2 WF coefficient is calculated and analyzed in detail, including observation height, the solar zenith angle, solar azimuth, surface albedo, spectral resolution and so on. And based on the sunlight spectrum of zenith direction, the performance of the instrument, CO2 vertical column concentration and disturbance, and CH4 and H2O vertical column concentration were analyzed. And a preliminary analysis of the inversion error is better than 1%.

The vertical column density (VCD) of tropospheric SO2 was obtained by using multi-axis differential optical absorption spectroscopy (MAX-DOAS). And differential slant column density (dSCD) of SO2 was got by using different reference spectra and different wave bands. By comparison, the error and the variation of error was minimum when using the sequential reference, and the error was less than 5%. Six wave bands were selected and compared to determine the optimal retrieval wave band which was 307.5～315 nm. Combined with meteorological data, a contamination process from 14th to 18th, October, 2015 was studied. The data analysis showed that wind speed and direction were two important factors affecting the concentration of SO2 in monitoring site. The SO2 produced from city and power plants was transported to monitoring site under the influence of east and south wind. According to the research, MAX-DOAS can accurately inverse the atmospheric SO2 VCD, which has great significance for exploring the tropospheric SO2 VCD of the city, satellite calibration, model validation and studying pollution transport.

Total hydrocarbons (THC) of vehicle emissions is the major cause of haze and photochemical smog, so THC monitoring must be one of the goals in the investigation of urban air pollution and research of effects on human health. The design of the THC measurement system based on hydrogen flame ionization detector (FID) for vehicle exhaust is developed and comparative experiments are carried out. The working principle of the FID for measuring THC is introduced, and the THC measurement module of vehicle exhaust is designed according to the measuring principle of FID. A micro-current amplifier circuit is designed for detecting the ion signals to meet a wide range, high-precision and micro-current signal testing requirements, and the peak signal of the THC is obtained. Moreover, the THC peak signals are analyzed qualitatively and quantitatively. The concentration of the THC are calculated by standard curve method. The comparative experiments are carried out by using SENSORS-FID THC analyzer of the United States, and the results show that the measurements of the THC measurement module is higher, and the measurement error is about 4.44%～8.43%, this systematic error is primarily caused by oxygen peak interference.

To meet the construction requirements of environmental monitoring and early-warning system, the environmental automatic testing system which based on the internet of things and optoelectronic technology was designed. Combined embedded technology, full-spectrum detection technology and e-tongue inspection technology, the online testing on a number of water quality parameters, such as pH, conductivity, ammonia nitrogen, turbidity, chroma and COD, et al. could be implemented successfully. This system could evaluate the change of water quality. According to the performance of the system verified by series experiments, the accuracies of pH, conductivity, ammonia nitrogen, turbidity and COD measurement were -0.02, 0.4%, 2.9%, 2.1%, 1.1%, respectively. The detection limits of conductivity, turbidity and COD measurement were 0.7 μS/cm, 0.9 NTU and 0.5 mg/L, respectively. The results of the actual water samples for comparison between the system and standard methods were acceptable. The consequence could satisfy the demand of on-line water quality monitoring.

Based on fast gas chromatograph and time-of-flight mass spectrum technology, fast gas chromatography/electron impaction source of time-of-flight mass spectrometer (FGC/EI-TOF MS) is developed. The FGC/EI-TOF MS is mainly composed of a sample system, a GC separation system, a GC-MS interface, an electron impaction source, a vertical introduction of reflection time-of-flight mass analyzer, a data acquisition system and hardware control system. A vaporization chamber, heating film of GC and three-wall-tube line are built. After the optimization of EPC, temperature of GC, and important voltage parameters, Alkanes standard gas mixtures are studied, and fast seperation and accurate classification different kinds of organic compounds are eventually realized rapidly and accurately. At last, the FGC-TOF MS is applied to test the characteristics of benzenes and isobutane standard gas mixtures. The results indicate that mass range is 4～502 amu, resolution is better than 500, detection limit is 500 ppb, analysis time is less than 5 min, reproducibility RSD is less than 10%, dynamic range is four orders of magnitude.

A single photon ionization time-of-flight mass spectrometry(SPI-TOFMS) was applied to the online monitoring of the volatile organic compounds (VOCs) emitted from a plastic company in Pearl River Delta. The source profiles of workshops and sewage treatment station were attained successfully. Over-standard concentration of VOCs has not been found during the monitoring. In addition, the contribution of every pollution source to the internal roads was also evaluated by comparing the source profiles of workshops. This online VOCs mass spectrometry has wide markets in aspects of online VOCs monitoring, emergency detecting, pollution mode recognition and underlying discharge monitoring.

The unmanned aerial vehicle (UAV) platform has broad prospects in a variety of forest fire monitoring and warning means, with characteristics of less operation cost, simple operation, flexibility and so on. Based on infrared detector for detection and warning of the fire point, a forest fire warning system was presented with low-cost rotor UAV as a platform, using the Beidou receiver module for UAV positioning. Theoretical calculation and practical examples were given to verify the practicability of this scheme. The results show that the system can locate the fire point quickly and accurately without personal real-time monitoring, which provides a solution for the prevention of forest fire and rapid fire fighting command.