**4. Lidars on metal vapor and Nd:YAG lasers**

Lidars on metal vapor and Nd:YAG lasers are used since mid 70-ties of the last century. The fast progress in development of Nd:YAG lasers of stable and very good output parameters, meeting the requirements for effective atmospheric probing, provided their wide applications in most of lidar sensing systems. In the first decade of the new century, the problems related to more precise range-resolved characterization of the atmosphere (aerosol and molecular content, clouds, air quality, atmospheric transport, etc.) became the most important challenge to the modern lidar systems and their future effective incorporation in the Global Atmospheric Watch (GAW) networks. An important approach for improving the characterization procedures is the combined use of lidars (and lidar networks) with some other instruments as sun-photometers, microwave (MW) radiometers, in-situ measurements, etc. In spite of the already proven synergy of this approach, the further improvement of the lidar atmospheric characterization remains an important task. In this sense, a possible solution could be the use of multiwave lidar sensing, especially in the visible range. That is because the typical aerosols, loading the troposphere and low stratosphere as dust, volcanic ash, thin clouds, etc., are of submicron and near micron size range. Thus, they are commensurable to the wavelengths in

remote sensing. Typically, output powers of 1-3 W are sufficient for probing simultaneously

Most of the available Cu-vapor lasers operate using one of the two active substances - pure Cu or CuBr. They emit two basic spectral lines (at 510.6 nm and 578.2 nm) of close mean output powers. The beam and temporal characteristics at both lines are practically equal. These lasers provide Gaussian beams of low intensity fluctuations and simultaneous emission of pulses at both wavelengths. The CuBr lasers are more preferable for lidars, because of their very low readiness time (~5-15 min), due mainly to the quite lower working

Lasers on Au-vapor emit radiation basically at a wavelength of 627.8 nm with temporal and beam parameters quite similar to these of pure Cu-vapor lasers. The combination of Cu- and Au-vapor lasers in a single laser tube is very attractive for lidar applications. Mean output powers of ~0.5 W for each of the three lines 510.6 nm, 578.2 nm and 627.8 nm will be

Conventional flashlamp-pumped Nd:YAG lasers, operating in Q-switching mode, provide pulses of 5-10 ns FWHM, at 10-100 Hz typical repetition rates, with extremely high pulse energy reaching more than 1 J at the fundamental wavelength (1064 nm) and up to hundreds of millijoules at the second (532 nm), third (355 nm), and fourth (266 nm) harmonics. These lasers can be designed for simultaneous operation at all the available wavelengths or at optional combinations of them, being by this manner powerful multiwavelength sources of UV, visible, and IR light. They exhibit perfect shot-to-shot energy stability (instabilities of ≤±2%), long-term power drift of less than ±3%, and temporal jitter of ≤1 ns. In addition, Nd:YAG lasers possess excellent beam-quality characteristics such as Gaussian spatial profile, beam divergence down to 0.5 mrad, pointing stability of < 50 μrad, and polarization ratio > 80%. Diode-pumped Nd:YAG lasers, using bars of powerful laser diodes instead of flash-lamps, are also commercially available. They combine the features

quoted above with compactness and high averaged power at kHz pulse frequencies.

Lidars on metal vapor and Nd:YAG lasers are used since mid 70-ties of the last century. The fast progress in development of Nd:YAG lasers of stable and very good output parameters, meeting the requirements for effective atmospheric probing, provided their wide applications in most of lidar sensing systems. In the first decade of the new century, the problems related to more precise range-resolved characterization of the atmosphere (aerosol and molecular content, clouds, air quality, atmospheric transport, etc.) became the most important challenge to the modern lidar systems and their future effective incorporation in the Global Atmospheric Watch (GAW) networks. An important approach for improving the characterization procedures is the combined use of lidars (and lidar networks) with some other instruments as sun-photometers, microwave (MW) radiometers, in-situ measurements, etc. In spite of the already proven synergy of this approach, the further improvement of the lidar atmospheric characterization remains an important task. In this sense, a possible solution could be the use of multiwave lidar sensing, especially in the visible range. That is because the typical aerosols, loading the troposphere and low stratosphere as dust, volcanic ash, thin clouds, etc., are of submicron and near micron size range. Thus, they are commensurable to the wavelengths in

temperature of about 450 0C as compared to 1500 0C -1700 0C for lasers on pure Cu.

the troposphere and stratosphere.

sufficient to cover heights above 15 km in PC mode.

**4. Lidars on metal vapor and Nd:YAG lasers** 

the visible range. The simultaneous use of laser radiations covering the green-to-red part of visible range can provide better description of the backscattering wavelength dependence on the aerosol particle sizes, especially in multimode size distributions. Efficient sources of radiation in this spectral domain are the mentioned above Cu- and Au- vapor lasers. The combination of lidars based on these lasers with Nd:YAG lidars operated at 1064 nm and 532 nm offers additional advantages to cover practically the mid-visible and near IR ranges, where the absorption effects can be neglected as a rule.

In Arshinov et al., (1983) the application of Cu-vapor lidar for measuring profiles of the atmospheric humidity and temperature is described. The Cu-vapor lidar system is used for detection of stratospheric aerosol layers at heights of up to 28 km (Kolarov et al., 1988). The scattering ratios (aerosol and molecular to molecular) exceeding 1.5 within the height range 22-28 km are measured at the wavelength of 510.6 nm (1-3 W mean power, 5 KHz repetition frequency, 19cm telescope diameter, and 10-100 s accumulation times) with a photon counting system. A complex system, containing Cu-vapor (510.6 nm) and Au-vapor (627.8 nm) lasers, combined with MW and IR radiometers and MW radar is described in Stoyanov et al., (1988). The two lasers are synchronized in time. The output powers are of 1 -3 W at 510.6 nm and 0.5 W at 627.8 nm, at pulse repetition frequencies 5-15 KHz. The experimentally demonstrated operational heights are 30 km for 510.6 nm and 22 km for 627.8 nm at 100 s accumulation times. The short accumulation times provided opportunities for studying some dynamic processes in the lower stratosphere. One of the most important results of these experiments is the demonstration of good lidar performance in the PC mode within the ranges from the planetary boundary layer (PBL) heights (~750 m) up to 30 km in the stratosphere. The first lidar and MW radiometers remote sensing experiment is performed using Cu-vapor laser of parameters as given above (Gagarin et al., 1987). The operational wavelengths of MW radiometers are 0.8 cm and 1.35 cm to be sensitive to the free water in cloud droplets and the water vapor. The measurement accuracy for the brightness temperature is below 0.1 K. The accuracy of determining cloud water content and water vapor is ~ 0.5% and 1%, respectively. In this schematic, the lack of range resolution of the radiometers is compensated by the good lidar resolution. In series of experiments the links between the time variations of the backscattered time-resolved lidar signals and the brightness temperatures at the radiometer wavelength are demonstrated and analyzed.

The aerosol lidar with CuBr-vapor laser developed at Laser Radars Lab of the Institute of Electronics (LRL-IE), Bulgarian Academy of Sciences, is shown in Fig.3a (Grigorov et al., 2010). The CuBr-laser generates pulses at high-repetition frequency of 13 kHz (10 kHz in the upgraded version), with duration of 10 ns at 510.6 nm and 578.2 nm. Laser beam is directed vertically upward. Two Cassegrain telescopes with 15 cm aperture and 2.25 m focal length receive the backscattered radiation at the two wavelengths. A registration in PC mode is applied. Received backscattered lidar signals are stored in the computer by means of a photon counting board providing spatial resolution of 15-30 m, in 1024 samples and averaging time of 1 min. The maximum height is 15 km, limited by the laser pulse repetition frequency. Under daytime conditions, the sounding height decreases to about 4-5 km, due to intensive sky illumination, reducing the signal-to-noise ratio (SNR). Each lidar measurement lasts about 3-4 hours and more. The lidar profiles, integrated over accumulation time of 1 min, are additionally averaged by summation of data from 30 profiles. Thus, the measurement time for each profile amounts to 30 min. The stored data are subsequently processed by Fernald's algorithm, using a program in MATLAB environment, developed in LRL-IE.

LIDAR Atmospheric Sensing by Metal Vapor and Nd:YAG Lasers 357

activities. Aerosols, having different size distributions and chemical or physical properties, can affect the climate over large regions. Aerosol particles reflect the solar radiation, act as cloud condensation nuclei, modify the scattering properties and lifetime of clouds, influence the precipitation cycles, as well as atmospheric radiative and thermal balance, etc. Ejecting immense amounts of ashes and gases into the atmosphere, active volcanoes can strongly affect for long periods climate, ecology, aviation industry, agricultural activities, and human health over regions of up to global scale. This is why observations and alerting on volcanic aerosols in the atmosphere are of great importance. Lidars can provide real-time sensing of

The significance of lidar information, provided by a single lidar station is essentially enhanced if working in a lidar network. Such an idea combines together researches in different lidar networks as EARLINET (http://www.earlinet.org), MPLNET (http://mplnet.gsfc.nasa.gov/), AD-Net (http://www-lidar.nies.go.jp/AD-Net/), etc. The primary goal of the project EARLINET (European Aerosol Research Lidar Network) is the creation of a common database, banding the results of observation of lidar stations located in the European countries (Bösenberg et al., 2003). Main result of such cooperation is the establishment of a quantitative lidar dataset describing the aerosol vertical, horizontal, and temporal distribution, including its variability on a continental scale. Such a dataset could be a comprehensive data source to address the four-dimensional spatio-temporal distribution of aerosols on a global scale

Analyses of lidar data require additional information to improve the interpretations of both the type and origin of aerosol layers. Several regional models for simulation and prediction of the dust cycle in the atmosphere have been developed (Kallos et al., 1997; Ozsoy et al., 2001; Nickovic et al.; 2001; Perez et al., 2006a, 2006b). The Dust Regional Atmospheric Model (DREAM) (Nickovic et al.; 2001; Perez et al., 2006a, 2006b) provides reliable operational forecast maps of dust load and concentration in the atmosphere over the North Africa and Euro-Mediterranean region. The model is operated by the Barcelona Supercomputing Center (BSC), (http://www.bsc.es/projects/earthscience/DREAM/). Maps present the cloud coverage, wind directions and speeds, and dust loads. Another source of information about the origin of the aerosol layers offers the HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) model (Draxler & Hess, 1998; Draxler & Rolph, 2011). It represents a complete system for computing simple air parcel trajectories to complex dispersion and deposition simulations. The model can be run interactively through the READY system at the site http://www.arl.noaa.gov/HYSPLIT\_info.php of the Air Resource Laboratory of NOAA, USA. Calculations of the forward/backward air mass trajectories give a plot of the path passed by the air mass for a chosen time period before arriving to the lidar station

The results of lidar applications in the remote sensing of the atmosphere, presented in this Section, are obtained in the LRL-IE using the two elastic backscatter lidar systems described in Sec.4. Some results are reported, concerning lidar mapping of aerosol fields over large industrial zones as one of the important applications of lidars in the regional ecological

The Sofia lidar station, being one of the stations working in the frame of the EARLINET

atmospheric aerosols over large areas with high spatial and temporal resolution.

(Pappalardo et al., 2010).

location.

studies and expertise.

project, is involved in the following research activities:

Fig. 3. Photographs of the CuBr-vapor (a) and Nd:YAG (b) laser-based aerosol lidars at LRL-IE involved in the European Lidar Network measurement programs.

The Nd:YAG lidar system of the LRL-IE (Fig.3b) is a 3-channel combined aerosol-Raman lidar (Peshev et al., 2010). The laser provides output pulse energies of up to 1 J at 1064 nm and up to 120 mJ at 532 nm, at a repetition rate of 2-5 Hz, with pulse duration of 15 ns FWHM. The pulse power is of up to 70 MW at 1064 nm and up to 10 MW at 532 nm. The corresponding values of the averaged power for the two wavelengths are of up to 2 W and 0.25 W, respectively. The output beam divergence is of 2.5 mrad (total angle). These performance characteristics of the laser allow one to carry out nighttime and daytime lidar measurements. The optical part of the receiver contains a Cassegrain telescope (35 cm aperture; 2 m focal length) and a 3-channel spectrum-analyzing module based on dichroic beam-splitters, narrowband interference filters (1-3 nm FWHM), edge-pass filters, and neutral densities. The electronic part of the lidar receiving system consists of three compact photo-electronic modules. Each module comprises a photon detector (photomultiplier or avalanche photodiode), 10 MHz/14-bit ADC, high voltage power supply, and controlling electronics. The aerosol lidar channels operate in analog mode with 15 m range resolution. Receiving modules are connected to a computer by highspeed USB ports. The acquisition system is controlled by specialized software providing the accumulation, storage, and processing of lidar data. It allows for evaluating and plotting profiles of range-corrected lidar signals, aerosol backscattering coefficient, and estimation error.

#### **5. Applications of LIDARS for remote atmospheric sensing**

Atmospheric aerosols originate from natural and anthropogenic sources such as desert windstorms, forest fires, volcanic eruptions, sea spray, and combustion products of human

(a) (b)

Fig. 3. Photographs of the CuBr-vapor (a) and Nd:YAG (b) laser-based aerosol lidars at LRL-

The Nd:YAG lidar system of the LRL-IE (Fig.3b) is a 3-channel combined aerosol-Raman lidar (Peshev et al., 2010). The laser provides output pulse energies of up to 1 J at 1064 nm and up to 120 mJ at 532 nm, at a repetition rate of 2-5 Hz, with pulse duration of 15 ns FWHM. The pulse power is of up to 70 MW at 1064 nm and up to 10 MW at 532 nm. The corresponding values of the averaged power for the two wavelengths are of up to 2 W and 0.25 W, respectively. The output beam divergence is of 2.5 mrad (total angle). These performance characteristics of the laser allow one to carry out nighttime and daytime lidar measurements. The optical part of the receiver contains a Cassegrain telescope (35 cm aperture; 2 m focal length) and a 3-channel spectrum-analyzing module based on dichroic beam-splitters, narrowband interference filters (1-3 nm FWHM), edge-pass filters, and neutral densities. The electronic part of the lidar receiving system consists of three compact photo-electronic modules. Each module comprises a photon detector (photomultiplier or avalanche photodiode), 10 MHz/14-bit ADC, high voltage power supply, and controlling electronics. The aerosol lidar channels operate in analog mode with 15 m range resolution. Receiving modules are connected to a computer by highspeed USB ports. The acquisition system is controlled by specialized software providing the accumulation, storage, and processing of lidar data. It allows for evaluating and plotting profiles of range-corrected lidar signals, aerosol backscattering coefficient, and

IE involved in the European Lidar Network measurement programs.

**5. Applications of LIDARS for remote atmospheric sensing** 

Atmospheric aerosols originate from natural and anthropogenic sources such as desert windstorms, forest fires, volcanic eruptions, sea spray, and combustion products of human

estimation error.

activities. Aerosols, having different size distributions and chemical or physical properties, can affect the climate over large regions. Aerosol particles reflect the solar radiation, act as cloud condensation nuclei, modify the scattering properties and lifetime of clouds, influence the precipitation cycles, as well as atmospheric radiative and thermal balance, etc. Ejecting immense amounts of ashes and gases into the atmosphere, active volcanoes can strongly affect for long periods climate, ecology, aviation industry, agricultural activities, and human health over regions of up to global scale. This is why observations and alerting on volcanic aerosols in the atmosphere are of great importance. Lidars can provide real-time sensing of atmospheric aerosols over large areas with high spatial and temporal resolution.

The significance of lidar information, provided by a single lidar station is essentially enhanced if working in a lidar network. Such an idea combines together researches in different lidar networks as EARLINET (http://www.earlinet.org), MPLNET (http://mplnet.gsfc.nasa.gov/), AD-Net (http://www-lidar.nies.go.jp/AD-Net/), etc. The primary goal of the project EARLINET (European Aerosol Research Lidar Network) is the creation of a common database, banding the results of observation of lidar stations located in the European countries (Bösenberg et al., 2003). Main result of such cooperation is the establishment of a quantitative lidar dataset describing the aerosol vertical, horizontal, and temporal distribution, including its variability on a continental scale. Such a dataset could be a comprehensive data source to address the four-dimensional spatio-temporal distribution of aerosols on a global scale (Pappalardo et al., 2010).

Analyses of lidar data require additional information to improve the interpretations of both the type and origin of aerosol layers. Several regional models for simulation and prediction of the dust cycle in the atmosphere have been developed (Kallos et al., 1997; Ozsoy et al., 2001; Nickovic et al.; 2001; Perez et al., 2006a, 2006b). The Dust Regional Atmospheric Model (DREAM) (Nickovic et al.; 2001; Perez et al., 2006a, 2006b) provides reliable operational forecast maps of dust load and concentration in the atmosphere over the North Africa and Euro-Mediterranean region. The model is operated by the Barcelona Supercomputing Center (BSC), (http://www.bsc.es/projects/earthscience/DREAM/). Maps present the cloud coverage, wind directions and speeds, and dust loads. Another source of information about the origin of the aerosol layers offers the HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) model (Draxler & Hess, 1998; Draxler & Rolph, 2011). It represents a complete system for computing simple air parcel trajectories to complex dispersion and deposition simulations. The model can be run interactively through the READY system at the site http://www.arl.noaa.gov/HYSPLIT\_info.php of the Air Resource Laboratory of NOAA, USA. Calculations of the forward/backward air mass trajectories give a plot of the path passed by the air mass for a chosen time period before arriving to the lidar station location.

The results of lidar applications in the remote sensing of the atmosphere, presented in this Section, are obtained in the LRL-IE using the two elastic backscatter lidar systems described in Sec.4. Some results are reported, concerning lidar mapping of aerosol fields over large industrial zones as one of the important applications of lidars in the regional ecological studies and expertise.

The Sofia lidar station, being one of the stations working in the frame of the EARLINET project, is involved in the following research activities:

LIDAR Atmospheric Sensing by Metal Vapor and Nd:YAG Lasers 359

in red. Parts of urban regions of Kalarash and Silistra are presented in dark orange. The position of the scanning lidar is marked as well. As seen, two main areas of intensive aerosol loadings are well displayed on the lidar map. The upper map area shows the aerosol field distribution in the vicinity of the steel plant near Kalarash. This plant is one of the main pollution sources in the region of the investigation. The obtained lidar maps related to adjacent sectors show that the pollution emitted in this region is transported approximately parallel to the Danube River, probably not crossing the border. The aerosol plume in the mid-part of the image is identified as to be emitted from the cellulose plant located far from the mapped region. The pollution of this plume is supposed to pass through the river. This estimate is supported by the parallel gas analysis. The measured phenol concentration (typical for cellulose manufacturing) exceeds about 8 times the tolerable level in Silistra.

Fig. 4. Lidar map of aerosol pollution distribution over the Silistra – Kalarash region;

Numerous observations on Saharan dust presence in the atmosphere over Sofia are carried out by both lidars of LRL-IE, in order to follow the concentration, spreading, and temporal

*Plume from the Cellulose Plant* 

*Steel plant pollution region* 

**5.2 Lidar observations on Saharan dust loadings in the atmosphere** 

14.10.1992.

*Lidar* 


#### **5.1 Lidar mapping of aerosol pollutions over industrial regions**

The monitoring of air pollution distribution over large industrial and urban zones is an important task for improving the quantitative and qualitative estimates of the pollution impact on the environmental conditions. The air pollution transport from local sources is connected with the motion of air masses driven by the wind. The high temporal and spatial resolution of lidar probing and the speed of measurements define the high efficiency of aerosol transport lidar mapping. The opportunity to scan areas of the order of 50-100 km2 by a single scanning lidar system can provide valuable information for in-depth analysis of the pollution dynamics over broad regions containing a large number of potential local sources.

Results are presented below from a lidar mapping of air pollution distribution and transborder pollution transport over the Danube River in the region of Silistra (Bulgaria) – Kalarash (Romania). The measurements are performed using a scanning CuBr lidar system (=510.6 nm) applying photon counting detection. The main lidar parameters are similar to those described in Sec.4. The scanning system provides 1.60 scanning step from 00 to 3600 in the horizontal plane with an angular elevation step of 10 within the range from -30 to 150. The scanning lidar is mounted on the roof of a high building located on the hill near the riverside of the Danube River. The total area scanned in successive measurement sessions in order to map different zones as urban and industrial ones, river ports, etc., is of more than 200 km2. The scanning zone is divided into sectors, separately mapped by the lidar. The integration time at a given angle is chosen to be 1 min, providing maximum sounding distances to more than 12 km in horizontal direction. The measurement time for scanning a 300 angular sector is of about 30 min. Under conditions of measurements, one can accept the concept of approximately frozen aerosol fields. Lidar mapping measurements are supported by simultaneous in-situ measurements (including gas analysis) by a specialized transportable laboratory (Mitzev et al., 1995).

The processing of lidar data includes solutions of lidar equation for each angular direction, using the Klett's inversion method and estimation of the volume aerosol extinction coefficient. The latter is converted into aerosol mass concentration, by using our previously derived approximate empirical expression (Mitzev et al., 1995). Finally, the processed data in mass concentration profiles (stored in radial coordinates) are transformed to rectangular coordinates and attached to the geographical map (see Fig.4).

The lidar maps, created from these measurements contain valuable quantitative information about the spatial distribution of aerosol fields as well as their dominated paths of propagation. Fig.4 presents a lidar map of aerosol distribution (dark blue isolines of constant mass concentration) over an area of ~ 70 km2 (12 km x 6 km). The Danube River and the Lake of Kalarash are displayed in blue. The contours of some geographical objects are given

 Observation of special phenomena, such as unusually high concentrations of aerosols in the troposphere (transportation of mineral dust from Sahara desert over the Mediterranean Sea to Europe, volcanic eruptions, formation of smoke layers resulting

Correlative measurements with space-borne lidars, in the frame of international

The monitoring of air pollution distribution over large industrial and urban zones is an important task for improving the quantitative and qualitative estimates of the pollution impact on the environmental conditions. The air pollution transport from local sources is connected with the motion of air masses driven by the wind. The high temporal and spatial resolution of lidar probing and the speed of measurements define the high efficiency of aerosol transport lidar mapping. The opportunity to scan areas of the order of 50-100 km2 by a single scanning lidar system can provide valuable information for in-depth analysis of the pollution dynamics over broad regions containing a large number of potential local sources. Results are presented below from a lidar mapping of air pollution distribution and transborder pollution transport over the Danube River in the region of Silistra (Bulgaria) – Kalarash (Romania). The measurements are performed using a scanning CuBr lidar system

=510.6 nm) applying photon counting detection. The main lidar parameters are similar to those described in Sec.4. The scanning system provides 1.60 scanning step from 00 to 3600 in the horizontal plane with an angular elevation step of 10 within the range from -30 to 150. The scanning lidar is mounted on the roof of a high building located on the hill near the riverside of the Danube River. The total area scanned in successive measurement sessions in order to map different zones as urban and industrial ones, river ports, etc., is of more than 200 km2. The scanning zone is divided into sectors, separately mapped by the lidar. The integration time at a given angle is chosen to be 1 min, providing maximum sounding distances to more than 12 km in horizontal direction. The measurement time for scanning a 300 angular sector is of about 30 min. Under conditions of measurements, one can accept the concept of approximately frozen aerosol fields. Lidar mapping measurements are supported by simultaneous in-situ measurements (including gas analysis) by a specialized

The processing of lidar data includes solutions of lidar equation for each angular direction, using the Klett's inversion method and estimation of the volume aerosol extinction coefficient. The latter is converted into aerosol mass concentration, by using our previously derived approximate empirical expression (Mitzev et al., 1995). Finally, the processed data in mass concentration profiles (stored in radial coordinates) are transformed to rectangular

The lidar maps, created from these measurements contain valuable quantitative information about the spatial distribution of aerosol fields as well as their dominated paths of propagation. Fig.4 presents a lidar map of aerosol distribution (dark blue isolines of constant mass concentration) over an area of ~ 70 km2 (12 km x 6 km). The Danube River and the Lake of Kalarash are displayed in blue. The contours of some geographical objects are given

Regular lidar measurements of the atmosphere performed twice weekly;

from forest or industrial fires, intense photochemical smog, etc.);

**5.1 Lidar mapping of aerosol pollutions over industrial regions**

transportable laboratory (Mitzev et al., 1995).

coordinates and attached to the geographical map (see Fig.4).

cooperation.

( in red. Parts of urban regions of Kalarash and Silistra are presented in dark orange. The position of the scanning lidar is marked as well. As seen, two main areas of intensive aerosol loadings are well displayed on the lidar map. The upper map area shows the aerosol field distribution in the vicinity of the steel plant near Kalarash. This plant is one of the main pollution sources in the region of the investigation. The obtained lidar maps related to adjacent sectors show that the pollution emitted in this region is transported approximately parallel to the Danube River, probably not crossing the border. The aerosol plume in the mid-part of the image is identified as to be emitted from the cellulose plant located far from the mapped region. The pollution of this plume is supposed to pass through the river. This estimate is supported by the parallel gas analysis. The measured phenol concentration (typical for cellulose manufacturing) exceeds about 8 times the tolerable level in Silistra.

Fig. 4. Lidar map of aerosol pollution distribution over the Silistra – Kalarash region; 14.10.1992.
