Bioaerosols dispersed in the atmosphere can adversely affect human health and the natural environment. Ultraviolet laser-induced fluorescence (UV-LIF) serves as an effective technique for bioaerosol detection. While current research primarily focuses on total fluorescence signals, the analysis of individual components within mixed spectra requires further investigation. This study proposes a method combining Random Forest (RF) with Genetic Algorithm (GA) to analyze mixed fluorescence spectra obtained from UV-LIF lidar. First, a RF model integrated with Bayesian optimization algorithms was developed to quantify the proportions of individual components in mixed fluorescence spectra. Subsequently, GA was employed to optimize the discrimination threshold for component identification. The fluorescence spectra included: Bacillus subtilis (BS), meadow oat pollen (MO), tobacco leaves (TL), and bovine serum albumin (BSA). Three-component mixed spectra with varying compositions and a four-component mixed spectrum were analyzed. RF quantitative model exhibited high predictive accuracy, with residual prediction deviation (RPD) values exceeding 3 for all components. The coefficient of determination (R-2) surpassed 0.9, and the maximum root mean square error (RMSE) remained below 0.082. After GA optimization, all average identification accuracies exceeded 92%. These results demonstrate the efficacy of the combined RF and optimization approach for spectral analysis. The method provides a valuable reference for quantitative and qualitative analysis of single bioaerosol components in the mixed spectra, as well as for complex spectral interpretation.

Real-time monitoring of atmospheric bioaerosol concentrations primarily involves analyzing the total fluorescence signal, which does not provide the specific content of each component. Since mixed fluorescence spectra contain information about the composition and content of bioaerosols, by employing multivariate analysis of the spectra and using multivariate regression algorithms, it is possible to extract the components from the mixed spectra and predict the proportion of each component. In this study, upon the extracted components in the mixed fluorescence spectra by vertex component analysis, we compared several quantitative analysis models for their accuracy in predicting bioaerosol proportions. The models can obtain the proportion of each component effectively when three components exist in the mixed spectra, with 2D convolutional neural networks (CNNs) having the best effect. To ascertain the algorithms' stability, discussions on scenarios involving multiple components (four-component and five-component mixtures) as well as low-frequency noise conditions have also been included. The degree of impact on Bacillus Subtilis is the biggest when the number of mixed components increased. For the CNNs, the interference caused by the increase in the number of components is relatively higher than that of low frequency noise. When the number of the component is increased, the prediction effect of the algorithms is affected with considerable low frequency noise. Principal component regression and back propagation neural networks can maintain a relatively stable and reliable prediction effect. It is possible to match the most suitable algorithms for different components, thereby obtaining more accurate measurement results and providing a certain reference value for the detection of bioaerosol concentrations.

Using UV lidar to induce bioaerosol fluorescence is an effective method for detecting atmospheric bioaerosols. In this work, vertex component analysis, combined with the genetic algorithm, is proposed to analyze the received fluorescence spectra from UV LIF lidar. The pure components in the mixed fluorescence spectra are extracted using the vertex component analysis method, and then the genetic algorithm is adopted to fit the proportion of each component in the mixed spectra. The echo signals of the UV LIF lidar are simulated using the Voigt function at the excitation wavelength of 266 nm. All components can be correctly identified. Low-frequency noise is one of the significant interference factors when non-biological particles are present in the atmosphere. For three mixed components—bacillus subtilis, escherichia coli, and meadow oat pollen—the genetic algorithm can predict the proportion of each component with high accuracy, even in the presence of low-frequency noise. The stability of the algorithm was also explored with mixtures of four and five components, along with low-frequency noise. The residual predictive deviation (RPD) for all cases ranged from 2.0 to 28.2. It was concluded that the genetic algorithm can effectively predict the proportions in the mixed fluorescence spectra.

Masson pine (Pinus massoniana Lamb.) is a dominant coniferous species in southern China, known for its rapid growth, abundant yield, and extensive utilization. Despite the robust adaptability of Masson pine and the rich annual precipitation in its distribution areas, this species still faces the mortality risk caused by the recurrent high temperatures in summer and low precipitation in subtropical regions. The mortality risk of Masson pine may increase in the future when facing a more frequent or intensive drought threat due to climate change. In this study, we conducted a manipulated drought experiment accompanying high temperature (~32.3 ± 0.7 °C in daytime and 28 °C in nighttime) to simulate a flash drought, aiming to explore the composite physiological response (hydraulic, gas exchange, and nonstructural carbon (NSC) characteristics) of Masson pine seedlings to extreme drought characterized by a high intensity and long duration. We found that, as the drought developed, the leaf water potential and gas exchange traits (net photosynthesis rate, stomatal conductance, and transpiration) significantly decreased while the percentage loss of hydraulic conductivity (PLC) significantly increased. In contrast, NSC remained a more constant trend before it was significantly reduced on day 30 after the beginning of the drought. Except for NSC, all the other traits had significant correlations between them. Additionally, hydraulic dysfunction indicated by the increasing PLC preceded the NSC depletion, which may indicate a more significant role for hydraulic failure than carbon starvation in drought-induced mortality. Conclusively, hydraulic and gas exchange traits showed a coupling response to drought, but NSC displayed an independent dynamic. The findings may improve our understanding of drought-coping strategies of Masson pine and provide some theoretical basis for Masson pine forest management.

Water is the only atmospheric parameter with three-phase state. An ultraviolet Raman lidar was developed for synchronous measurements for water vapor, liquid water and ice water in Xi’an University of Technology, Xi’an, China (34.233°N, 108.911°E). An accurate retrieval method on the basis of interference degree is proposed for synchronous three-phase water mixing ratio profiles. Preliminary measurements are carried out in the Laser Radar Center of Remote Sensing of Atmosphere (LRCRSA). Several representative examples are obtained and validated the performance of Raman system. Combined with atmospheric temperature profiles, the synchronous water vapor, liquid water and ice water profiling are retrieved and revealed the variation characteristics in three-phase water. The effective detection can reach up to a height of 5 km under cloudy weather, and synchronized growth in water vapor and liquid water content was obtained in cloud layers. Continuous observations are also made under haze weather condition, and the temporal and spatial evolution trend of three-phase water in clouds at 2 km altitude are successfully realized.

Poisson distributions have the characteristic of equality between their variance and mean values. By constructing a calculation model of the temporal variance and spatial variance, the relationship between the variance and mean values of lidar analog data and photon-counting data can be analyzed. The calculation results show that the photon-counting data from far field have the distribution property of equality between the variances and the corresponding mean values, while the analog data for the whole probing traces do not. In this paper, by analyzing the distribution properties of the spatial variance and temporal variance of lidar data, the dead time of photon-counting data was estimated, and the threshold voltage of the photon-counting system and the linear working range of photomultiplier tube were evaluated. The results show that the linear working range of the high voltage for the photomultiplier tube in the ultraviolet elastic scanning lidar is between −500 V and −1000 V, and the dead time and threshold voltage of the photon-counting system in the Licel transient recorder are 3.488 ns and 1.20 mV, respectively. Meanwhile, a novel gluing method between analog data and photon-counting data is presented, based on the calculation results of the variance distribution of lidar data. The linear transfer coefficients were determined by minimizing the differences between the variance and mean of the transformed photon-counting data in the near filed with high signal to noise ratio. The glued data were distributed to express the atmospheric conditions uniformly.

With the rapid development of precision automation and intelligent equipment industry, the need of miniaturized and high-precision grating displacement sensors is increasing. It is still a challenge for highly reliable mini sensors due to the limitations of current employed light sources, such as their sizes, features, etc. To develop novel grating displacement sensors, light source needs to be met the requirements. Based on the principles of grating signals, we developed a novel grating displacement sensor by employing a Honeywell infrared LED as the light source. The influence of the main parameters (e.g. divergence angle, width of light source, etc.) of the infrared LED light source on the grating Moire Fringes has been analyzed in detail. The intensity distribution of the Moire light field model has also been established to estimate the quality of the grating signals with new light source. Both mathematical modeling and experimental results verify the feasibility of the design showing two clear orthogonal sinusoidal signals. Compared with traditional light sources (visible point light source), the infrared LED light source can significantly shorten the grating optical structure, thereby facilitating the miniaturization of the grating sensor. Due to the transmission characteristics of long-wavelength light, the infrared source can both improve the contrast of the Moire Fringes and the quality of the grating signals. We believe that the designed grating displacement sensor will open a new phase in application with the advantages of infrared LED source, such as stable intensity, long lifetime and low prices.

In the detection of atmospheric temperature profile by Rayleigh scattering, the influence of Brillouin scattering is usually ignored and the accuracy of temperature detection is reduced. Current researches on Brillouin scattering are mainly focusing on hydrodynamic and Knudsen regime, few researches has been done on the kinetic regime. In order to improve the precision of atmospheric temperature measurement, a mathematical model based on three Gussian distributions was adopted to study the Rayleigh-Brillouin scattering spectrum (RBS) in kinetic regime, the mixed Rayleigh and Brillouin signal in atmospheric echo signal is separated to obtain independent Rayleigh and Brillouin spectrum. Finally, the experimental platform was set up to control simulation of atmospheric environment system and establishes a hyperspectral splitting optical system based on Fabry-Perot interferometer. The spectrum obtained by the experiment was used to optimize the mathematical model and improve the detection accuracy of atmospheric temperature profiles.

Aerosol optical depths (AODs) arc calculated for the Xi'an region based on the observation data of a sky radiometer (POM-02) and Beer-Lambert-Bouguer law. These AODs arc used as true values to compare with three inversion algorithms' results of AODs in the MODIS C061 product of NASA Terra satellite, and the accuracies and applicability of those three inversion algorithms for the Xi' an region arc discussed. Furthermore, we statistically analyze the spatial distributions and seasonal variation characteristics of AODs for the Xi' an region and its surrounding region using MODIS aerosol products. The results indicate the following: there is the best relationship between the AODs inverted from the MODIS DT&DB product and POM-02 data among those three MODIS inversion algorithms of Terra satellite, and its correlation coefficient is approximately 0.92. Thus, the MODIS DT&DB product is the most appropriate product for studying climate change and air pollution in the Xi'an region. Dust storms and human activities arc the main sources of aerosols in the Guanzhong Plain. The AODs arc greater in the eastern part and less in the western part of Guanzhong Plain, and the high-value centers arc mainly distributed in the regions such as Xi' an, Xianyang, and Weinan. The AODs in the Xi' an and Xianyang regions reach their maximum and minimums in the spring and autumn, respectively. However, the AODs from other regions in the Guanzhong Plain show apparently seasonal variation characteristics with greater values in spring and summer and lower values in autumn and winter.

To minimize the influence of laser speckle jitter from a non-ideal measurement environment on fluorescence-lifetime imaging microscopy (FLIM), a cross-correlation analysis method was proposed based on the changes in displacement in the fluorescence particle inverse calculation. The displacement of each pixel in sequential images was obtained from the displacement of the laser speckle, using a fast Fourier transform cross-correlation analysis. Through the use of an accurate time-gated time resolution, FLIM was applied to investigate the test point displacement affected by the laser speckle jitter. The fluorescence intensity of the test points was then replaced based on the intensity of the displacement points. The fluorescence lifetime can finally be calculated using the iterative deconvolution algorithm. Experimental results show that the average relative error of the inverted fluorescence lifetime is less than 30%, and the effective rate of the test points without the lifetime is greater than 18%. We believe that the proposed method can provide valuable guidance toward improving the accuracy of a fluorescence-lifetime measurement. (C) 2018 Elsevier Ltd. All rights reserved.

In order to perform hyperspectral remote sensing, we present a continuous tunable cavity Fabry-Perot interferometer (FPI) by using potassium dideuterium phosphate (DKDP) with two ring electrodes. DKDP has the good performances of high transmittance in the ultraviolet band and large aperture of 20 mm (the maximum aperture can be 100 mm). Since the resonant frequency of an FPI can be continuously varied with the refractive index change of DKDP caused by the electro-optic effect, the influence of moving parts on resonant frequency can be eliminated. Digital holographic interferometry based on a Mach-Zehnder interferometer is employed to measure the refractive index modulation of DKDP. The parameters of FPI are characterized by using an experimental setup with frequency locking and temperature control technologies. Taking the temperature-measuring high-spectral-resolution lidar based on Rayleigh-Brillouin scattering as an example, a continuous tunable cavity FPI with the full width at half-maximum of 200 MHz and free spectral range of 11.12 GHz is realized. The results are in good agreement with the designed parameters.

The objective of this study was to check whether different water and nitrogen treatments and, even the water-nitrogen coupling effect of plants could be correctly differentiated via chlorophyll a fluorescence image. We developed a classification method using the imaging analysis of chlorophyll a fluorescence induction based on Artificial Neural Network. The measurements were carried out on scheffera octophylla (Lour.) Harms, and the images were recorded at 690 nm with a high-resolution imaging device consisting of LEDs for an excitation at 460 nm and an Electron-Multiplying CCD camera. The effect of three different water and three different nitrogen treatments on the fluorescence parameters were obtained by hundreds of time-resolved fluorescence images. We used a Radial Basis Function neural network to model and test the sample data. The results showed that the different water and nitrogen statuses of plants were identified by the chlorophyll a fluorescence images and showed a high recognition accuracy. Compared with nitrogen, water had more of an influence on chlorophyll a fluorescence and was easier to identify. However, because the water and nitrogen restrict and promote each other, studying the coupling effect of water and nitrogen is necessary. Nine levels of water-nitrogen coupling plants were tested and classified. We discovered that a significant decrease on the classified accuracy was observed for the high nitrogen and low nitrogen treatments, while under a medium N-supply, the recognition rate was high. The method in this paper allowed plants to be classified under different water and nitrogen treatments, and has the potential to monitor the water and nitrogen coupling effect of plants in situ. (C) 2019 Elsevier Ltd. All rights reserved.

Variation of output laser pulse energy has great influence on the measured data of radar,so the output laser pulse energy must be measured accurately in real-time.Aiming at the problem that output laser pulse width is very narrow,and it is difficult to achieve high frequency acquisition circuit,a pulse energy monitoring system is designed based on the peak sampling and holding circuit.In situ monitoring of simulated pulses with pulse width of about 10 ns and repetition rates of 20,60,100 Hz is carried out.Results show that the system can accurately detect laser emitting energy without any pulse missing.There is a good linear relationship between the system output voltage and measured laser pulse energy in lidar observation experiments,which provides a correction basis for lidar data inversion.

A laser-induced fluorescence lidar has been developed for detecting the concentration of fluorescent aerosols in the air. The fluorescence lidar was constructed with a pulsed fourth-harmonic Nd:YAG laser at the ultraviolet wavelength of 266 nm with a repetition rate of 10 Hz. A 250 mm diameter custom telescope was used to collect optical spectra ranging from 260 to 560 nm. Fluorescence signals with wavelengths ranging from 310-440 nm were extracted, exploring a filter with a bandwidth of 130 nm. The preliminary experiments were conducted at the campus of Xi'an University of Technology, in which the fluorescence signals of atmospheric fluorescent aerosols were continuously collected from 20:00 to 23:00 CST on 13 December 2017. Based on the fluorescence lidar equation, the density of fluorescence signals was calibrated using Rayleigh-Mie scattering signals and ozone (O-3) concentration data at the ground level. Measured ranges show a strong dependence with the O-3 concentrations due to its absorption characteristics at ultraviolet 266 nm. Moreover, the concentration of the biogenic particles was also calculated based on the raw data of the fluorescence channel. Obtained results show that the concentration of biogenic particles in the Xi'an area varied greatly, ranging from 3456 particles . m(-)(3) to 8835 particles . m(-)(3) during winter. (C) 2018 Optical Society of America

From August to October 2010 lidar measurements of aerosols in the troposphere were performed at Otlica observatory, Slovenia, using a vertical scanning elastic lidar. The lidar data sample, which contains 38 nighttime vertical profiles of the mean aerosol extinction, was combined with continuous ozone concentration (O-3), particulate matter concentrations (PM) and daily radiosonde data. The obtained radiosonde and lidar-derived heights of the atmospheric boundary layer (ABL), which varied considerably from day to day, were found to be in good agreement. The mean values of the aerosol optical depth (AOD) at 355 nm, were calculated separately for the ABL and for the free troposphere (FT). A ten-fold increase of the FT AOD was observed during the days with predicted presence of Saharan dust above the lidar site. To correlate AOD values with the type and origin of aerosols, backward trajectories of air-masses above Otlica were modeled using the HYSPLIT model and clustered. High ABL AOD values were found to be correlated with local circulations and slowly approaching air masses from the Balkans and low values with northwestern flows. The highest values correlated with southwestern flows originating in northern Africa. (C) 2018 Elsevier Ltd. All rights reserved.

A novel method for particulate matter mass concentration measurement has been proposed based on a multi-wavelength lidar covering the ultraviolet to the near infrared spectra. The proposed method combined extinction coefficients at working wavelengths with quantity of mass extinction efficiency (MEE), which is defined as the ratio of extinction coefficient and mass concentration of particulate matter in unit volume, makes the mass concentration of particulate matter retrievable. To determine the values of the MEE, a mathematical model was developed based on the extinction efficiency data of certain wavelength reported with the Mie theory and the particle size distribution data derived from a multi-wavelength lidar. Retrieved results of mass concentration from two experimental cases with clear and fog/haze weather conditions, which s in line with the monitoring results at the ground level reported by the Environmental Agency. The proposed method invert the recent problem on aerosol sources monitoring and open new lidar capabilities on atmospheric research.

In order to investigate biological aerosols in the air, a fluorescence lidar has being developed at Laser Radar Center of Remote Sensing of Atmosphere, Xi'an University of Technology. The fluorescence lidar is constructed with a pulsed Nd:YAG laser, employing at based harmonic (1064 nm), second harmonic (532 nm) and fourth harmonic (266 nm) simultaneously, with a repetition rate of 10 Hz. A 250 mm diameter custom telescope is used to collect optical spectra ranging from 260-1100 nm. In the Infrared detection, an avalanche diode (APD) is used, and two photomultiplier tubes (PMTs) for two linear orthogonal polarization detection at a wavelength of 532 nm. Range-resolved fluorescence signals are collected in 32 channels of compound PMT sensor coupled with Czerny-Turner spectrograph. Based on the current configurations, we performed a series of numerical simulations to estimate the maximal detectable ranges and the minimal detectable concentrations of biological aerosols with various conditions. With a relative error of less than 10%, simulated results show that the system is able to monitor biological aerosols within detected distances of 1.3 km and of 2.0 km at daytime and nighttime, respectively. The developing fluorescence lidar is also capable to identify a minimum concentration of bio-aerosols at about 150 particles.L-1 with daytime operation and 100 particles.L-1 with nighttime at a distance of about 0.1 km. We truly believe that the fluorescence lidar could be spread in the field of remote sensing of biological aerosols in the near future.

Accurate aerosol optical properties could be obtained via the high spectral resolution lidar (HSRL) technique, which employs a narrow spectral filter to suppress the Rayleigh or Mie scattering in lidar return signals. The ability of the filter to suppress Rayleigh or Mie scattering is critical for HSRL. Meanwhile, it is impossible to increase the rejection of the filter without limitation. How to optimize the spectral discriminator and select the appropriate suppression rate of the signal is important to us. The HSRL technology was thoroughly studied based on error propagation. Error analyses and sensitivity studies were carried out on the transmittance characteristics of the spectral discriminator. Moreover, ratwo different spectroscopic methods for HSRL were described and compared: one is to suppress the Mie scattering; the other is to suppress the Rayleigh scattering. The corresponding HSRLs were simulated and analyzed. The results show that excessive suppression of Rayleigh scattering or Mie scattering in a high-spectral channel is not necessary if the transmittance of the spectral filter for molecular and aerosol scattering signals can be well characterized. When the ratio of transmittance of the spectral filter for aerosol scattering and molecular scattering is less than 0.1 or greater than 10, the detection error does not change much with its value. This conclusion implies that we have more choices for the high-spectral discriminator in HSRL. Moreover, the detection errors of HSRL regarding the two spectroscopic methods vary greatly with the atmospheric backscattering ratio. To reduce the detection error, it is necessary to choose a reasonable spectroscopic method. The detection method of suppressing the Rayleigh signal and extracting the Mie signal can achieve less error in a clear atmosphere, while the method of suppressing the Mie signal and extracting the Rayleigh signal can achieve less error in a polluted atmosphere.

Pollen is one of the important allergens in allergic diseases. The objective of this research effort is to investigate the use of a fluorescence lidar to measure concentrations of pollens in the atmosphere. The fluorescence lidar was constructed with a pulsed Nd:YAG laser, employing at fourth harmonic and second harmonic. A 250mm diameter custom telescope was used to receive optical spectra from 260 nm–560nm excited by 266nm, and R7506 Photomultiplier tubes for two linear orthogonal polarization applications at 532nm. Range-resolved fluorescence signals are collected in 32 channels of compound PMT sensor coupled with Czerny–Turner spectrograph. Based on the current configurations, we performed a series of simulations to estimate the measurement range and the concentration resolutions of pollens. With a relative error of less than 10%, theoretical analysis shows that the system is capable of identify a minimum concentration of pollen grains at 19 particlesL−1 and 1 particlesL−1 at daytime and nighttime within a range of 1.0km, which is capable of identify a minimum concentration of pollen grains at 903 particlesL−1 and 10 particlesL−1 at daytime and nighttime within a range of 5.0km.

Biological aerosols widely spreading in the atmosphere will easily result in various epidemic diseases, meanwhile, biological aerosol weapons pose a severe threat to the safety and security of military forces and civilians. It is critically important to remotely detect biological aerosols at real-time. In this work, a double-wavelength laser induced fluorescence lidar was constructed for atmospheric bacterial spores' identification and thus the early warning. The device employed a Nd YAG laser operating at 1 064 and 266 rim, with a repetition rate of 10 Hz. Based on lidar detection principle, a series of numerical simulations were performed to estimate the measurement range of the elastic scattering signals in the infrared band and the fluorescence signals induced by ultraviolet laser. In the ultraviolet band, the signals were analyzed with a spectrograph to evaluate the minimum concentrations of bacterial spores at different pulses. With a relative error of less than 10%, theoretical analysis shows that, within a range of 1.0 km, the system is capable of identifying a minimum concentration of bacterial spores at about 15 000 and 8 400 particles center dot L-1 at daytime and nighttime with the single laser pulse excitation. With an integrated pulses of 10 000, the detectable abilities of the fluorescence lidar greatly improves, identifying a minimum concentration of bacterial spores at 144 and 77 particles center dot L-1 at daytime and nighttime, respectively. In the lidar operation, when bacterial spores are located by the infrared elastic signals, one could actually extend the collected intervals in the fluorescence detection to improve the Signal-to-noise ratio, which may lose acceptable temporal resolution.