This paper describes the flame vision system developed by CISE, on behalf of Thermical Research Division of ENEL, allowing a non-intrusive analysis and a probabilistic classification of the combustion process inside the gas turbines. The system is composed of a vision probe, designed for working in hostile environments and installed inside the combustion chamber, an optical element housing a videocamera, and a personal computer equipped with a frame grabber board. The main goal of the system is the flames classification in order to evaluate the occurrency of deviation from the optimal combustion conditions and to generate warning messages for power plant personnel. This is obtained by comparing some geometrical features (baricenter, inertia axes, area, orientation, etc.) extracted from flame area of images with templates found out during the training stage and classifying them in a probabilistic way by using a Bayesian algorithm. The vision system, now at the test stage, is intended to be a useful tool for combustion monitoring, has turbines set-up, periodic survey, and for collecting information concerning the burner efficiency and reliability; moreover the vision probe flexibility allows other applications as particle image velocimetry, spectral and thermal analysis.

We present a low angle light scattering optical particle sizer (measuring range 1-100 micrometers ) utilizing an innovative scheme that enables the instrument to operate at extinction values as small as 10^-5. At variance with existing systems, the illuminating beam is properly modulated so to allow a coherent detection scheme to be used. In this way optical and electronic noise can be detected and subtracted from the measured signals therefore increasing the S/N ratio.

A new instrument specifically developed for measuring NH₃ concentrations at the inlet and outlet of an SCR reactor is described. It is based on the photoacoustic technique and can perform an online and in-situ n-point measurement grid. The major problem is the severe interference of CO₂ combustion gas that absorbs at the same wavelength used for NH₃, lowering the sensitivity of the technique. A ¹³CO₂ tunable laser emitting at three wavelengths has been used in order to evaluate simultaneously the CO₂ and NH₃ concentration. The integrated measurements have been done by a simple differential absorption technique by using a pyroelectric detector; the spatially resolved measurements have been carried out through the analysis of the acoustic wave generated by the laser beam and collected by four microphones perpendicularly disposed. Experimental tests on a 35 MW industrial plant and comparison with traditional sampling methods are presented.

Fourier transform infrared spectroscopy is a powerful tool for measuring the spectral radiation from flames and industrial processes. The effects on the spectrum from flame fluctuations occurring naturally during the measurement are analyzed briefly and illustrated by experiments and simulations. Results from applications on an oil flame, a pulverized coal flame, and a full- scale waste incinerator are described. The experiments show that important information about the composition and temperature of gas and solid temperatures might be extracted from the emission spectra. The practical and technical limitations of the technique and possible future applications are discussed.

New environmental control strategies and innovative and cost-effective measurement techniques are needed more and more because of increasing regulatory requirements for the limitation of greenhouse gas and other pollutant emissions from industrial facilities. Fourier- transform-infrared-spectroscopy (FTIS) offers a great potential for remote sensing of plume emissions. A high-resolution K300 spectrometer together with a multicomponent analysis software was used inside a mobile environmental laboratory to quantify effluent concentrations in plumes of smoke stacks and flares. The software is based on radiative transfer line-by-line calculations and least-square fit procedures. Several measurement results are described applying the remote emission control FTIS technique. Estimates of the regional air pollution due to an ensemble of small building smoke stacks are given. Measurements are reported to quantify unburned methane from elevated flares.

To quantify aircraft engine emissions with respect to their impact on the upper and lower atmosphere, remotely working systems are required. The combination of passive FTIR gas concentration measurements with LDA gas velocity-data from exhaust plumes leads to an absolutely remote determination of emission rates of several molecule species. The combined system was tested at the Munich-North power plant where the obtained fluxes of CO₂, CO, NO, and HCl showed good agreement with the in-situ measurements. In case of aircraft emissions of CO₂, H₂O, CO, and No, data shall be determined. First tests of the system with the DLR-Falcon aircraft in October 1994 will be discussed.

To evaluate the impact of air traffic on the upper and lower troposphere, one must find an effective method to measure the actual gas emissions of aircraft engines at defined thrust levels and at all altitudes. FTIR-emission-spectroscopy detects the thermal radiation of hot exhaust gases, yielding all information about its compounds during one measurement. This remote technique can be used under ground- and flight-conditions. The theoretical line-by-line retrieval of the measured spectra simulates the radiative transfer through several plume- and foreground-layers and is based on the HITRAN 92 database. After the spectroscopic determination of the plume temperature and its profile from the CO₂-band around 2400 cm^-1, one obtains the toal mass of the single gas species in the field of view of the spectrometer. Comparing the measured data with the theoretical emission index of CO₂ from ideal stoichiometric combustion, one obtains the emission indices for the other measured species. Knowing the fuel consumption of the engine, one may get the emission rates of the compounds in g/a. Several engine types, old fashioned engines (no bypass) and modern JT8 and CFM56 bypass at different thrust levels have been analyzed. H₂O, CO₂ CO, and NO concentrations can be derived immediately from the measurements right behind the nozzle exits, where the temperature profile is known to be homogeneous. The retrieval of the measured data far behind the nozzle exit uses a computer time consuming multilayer model. Formaldehyde and other hydrocarbon species are seen in some spectra and shall be implemented in the computer code. Apart form future applications for the turbine development and the engine-status control after a certain flight time, this remote sensing system can deliver emission data of aircraft engines and the temperature decay of the exhaust plumes at all altitudes.

A frequency stabilized distributed feedback dye laser is applied to give Fourier-limited excitation pulses with a duration of 50 ps at 226 nm and 309 nm. Then NO and OH radicals are excited in an atmospheric standard flame (McKenna burner) or in a pressure cell. The time-evolution of the flurorescence is recorded by a Streak camera in a pressure in the picosecond time scale. Quenching rates for NO, with CO₂, O₂, NO, and N₂ are reported. Systematic influences of the direction of polarization of the exciting laser beam on the LIF-decay are discussed as well as sensitivity and detection limits of the picosecond spectroscopy compared to the nanosecond LIF.

A method for the simultaneous detection and separation of two different 2D signals with one single laser and camera system will be presented. Laser Rayleigh scattering was used to determine the temperature field. The laser induced fluorescence (LIF) technique was applied to measure relative concentrations of either the OH radical or a tracer. The tracer was used for the detection of unburned methane. Signal separation was performed by a liquid filled prism specially designed for this application. First measurements prove the general suitability of this method inside technical burner systems burner systems and show the way for an improvement of the accuracy.

A mobile CARS system has been used to determine temperatures and their fluctuations in a high pressure combustion test rig with a maximum thermal power of 8 MW. Measurements have been performed in turbulent natural gas and oil flames up to a pressure of 14 bar. A high degree of turbulence and an increased pressure together with an extended size of the combustion chamber caused beam propagation effects to become a serious problem for this measuring technique. Temperature determinations in sooty oil flames became even more difficult due to the strong absorption of a part of the fundamental band of the N₂ CARS spectrum by C₂ radicals. Probability density functions of the temperature, each based on 1000 single shot measurements, have been determined at different radial and axial locations within the combustion chamber.

This paper describes a UV laser diagnostic system by means of which laser spectroscopic experiments were performed under microgravity conditions in a ground-based drop tower for the first time. A tunable, narrow bandwidth excimer laser is positioned at the top of the drop tower. The laser beam enters a falling drop capsule containing a specially adapted burner or combustion chamber. By the use of laser induced fluorescence spectroscopy measurements of 2D concentration and temperature profiles can be performed. Solutions of selected experimental problems such as laser beam collimation over a distance of more than 120 m, compensation of capsule drift, signal detection, and data acquisition (250 frames/s, 4.7 s measuring period), are discussed in detail. First measurements of laser induced predissociation fluorescence of OH radicals in a methanol flame under microgravity conditions are presented.

In order to get an understanding of the turbulent air-fuel mixing behavior and the subsequent combustion process of technically applied spray flames, spontaneous Raman scattering investigations were carried out in the flame of a commercial oil burning furnace. In this furnace a spray of liquid n-heptane is turbulently mixed with air by a swirl generating mixing device. The present geometry of the burner head creates a nonsymmetrical flame so that it was necessary to observe the whole flame volume in order to achieve meaningful data of the mixing behavior. For this reason spontaneous Raman scattering in combination with spatially resolving OMA-detection was applied. Using a simple polarization technqiue in addition to signal averaging over several hundred laser shots, broadband emissions of reacting hydrocarbons are suppressed effectively. This experimental setup facilitates simultaneous 1D spatial resolution of all majority species densities (fuel, N₂, O₂, CO₂, and H₂O) along the axis of the focused laser beam. Scanning the whole flame volume by the laser beam then offers 3D information. Quantitative densities of all majority species and the mean temperature were gained using calibration objects. The investigations offer information of stable recurring structures inside the turbulent spray flame. New insights into the mixing and combustion behavior of technically applied combustion systems were gained which are very promising for designing new burner head geometries in order to achieve more effective NO- reduction.

Regarding monitoring of industrial regions the performance of a mobile CO₂ lidar sensor detecting and localizing organic air pollutants is demonstrated.

This paper describes a mobile differential absorption LIDAR system, which operates in the UV, visible, and IR spectral regions. This system can measure a range of important air pollutants emitted by industry, including SO₂, NO₂, NO, HCl, benzene, toluene, and a large range of other VOC's. These species can be monitored at fugitive and flammable levels at ranges of up to 1 km (for IR measurements) and 3 km (for UV measurements). Examples of measurements of fluxes emitted from large scale industrial sties are presented and discussed. Comparisons are given between measured fluxes and those calculated using the US Environmental Protection Agency's and American Petroleum Institute's standard procedures for estimating industrial emissions. The fluxes measured by DIAL are higher than the values derived from the API procedures. Possible reasons for discrepancies between the measured results and the EPA/API estimation procedures will be discussed.

The diurnal variation of the Planetary Boundary Layer (PBL) structure and of the vertical profiles of the suspended particlates attenuation coefficient, are presented for the case of a semi-rural area near the north-eastern exit of the Athenian basin. These data were obtained using a ground-based LIDAR system operating at 355 nm and 532 nm, during the MEDCAPHOT-TRACE Campaign in the summer of 1994. The LIDAR system was pointing vertically and was operated during various air pollution (low up to severe photochemical episodes) and meteorological conditions (synoptic flow circulation, land-, and sea-breeze circulation). The LIDAR data are intercompared with meteorological and photochemical pollution data obtained by an instrumented aircraft overflying the Athens basin.

The SNCR-Method (Selective Non Catalytic Reaction) uses injection of ammonia to reduce nitrogen oxides present in stack gas produced by combustion facilities (e.g. power plants, waste incineration plants): 6N02+8NH3 —<7N2+12H20 4N0+02+4NH3—*4N2+ 6H20. In order to achieve an optimum injection of ammonia the concentration must be determined immediately after the combustion process (Fig. 1).

Atmospheric eddies, which have slightly different properties than their environment and are believed to be transported by the wind (Taylor's hypothesis), are used as tracers for remote wind measurements with a fast incoherent lidar. Horizontal measurements, parallel with the wind, have shown that the atmospheric structures can be traced in space and time and that the horizontal wind speed can be determined from a set of subsequent measurements. Also, the characteristic size and life time of the strucutres were inverted from the lidar system with their axes pointing in slightly different horizontal directions. Using the cross-correlation technique, the wind vector was derived from the geometry of the sysetm and the transient times of the structures crossing the two lidar field-of-views. The measured wind vectors are comparable with the in-situ measured wind vector. The same technique was applied to measure the vertical profile of the wind vector using a single lidar in the triangulation mode using fast adjustable platform pointing subsequently in three different (azimuth and elevation) directions. The wind vector could be measured to altitudes of about 1 km and were in agreement with the in-situ measured data from sensors on a 200 m high meteo mast and from data provided by a Doppler sodar.

Strategies for the reduction of emissions of diffuse sources or measures for the stabilization of these inputs into atmosphere require monitoring techniques providing high quality data. Because in many cases these sources are difficult to access, an approach is considered combining optical remote sensing techniques for the estimation of path-integrated concentrations near sources with dispersion modeling. Different methods are considered and compared for this transport modeling. By these methods, measurement configurations can be found which are optimal in different aspects for estimating emission rates. Hence, measurements can be done matched to specific source configurations and the quality of these estimatiosn can be assessed.

Laser equipment and methodology have been developed to monitor the power of dust sources ejecting aerosol pollution to atmosphere, and the distribution of aerosol mass concentration at regions of large factories. This work is directed to describe lidar technique of atmospheric aerosol monitoring. Technical parameters of lidar stations and results of experimental investigation are presented. The measurements were implemented by the multiwavelength lidar 'Glory-M', which has 7 working wavelengths within the range 0.38-1.0 micrometers , and mobile lidars station (MLS). The lidar design provides measuring the intensity and the depolarization of backscatter signals. Two problems have been considered. The first one is ivnestigating aerosol optical parameter spectra over the boundary atmoshperic layer both at background and industrial regions. The extinction spectra for industrial regions are different from low power, as for background ones. The obtained data will enable one to correct the atmosphere spectra deformation in observing different regions of the Earth from satellites and aircraft. The second problem is invesitigating spatial distribution of dust over regions of industrial centers. The dust flows and power of outbursts have been estimated. The maps of dust mass concentration at the regions of the large factories have been constructed.

A scanning single-frequency aerosol lidar is described in which a system of control, recording, and processing of information is based on a personal computer. The problem of processing the lidar returns with instrumental noise is considered. The examples of lidar application to monitoring of the spatiotemporal distribution of aerosol polluting fields, as well as determining the strength of emissions from stacks of local pollution sources are considered.

Nitrogen oxide is one of the main pollutants in the atmosphere of Moscow city. Its content in the air of Moscow is in many cases much more than the maximum permissible concentration. Regular measurements of NO₂ concentrations were performed from 1986 with the help of long-path laser gas analyzer at the station of optical monitoring of atmosphere pollution in the northeastern part of Moscow. Chemiluminescent gas analyzers were used for simultaneous measurements of NO, NO₂, and NO_x at the heights of 3 m and 82 m above the ground level. The ratios NO₂:NO_x at the high level were in all cases more than at the low level except for the days with the winds from the forest-park zone of Moscow. The results permitted the estimation of the rate of NO to NO₂ conversion assuming that sources of NO_x emission are placed at the surface (automobile transport). Mean value of NO to NO₂ conversion rate is 1.24 + 0.57 hr$_-1). This value may be used for model calculations of NO₂ contents in an urban air. The revealed tendencies in the grade of conversion NO to NO₂ according to seasons and day to night changing promotes to enhance the reliability of NO₂ emissions estimation in Moscow and other cities. The research was supported by US DOE award.

This paper delas with the remote sensing system dDIM, which operates in the pulse mode using tunable diode lasers. The dDIM system enables the simultaneous sensing of two gaseous pollutants through the parallel operation of two diode lasers using the same optical path. The respective laser pulses are offset in time. The special features of this system are its compact size (325mm by 142mm by 375mm) and its stand-alone, mains-independent operation, which permits flexible and mobile use. This paper introduces the initial results of the application of dDIM in measuring methane while monitoring a landfill. The measurements were undertaken in comparison to a conventional measuring device. The influence of band overlapping in measuring methane at atmospheric pressure will be further dealt with using simulations.

COSPEC (COrrelation SPECtrometer) remote sensors constitute a family of spectroscopic instruments that was developed in Canada in the 70's by Barringer Research Ltd. The last and also the best commercial model of this family was the COSPEC V (1981). When COSPEC V is utilized as a passive remote sensor, it uses a part of the UV-Vis spectral content of the zenith sky natural background to detect the presence of SO2 or NO2 traces in the atmosphere. COSPEC instruments became quite popular in the 70's and 80's and they were used especially in studies on air pollution coming from the industrial sector. Their most common application was the tracking from ground level of SO2 enriched industrial plumes. Step by step, the parallel growing of LIDAR's family, with more and more powerful instruments, together with the COSPEC's own technical limitations have driven this sensor to be unfairly forgotten. In the last few years and in spite of all these problems, COSPEC V has been widely used again by our research group in different projects related to transport and diffusion of air pollutants. In these occasions, this instrument has not been applied to study single industrial sources but urban atmospheres rich in NOR, where NO2 is the suitable target gas. The results of these studies have demonstrated the big potential of this remote sensor and possibly the necessity of its technical improvement and renewal. In this work, the most important instrumental problems of the original COSPEC V together with the optical and electronic solutions introduced are presented. As a result of the COSPEC's redesign, a new instrument has been developed: DICOSPEC (DIgital COrrelation SPECtrometer), whose performance is clearly better than COSPEC V's and whose handling by non-experts has been proven to be easier.

Effective measurement techniques are required to assess the environmental impacts of various emission sources in industry and agriculature. Open-path Fourier-transform-infrared- spectroscopy is an efficient method especially for diffuse source monitoring. Based on measurments with the high-resolution K300 spectrometer operating inside a mobile laboratory for air pollution remote sensing, path-averaged concentrations of CO₂, N₂O, CH₄, NH₃, CO, NO, and HCNO in extended area plumes are retrieved. The analysis software uses radiative transfer line-by-line calculations and least-square fit procedures. Emission rates are calculated from these values and simultaneously recorded meteorological data using a Gaussian plume dispersion model. The methodology is discussed and examples are given for emission rate estimates including industrial and agriculatural areas, highways, waste disposal sites, and forest fires.

There was an unusual method of gas equipment remote control realized during field proved at 1994 at the outskirts of St. Petersburg's. The gas equipment worn and infringings of the technology made the gas station in the field at high risk of fire or air toxins. Escaping gas can be detected only with an apparatus which either controls the element or exercises control over the general arrangement. Therefore, it is appropriate to equip an operator with a device, which he would use at any instant and at any lcoation. The hand-held thermovision camera makes possible more detailed remote observation of the hard-to-reach spots independently of natural illumination. It shows the smallest temperature differences in elements of equipment to the operator and enables him to detect internal flows.

COMPAS is a decision support system designed to assist in the assessment of the consequences of accidental releases of toxic and flammable substances. One of the key elements of COMPAS is a feedback algorithm which allows us to calculate the source term with the aid of concentration measurements. Up to now the feedback technique is applied to concentration measurements done with test tubes or conventional point sensors. In this paper the extension of the actual method is presented which is the combination of COMPAS and an optical remote sensing system like the KAYSER-THREDE K300 FTIR system. Active remote sensing methods based on FTIR are, among other applications, ideal for the so-called fence line monitoring of the diffuse emissions and accidental releases from industrial facilities, since from the FTIR spectra averaged concentration levels along the measurement path can be achieved. The line-averaged concentrations are ideally suited as on-line input for COMPAS' feedback technique. Uncertainties in the assessment of the source term related with both shortcomings of the dispersion model itself and also problems of a feedback strategy based on point measurements are reduced.

Deposition processes are a very important part of the atmospheric dispersion models. The removal from the atmosphere of the different chemical compounds which are deposited over different surfaces which are found in a complex environment such as urban centers and surroundings. Mesoscale transport models use the deposition modules to remove pollutants from the atmosphere and the quality of the deposition parameterization processes is out of the question. The classical approach of deposition processes into the dispersion models has been based on the so-called 'big-leaf' approach which is based on a former version of the resistance approach introduced in the 1970's. This is based on the accurate parameterization of the resistances, particlarly on the canopy resistance. The characterization of the different land-use types and the quantification of the different parameters for the different chemical compounds is based on field experiments carried out with ighly sophisticated instrumentation usually based on the eddy-correlation technique. In this contribution, we present first preliminary results on a field deposition experiment based on the data obtained from the eddy-correlation and modified Bowen-ratio technique for ozone (eddy-correlation) and SO₂ and ammonia (Bowen ratio). The location of the field experiment is a military airport (Cuatro Vientos) located 10 km to the west of the Madrid urban area. Deposition velocities are modeled following different approaches based on the 'big-leaf' concept. Ammonia emission/deposition processes are parameterized focusing on the 'canopy compensation point'.

A very important element in the description of the atmospheric chemistry is an accurate spatial and temporal inventory of the different compounds. Some of them are passive (SO₂) and others are very reactive (NO_x, VOC's). In this contribution, the anthropogenic emissions are usually accounted following the CORINE and EPA methodologies. In this paper, we present an accurate point, line, and area emission distribution over the Madrid mesoscale urban and suburban area which comprises 80 by 100 km² with 2000 m resolution. A special part of the inventory is that which is produced by the traffic. We have considered different types of vehicles, emission factors, driving modes, and pollutants. To account for the biogenic hydrocarbon emissions produced by the different vegetation canopy environments we are using data from the thematic mapper Landsat-5 satellite which has 30-m resolution. A Bayesian algorithm enalbles us to cluster the different spectral signatures into several groups which can be identified as different land-use types. Isoprene and monoterpene emissions are strongly dependent on the type of forest, deciduous or coniferous, intensity of the light and leaf temperature which depends on the net radiation balance. The integrated environmental system NUFOMO is used to compare emission results from handmade land-use and satellite land-use classification.

Photochemical air quality models provide the most defensible method for relating future air quality to changes in emission, and hence are the foundation for determining the effectiveness of proposed control strategies. In this contribution, we will show results from different photochemical schemes under typical emission conditions for a summer day in the Madrid mesocsale urban area. We will show that complex numerical integrated urban mesoscale models are a powerful tool to predict the ozone levels on this area. The comparison of model simulations at different grid points show an acceptable preliminary behavior. The results presented in this paper are prepared for August 15th, 1991 and the predicted ozone values are compared with those measured at two stations of the Madrid city monitoring network. Results show that the shape is successfully predicted by using the NUFOMO (numerical photochemical model) model. Because of the computer limitations, we have limited the results to this case study. Further investigations will provide additional information to produce a statistical analysis of the results. However, preliminary results show that the NUFOMO model is able to reproduce the measured ozone values.

The transport of air pollution from local sources (industrial stacks) is connected with the motion of air masses. We report an investigation of the air quality in the region of Silistra (Bulgaria)-Kalarash (Romania), characterized by the possibility of transborder transfer of air pollutants. We applied a combined technique of lidar sounding and mapping of the aerosol field and conventional sampling of the contaminating harmful substances at stationary stations. The concentrations of pollutants above the limit value were measured. We identified and localized the unknown emission souce of the organochlorine compounds and proved the transborder origin of the pollution in the town of Silistra.

Riga, the capital of Latvia is a city with nearly 900,000 inhabitants and various highly concentrated industries. Air pollution in Riga is a serious problem affecting health and damaging valuable buildings of historical importance, as acid rain and smog take their toll. Therefore the Air Quality Management System with significant assistance from Swedish Government and persistent efforts from Riga City Council was arranged in Riga. It contains INDIC AIRVIRO system which simulates and evaluates air pollution levels at various locations. It then processes the data in order to predict air quality based on a number of criteria and parameters, measured by OPSIS differential absorption instruments, as well as data from the Meteorological Service and results of episodic measurements. The analysis of the results provided by Riga Air Quality Management System for the first time allows us to start comprehensive supervision of troposphere physical, chemical, and photochemical processes in the air of Riga as well as to appreciate the influence of lcoal pollution and transboundary transfer. The report contains the actual results of this work and first attempts of analysis as well as overview about activities towards research and teaching in the fields of spectroscopy and photochemistry of polluted atmospheres.

Instantaneous air pollution fields in towns have mosaic structure and to measure its probability function one needs a measuring network with a great number of sensors. The problem cannot be solved purely by means of computer simulations because of the turbulent character of the air flows among buildings. In this paper we studied this problem by using lidar measurements and computer simulations simultaneously. The lidar data was used for the model parameterization. To simulate the atmosphere transport of the pollution puffs over city buildings we developed the special computer trajectory model. It takes into account the processes of plume widening while spreading over obstacles. The method is easily adjusted on the base of a plume geometrical and transparency lidar measurements. It uses the approximation of a quasipotential flow. The model accounts the widening of a plume flowing around buildings through effects of air volumes scattering on the potentials that represent building configurations. This method was used to study air polluted zones from one of Moscow's heat power stations.

The aircraft laboratory AN-30 'OPTIK-E', constructed at the Institute of Atmospheric Optics, is capable of measuring gas and aerosol composition of air, hydrosol, oil film, and spectral characteristics of underlying surface. In 1989-1991 tha airborne observations were performed using the AN-30 at the Baikal Lake, in Kamchatka, Buryatia, and a number of cities in Russia and Kazakhstan. The totality of experience of its operation has made it possible to develop an origianl technique for airborne ecological sounding, which gives a 3D representation on the distribution of air pollutants.

Some results of experimental investigations of photochemical processes in air over industrial centers of the Far East of the USSR are presented. It is shown that an almost permanent presence of two temperature inversions (near-ground and elevated ones) in the atmosphere over this region results in the formation of two layers of pollutant over the cities. The first layer accumulates atmospheric emissions from low-altitude sources and the second, from high- altitude sources. The photochemical change of the primary emissions into the secondary products takes place in the upper layer within the 200-400 m altitudes where enhanced content of photochemcially active gases takes place. There are no photochemical processes in the near- ground layer because the elevated layer of pollutants screens the UV radiation. However, the secondary products have been detected near the ground. They come here from the upper level due to sedimentation at night. It was noted that over the cities the secondary products of the near-ground layer are rapidly destructed, while outside the cities they live for a longer time. This is caused by the destruction of the secondary products by the atmospheric aerosol especially by its photochemical fraction.

Simultaneous water vapor measurements from satellite and ground based systems, have been performed in Napoli and Potenza throughout the last two years. Night-time water vapor profiles were provided by two lidar systems based on the Raman technique. The lidar system in Napoli, Italy is based on Xe:F excimer laser ((lambda) equals 351 nm), the system in Potenza, Italy is based on a Nd:YAG laser operating both on the 2nd (532nm) and 3rd (355nm) harmonics. Precipitable water (PW) measurements were provided during daytime by a grating spectrometer observing solar spectral irradiance at differentially absorbing channels in the wing of (rho) (sigma) (tau) water vapor band. Night and daytime PW measurements were provided using TOVS and AVHRR packages flying on NOAA satellites. Split window technique on two differentially absorbing AVHRR channels in the 11 micrometers region has been used in order to provide PW amounts with a spatial resolution of approximately 2 Km. The standard ITPP processing for TOVS package has been used in order to obtain PW estimates at lower spatial resolution (about 40 Km). Night and daytime radiosonde measurements have been performed during the last part of the campaign. All the simultaneous independent measurements have been analyzed and the results compared.

In this work the stratosphere was used as archive of former tropospheric conditions. Laboratory measurements (GC/MS) of CF₄, C₂F₆, and SF₆ were carried out on stratospheric air samples obtained by cryosampling between 1983 and 1995. Characteristic relations between height and age of stratospheric air were derived from SF₆ values. For heights above 25 km ages of 6-7 years for mid-altitudes and 7-9 years for the arctic winter stratosphere were found. By correlating this age information with mixing ratios of the perfluorated carbons increase rates of 1 ppt/y for CF₄ and 0.09 ppt/y for C₂F₆ are determined for the period 1980-1995, which translates into annual injection rates of 14000 t/y and 2000 t/y respectively. Assuming aluminium production as the only major source average emissions per ton of primary aluminium are thus calculated to be 0.75 kg for CF₄ and 0.11 kg of C₂F₆ in 1990. The mass ratio of CF₄/C₂F₆ in the atmosphere is observed to have decreased from well above 30 in 1980 to below 20 in 1995.

Some results of six years observation of the stratospheric ozone over Tomsk city (Western Siberian, 56.5N, 85.1E) are presented. The observational period covers time from December 1988 til May 1995, including Pinatubo volcano eruption. In our study we used two receiving telescopes with 1-m diameter and 2.2-m diameter mirrors and XeCl laser with hydrogen Raman cell. It was detected, in particular, the appearance of the volcanic aerosol cloud in the stratosphere led to a partial destruction under possible effect from Pinatubo eruptive cloud are discussed in detail.

The increase of particle surface-area concentration in the stratosphere after strong volcanic eruptions is believed to exacerbate the manmade ozone depletion because of the modified conditions for heterogeneous chemical reactions. To quantify this effect longterm observations of both particle surface-area and ozone concentrations are performed since the eruption of Mt. Pinatubo in June 1991. Surface-area concentrations are measured with a Raman elastic- backscatter lidar at Geeshacht (53.5 degrees N). Ozone measurements are made with ECC sondes by the German Weather Service at Lindenberg (52 degrees N). Surface-area concentrations of the order of 20 micrometers ²cm^-3 have been found between 10 and 20 km height during the first two years after the Mt. Pinatubo eruption. In the same height region, ozone losses of more than 20% during winter and spring seasons 1991/92 and 1992/93 have been observed.

In frame of building up the high-altitute environmental research station Zugspitze/Schneefernerhaus a measurement system for the investigation of trace gases on the basis of FTIR spectrometry is developed. A spectrometer with a spectral reolution up to 0.0014 cm^-1 is the main part and first solar absorption spectra will be shown. Complementary a transportable spectrometer at spectral resolutions 0.06 cm^-1 was used in the last years. These spectra are interpreted by the SFIT software package to retrieve atmospheric column densities of HCl, HF, N₂O, O₃, H₂O, HNO₃, and other gases. The results will be discussed in connection with spatial and temporal atmospheric variations.

Spectrometric measurements of column NO₂ abundances with a zenith viewing spectrometer working in 435-450 nm wavelength range, and a special numerical technique are used to retrieve NO₂ vertical distribution in the stratosphere and to estimate the level of tropospheric NO₂ contaminations. We present: 1) results of systematic NO₂ measurements at Zvenigorod Scientific Station since 1990, and 2) meridional distribution of NO₂ in the Atlantic Ocean measured from a ship in November 1989. The low- tropospheric NO₂ contents at Zvenigorod during pollution episodes can considerably exceed stratospheric columnar contents. Stratospheric NO₂ decreases dramatically in 1992, after the Mt. Pinatubo eruption, just in the volcanic aerosol layer. Measurements of NO₂ in the Atlantic Ocean reveal the latitudinal structure of stratospheric NO₂; tropospheric NO₂ contents are usually negligible but reach significant values when the ship approaches continents or intensive sea roads.

A relatively simple eye-safe compact GaAlAs lidar with solid state elements for the indoor/outdoor detection of aerosol/dust pollution layers and measuring its range and height has been developed and tested. An important feature of this full trouble-free instrument is the low eye-safe energy level compatible with ANSI standard, low mass, and low voltage circuits with only digital exchange technique, simplicity, and reliablilty. In active operation mode the lidar produces the backscatter coefficient profiles within a hundred meters and estimates of atmosphere turbidity over the road. In a passive mode it can measure a solar radiation flux. An application of this lidar's data for the model tuning in situ is to predict heavy toxic aerosol plume spreading from low sources over the city building or the territory with complex terrain is illustrated.

The data analyzed on the direct scattering coefficient of the aerosol 'dry matter' (mu) (45 degrees), parameters of condensation activity of aerosol particles (gamma) , and thermo- optical parameters F equals (mu) (25 degrees C)/(mu) (100 degrees C) and Q equals (mu) (250 degrees C)/(mu) (100 degrees C). Resulting from the analysis of the power spectrum of the scattering coefficient (mu) (45 degrees), it is revealed that the principal frequencies of this parameter are observed both on the inner-day values (12 and 24 hours) and on the periods of approximately 2, 3, and 8 days. It is more reasonable to relate them with synoptic processes. Thermo- and hygro-optical parameters (gamma) , F, and Q have the more weak variations than (mu) (45 degrees).

Continuous monitoring of Raven's seaboard since 1978 has evinced high nocturnal ozone concetrations. The present study of the averaged ozone values at three monitoring stations in the summers of 1993 and 1994, when there were no significant variations in precursor emission sources or in the usual weather conditions, shows nocturnal minima generally occurring between 3 o'clock and 7 o'clock am local time in a 20-30 (mu) g/m³ range. Nocturnal values above the normal range (> 70 (mu) g/m³) has a 6% frequency and, when compared to weather conditions, occured simultaneously at all sites (spaced 5-10 km apart) and coincided with the passing of a front or a low-pressure situation.

Mt. Cimone, the highest peak in Italy's north-central Apennines, divides two distinct climatic regions: the continentally temperate Po Valley to the north and the Mediterranean Basin to the south. It is considered a nonpolluted site, and the Italian Meteorological Service Observatory atop it, which is part of the WMO BAPMoN for the measurement of CO₂, turbidity, and suspended particulate matter, has been continuously monitoring surface ozone since March 1991 together with the FISBAT Insititute. The present, preliminary study of ozone transport episodes indicates that the air masses reaching Mt. Cimone exhibit low surface-ozone concentrations under certain weather conditions, i.e. the arrival of dust-laden air from the Sahara area or in concomitance with particular low-pressure systems.

This paper presents data on multifrequency sounding of stratospheric aerosol at the wavelengths of 353, 532, 628, and 1064 nm and the results of their interpretation. The aerosol sounding has been carried out in Tomsk at the Siberian lidar station using the telescope with the 2.2 m diameter receiving mirror.

In this paper we discuss statistical properties of aerosol properties of aerosol extinction of visible and IR radiation from field measurements of spectral transmission of the atmosphere along a horizontal path. The measurements have been carried out in winter under three different types of optical weather (haze, ice mist, and snowfall). We also compare in this paper the spectral behavior of the aerosol extinction coefficients measured in Tomsk and near Moscow.

Velocity measurements of drifting atmospheric clouds have been regularly carried out by collecting imaged sites above the ground and statistical data processing of successive images. Time variations for in-plane moving with variable pre-set velocities dots, related to different parts of the imaged area, are obtained after the in-line selection of image pixels from non- collinear directions in the image plane. The temporary instability dependencies are constructed afterwards and their minima are located to determine the cloud main drift velocity direction first. New in-line data selection in the image plane, as close as possible to the already obtained velocity direction, provides the calculation of time variations for getting the velocity module for various sectors in the imaged area, possibly drifting with different velocities.

Continental Argentina and Chile are extended on a very dangerous geographic situation in relation with the ozone layer depletion. Global reduction in total ozone can be related in part with the Antarctic's ozone hole and its dynamics of dilution; this phenomena is not well studied yet. To follow this process, a network of 16 simple and portable ozone meters has been developed. The instruments, installed from Arica (18 S) to Rio Grande (54 S), work by measuring the direct solar intensity at 300 nm and 313 nm. The ratio of these two measurements was related directly with ozone contents. Measurements during 1993 and 1994 have shown that when the ozone hole reached South America, a correlated reduction in the total ozone content can be observed in latitudes as low as 35 S. The origin of the ozone depletion detected over Argentina and Chile is discussed in relation with solar flares and circulatory dynamics.

This paper describes a lidar technique for the remote sensing of microphysical and optical properties of fog and clouds. The technique is based on recovering the information contained in the multiple scattering contributions to the lidar signals. The multiple scattering contributions are measured via detection at three or more fields of view ranging from a value slightly greater than the laser beam divergence to a maximum less than the width of the forward peak of the phase function at the lidar wavelength. The inversion is performed by least squares fitting these measurements to a multiple scattering lidar equation obtained in analytic form from a phenomenological model of the scattering processes. The solutions are the scattering coefficient and the effective radius of the fog or cloud droplets. This is sufficient information to determine the parameters of an assumed gamma distribution for the droplet sizes from where cloud properties such as the liquid water content and the extinction coefficients at visible and infrared wavelengths can be calculated. Typical results on slant path optical depth, vertical extinction profiles and fluctuation statistics of clouds are compared with in situ data. The agreement is very satisfactory. Sample appliication results on monitoring the visual and infrared detection ranges through clouds are discussed.

A cloud base height indicator using a single pulse solid state laser, called ALTO, has been developped for aeronautical purposes by SOPELEM-SOFRETEC company in association with the French Meteorological Office. Meteo-France investigates a method to obtain a vertical profile of atmospheric extinction coefficient. This method is used to determine cloud height, to locate backscattering layers and to allow the identification ofthe hydrometeor nature in the lower part ofthe atmosphere (mist, fog or precipitation). The performances of this equipment in term of LIDAR capabilities were presented in detail in a previous paper 1,

A cloud observational system, which combines ground and satellite measurements, is used to characterize cloud fields over a central area in the Netherlands. The results will be used for the improvement of parameterizations in climate models. In the Netherlands a ground-based cloud detection network of 11 stations is installed. Each station consists of a lidar-ceilometer and an upward looking narrow beam infrared radiometer. The observations of the network are combined with cloud parameters derived from NOAA/AVHRR and Meteosat. Information on the actual atmospheric conditions like standard rawinsonde data, and numerical weather forecast analysis data are also used. The system aims at a detailed quantitative description of cloud cover, cloud structure and radiative properties in an area of 120 by 120 km², the typical size of a general circulation model (GCM) grid box. The cloud detection system (CDS) will be operated for a period of two years. A first analysis of the measurements from the CDS for September 11, 1994 is presented. The ground based measurements are combined with the satellite measurements. It is concluded that this combination of measurements gives a good description of the temporal and spatial variability of the clouds over the TEBEX area. The concept, to use the lidar-ceilometer measurements to derive cloud characteristics for meteorological purposes, has much potential. Thousands of lidar ceilometers are already installed all over the world. The investiments needed to use these measurements for meterological applications would be limited.

The main obstacles for a reliable cloud base height determination with a lidar system are hydrometeors and precipitation scattering the emitted laser light on its way into the atmosphere. In order to fulfill the requirements of airport and aviation safety, modern commercial ceilometers have to be designed in a way that especially takes care of this problem, and the fact that a great part of the light received by the lidar originates from multiple scattering. Results of measuring campaigns performed in Canada and Germany show that the new Hagenuk cloud height lidar LD-WH X 06 also gives reliable results in difficult weather situations. Additionally it offers features like easy servicability, high MTBF, extended maintensance intervals, and long lifetime of laser being of special interest for commercial applications. This presentation contains some examples of the measuring compaigns mentioned above along with a comparison of the cloud heights measured with two different ceilometers in Canada.

The vertical structure of the atmospheric aerosols was investigated during cable car ascents along a steep mountain slope in the German Alps. Aerosol size distributions were measured by operating two particle spectrometer probes with diameter ranges of 0.15-3 micrometers and 1-95 micrometers . Aerosol and hydrometeorits were sampled under the meteorological conditions of clear atmosphere, haze, and thick clouds. For each atmospheric layer the phase function P((theta) ) and the extinction coefficient (sigma) _ext were calculated for the visual ((lambda) equals 0.55 micrometers ), the near IR with Nd:YAG laser radiation ((lambda) equals 1.064 micrometers ) and iodine laser radiation ((lambda) equals 1.315 micrometers ) and the middle IR with CO₂ laser radiation ((lambda) equals 10.59 micrometers ) using the computer code AGAUS of the EOSAEL program library of the US Army Research Laboratory. In the visual and the near IR the aerosol extinction coefficient is not sensitive to variations of the complex refractive index. However, for the CO₂ laser radiation, the uncertainty in the values of optical constants causes large deviations in calculated aerosol extinction coefficients.

Optical remote sensing techniques play an increasing role in assessing air pollution by various industrial emission sources. The interpretation of radiation measurements of distant sources requires modeling of tropospheric radiative transfer which is determined by molecular and aerosol absorption and scattering. The paper describes a radiation simulation algorithm developed to analyze Fourier-transform-infrared-spectroscopic (FTIS) measurements of both distant hot sources and ambient air diffuse emissions with special emphasis to the influence of aerosol and cloud absorption and multiple-scattering. The retrieval software is based on line- by-line and Mie-scattering routines. It is shown, how measured FTIS plume spectra depend on different meteorological conditions. Consideration of aerosol absorption is very important for the quantification of air pollutants from measured spectra, whereas effects of multiple- scattering can be neglected in most cases.

The wind field and turbulenceof the Atmospheric Boundary Layer are important parameters for various application fields, like meteorology, atmospheric physics, environmental protection, wind-energy utilization, air-traffic control, and-so-on. Their high variability in spatial and temporal scales necessitates a fast remotesensing method. From the available techniques the Doppler Lidar has proved to be the most promising candidate. Therefore, DLR has combined its well established Laser Doppler Anemometer (LDA) and its laser-based Ceilometer with a Sonic Anemometer.

Theoretical grounds are given in this paper for two methods of determining preferred orientation of crystal particles in a cloud. The methods proposed in the paper enable one to do this in a much simpler way than it could be done when measuring full backscattering phase matrix. One of the methods proposed assumes that a polarization lidar can be rotated as a whole, while the second technique uses rotation of the polariztion plane of a linearly polarized sounding beam. Feasibility of the former technique is illustrated in the paper with the results of field experiments on sounding of a snowfall. Recent experimental studies of crystal clouds conducted with a polarization lidar capable of measuring backscattering phase matrices (BPM) have revealed the fact that preferred orientation of symmetry axes of particles in crystal clouds is very often observed to be in horizontal plane. This conclusion is drawn from the fact that off-diagonal elements of BPMs measured differ from zero. Using a model ensemble of crystal particles of axially symmetric plates and columns one can determine the direction of preferred orientation and the degree of particles orientation about this direction. For many practical reasons it is quite desirable to try to construct a technique for detecting situations in clouds under study when a preferred orientation of crystal particles occurs, which is more simple than that based on measurements of BPMs of clouds. Below we describe two possible versions of lidar measurements using a polarization lidar with a linearly polarized sounding radiation. Such a lidar can record two cross polarized components of lidar returns from scattering medium, i.e. two first Stokes parameters. One of the versions assumes that a lidar facility can be turned around the sounding beam axis as a whole, white in the second version we need to use a (lambda) /2 phase plate in the lidar transmitter to enable changes of sounding beam polarization. In order to make understanding of the techniques proposed easier, let us remind basic relationships for a polarization lidar sensing scheme.

Backscattering phase matrices (BPM) have been measured by a polarization lidar with controllable polarization of output laser radiation for measuring all the Stokes parameters at (lambda) equals 532 nm. The degree of orientation and the preferred orientation of particles are determined. To this end, the properties of BPM for the model of axisymmetric prolate particel (APP) ensemble are used. It has been suggested that scattering anisotropy of an aeosol layer, whose BPM is not described by the APP model, is caused by birefringence. In laser sensing of the atmosphere ensembles of aerosol particles being sounded are characterized by the backscattering coefficient. Very often, in addition to this parameter, the intensities of polarized and crosspolarized components of backscattered radiation are determined provided that lineraly polarized laser radiation is used. The ratio of these components is called depolarization and it is assumed a measure of the particle nonsphericity. Use of the above characteristics is based on the concept that atmospheric aerosols are ensembles of spherical or nonspherical randomly oriented particles. An experience of optical studies has shown that such a concept is quite justifiable for the majority of atmospheric aerosols. However, these exists quite a wide class of natural aerosols in the atmosphere, namely, the crystalline clouds, for which the lidar equation in scalar form is insufficient since such aerosol ensembles should be described with a backscattering phase matrix. Below we shall demonstrate this by an example. Of course, the necessity of using the BPM to describe such aerosols is, in certain sense, obvious because anomalous optical phenomena resulting from a pronounced anisotropy of light scattering by crystal clouds have been known long ago. Nevertheless, such phenomena are too rare and it is not a proiri clear how often essential deviations from the scalar approximation occur. Thus, the experimental material available for our analysis at present and partially described in references allow us to arrive at the conclusions that in 30-40 percent lidar observations of crystalline clouds either the backscattering coefficient depends on the direction of sounding radiation polarization or the polarization of scattered light becomes elliptical, or both these effects occur simultaneously.

Lidar was applied to identify atmospheric inhomogeneities by different scattering behavior of the aerosol particles. Aerosol is a general term, it includes also the aggregation to clouds. In the stratosphere there exist droplets of sulfuric acid (from volcanic eruptions), nitric acids and water vapor in mixed phases. In the polar region polar stratospheric (aerosol) clouds form during winter at temperatures around 195K. Usually the nucleation of nitric acid trihydrate (NAT) and nitric acids dihydrate (NAD) happens in a complicated mechanism. A lidar can identify the result of the nucleation process, not the gas phases. It can distinguish between droplets (as Mie particles) and crystal (frozen droplets of NAT and NAD) by polarization. Because the aerosols are driven with the wind and have a tendency to sediment, the orientation of the lidar to the aerosol particles is an important factor. Improvement of lidar measurements to distinguish between stratospheric aerosols and polar stratospheric clouds (i.e. by the given definition the frozen aggregation) was focused: 1) on the use of multiple wavelength lidar with a polarization channel, 2) on a theoretical study on the possibility to use multiple scattering as an additional discriminant and, 3) on scanning. Points 2 and 3 require a better detection and signal processing system. Statistical problems arise for the comparison of measurements, for example if one compares ground-based and airborne measurements of the same cloud. Airborne measurements can contribute to the problem because one can reduce the distance to the object and therefore the 1/R2-dependency leads to larger signals. Averaging with respect to ensemble statistics are covered in this report. It is accepted that ground-based lidar systems especially with a Raman channel measure with a high pulse repetition rate over a few minutes to get a signal which can be processed.

Time and range resolved multiple-scattered radiation from the aureole around a traveling radiation, using an additional off axis scanning lidar-receiver, on top of an existing lidar. This system can be considered as an analogue equivalent of Bissonnette's fixed multi-field-of-view lidar. Because multiple-scattering is effective over much larger volumes than the illuminating volume by the laser pulse itself; the variations in the signals due to atmospheric structures are reduced to a certain extent. A simple second-order scattering model has been developed to understand this volume averaging effect. It has been shown that this model can be used to estimate the extinction coefficient of clouds and the average diameters of the cloud droplets. Lidar waveforms, obtained during a field experiment, have been inverted with the proposed model. The results appear to be in reasonable agreement with the in situ measured extinction using a nephelometer mounted on a cable car which was moving up and down a mountain slope through the clouds.

With an accurate scanning backscatter lidar cross sections of contrails are repetitively measured. Horizontal and vertical growth and effective diffusion coefficients are determined. The lidar ratio and optical depth are calculated by means of shadow calibration.

We present some examples of calculations of polarized multiply scattered return signals of ground-based and airborne multichannel LIDARs (such as the Microlidar of the DLR) and we discuss the applied numerical methods. These methods are variance reduction Monte Carlo algorithms allowing for controlling the empirical variance. We show how variance reduction Monte Carlo algorithms allowing for controlling hte empirical variance. We show how variance reduction Monte Carlo methods may be combined with properly chosen uniform grid search or random search procedures to retrieve one or more environmental parameters from a given (measured or calculated) LIDAR return signal. Such a procedure is time consuming, of course, but it will allow for the retrieval of several parameters simultaneously and for a sensitivity analysis. A retrieval method based on Monte Carlo methods and random search including such a sensitivity analysis gives much more information than the usual inversion procedures (e.g. based on integral equations and often uncheckable environmental assumptions). We present some examples of the (simultaneous) retrieval of the extinction coefficient and the particle size distribution of a cloud from the multiply scattered return signal of gorund-based and airborne LIDARs and a sensitivity analysis of this retrieval.

A series of laboratory experiments has been conducted to investigate the initiating and guiding effects of laser plasma channel on electrical discharge. It was confirmed that the plasma channels have strong guiding effects and reduce the required electrical field strength for electrical discharges to occur. A field experimental site targeting natural lightning is being prepared to develop the thunder storm monitoring system and to test the laser and optical systems against various weather conditions. The results from the laboratory experiments and laser transmission in snowy conditions as well as attempt of initiating electrical leader will be discussed.

At a lidar sounding of atmosphere meteorological parameters, (wind velocity and direction profile) the precipitation more often is considered as a hindrance, which limits the distance and accuracy of the measurements and in some instances performs the impossible measurements. But in the course of investigations it has been found that in certain situations, the precipitation increase efficiency of lidar sounding. In this case the distance and accuracy of sounding are increased. Primarily this is true for precipitation, which has the intensity no more than 1.5 mm/h. Furthermore, the possibility of determinations of the microstucture and integral characteristics of precipitation area is existed. The investigations were performed using the correlation scanning three-path lidar with vertical scheme of sounding. In the course of measurements the spatial-temporal series reflected by precipitation optical signals were amassed. The handling of the lidar data was based on the correlation-spectral method. The study of the optical signals reflected from preciptiation showed that the increase of the wind velocity sounding efficiency is available by amplification of the signal fluctuation components. Limitations on the rain rate are caused by powerful rain. A maximum of the dropsize distribution function is displaced inside of the high quantities which are not carried by air flows and useful information for the estimation of wind velocity gives the raindrops with dimensions no more than 0.6 mm in diameter. As the lidar allows to make a vertical cut of the precipitation area, that it was seen the moving of forward and reverse fronts of the precipitation area at the spatial range of the optical signals. From a comparison of the front positions in different times the drop speed of the different rain group is determined. This allowed us to estimate the range size raindrops and raindrop size distribution function was determined. The knowledge of precipitation microstructure parameters allows obtained integral characteristics of precipitation area: rain rate (mm/h), watery (g/m³), raindrop concentration (1/m³).

The importance of calibrating optical open-path measurements of atmospheric gases is discussed. Work carried out at NPL has involved calibrating integrated-path and range- resolved measurement instruments. The designs of the calibration gas cells used for this work are described and some results are presented. These results include the calibration of a DIAL measurement system using a calibration cell in which a stable flow of the target gas is maintained.

There is a growing need for standardizing the procedure concentration data from Fourier transform remote sensor instruments. To that end, the ManTech Environmental Technology, Inc., under contract to the US Environmental Protection Agency (EPA) is preparing a document that has become known simply as TO-16. When complete, the document will become a part of a compendium of methods on toxic organic chemical species that has been compiled and is maintained by the US EPA. The document will actually be a procedure that when followed will allow the consistent production of atmospheric gas concentration data. This paper discusses the contents of the procedure. It also addresses some of the difficulties encountered in the production of such a procedure.

In this paper, a multiwavelength (355, 532, and 1064 nm) lidar sounding across a stack plume has been computer simulated. The lidar data were then inverted using Klett's iteration method. It is shown that the obtained dust concentration profile is strongly affected by both the strength of the dust emission source and the particle size dispersion. Satisfactory results were only obtained for source strengths below 0.1 g/s. For higher source strengths, (that is more dense plumes) the errors increased rapidly with the increasing plume density. We show that by multiplying the obtained dust concentration profile by a factor dependant on the optical density of the plume satisfactory results can be obtained for source strengths of 1 g/s or higher.

The main requirements for in situ trace gas analysis, process and air pollution monitoring are specifity, high time resolution, and high sensitivity. Tunable diode lasers absorption spectroscopy is increasingly being used to measure trace gas concentrations down to low ppbv- levels (10^-9 volume mixing ratio). The applications of high frequency modulation (FM) schemes can further improve sensitivity and detection speed of modern instrumentation. In this paper the FM technique will be reviewed with respect to current limitations of tunable diode laser spectroscopy. Recent investigations and findings will be described with emphasis on the work done in our laboratory and novel approaches will be discussed. Some applications related to quality control and ISO calibration as well as fast and high sensitivity measurement challenges will be discussed.

We have combined a LiNbO₃ optical parametric oscillator (OPO) and a photoacoustic (PA) open cell to perform air pollution measurements. THe OPO is pumped by a Nd:YAG laser working at 10 Hz. It is tunable between 1.55 micrometers and 4.5 micrometers . The PA cell is designed to be isolated from outside acoustic noise. The OPO output wavelength is tuned between 3.39 micrometers and 3.44 micrometers to monitor absorption lines of methane. The characteristics of the pulses are E equals 2 mJ, (tau) equals 15 ns, (Delta) (nu) equals 2 cm^-1. The minimum detected concentration is 1 ppm. Decreasing the background noise, detection of concentrations as low as 20 ppb is expected.

We report on the use of laser based cavity ring down spectroscopy in the near UV. It is this part of the spectrum that is particulary well-suited for trace gas detection, as many molecules have strong, well characterized absorption bands in this region. We show that the detection limit for e.g. NH₃ is in the order of 10 ppb, for OH below 1 ppb and for Hg at the ppt level. We propose a Fourier transform based cavity ring down spectrometer, in which the advantages of cavity ring down detection are combined with the multiplex advantage of a Fourier transform spectrometer. A theoretical description of this spectrometer, as well as a proposal for the construction of a prototype are worked out.

PbSe is a very important photoconductive material extensively used as IR detector for military applications and may be considered one of the most useful materials for detection in the MIR range. In the last years the opening of its production for wide civil use has allowed the conception of new detection systems based on this semiconductor. Considering some possible applications of it in environmental control, PbSe can provide, for instance, good response band to monitor several gases of major importance (SO₂, NO, CO, etc.), especially when their concentrations are high. In this paper, we present applications of this semiconductor for this purpose: the developemnt of a new continuous emission monitoring system (CEMS) using a PbSe detector array in a nondispersive configuration. The basics of this prototype and some experimental results related to the detection of different typical emission gases with this system are presented here.

Cost-effective measurement techniques are required to assess environmental impacts of various emission sources in industry and agriculture. Fourier-transform-infrared-spectroscopy (FTIS) has great potential in this field due to the multicomponent nature of the measurements. Many compounds can be quantified from one single infrared spectrum. Gas releases from large compost heaps strongly contribute to the emission of greenhouse gases and other pollutants. Both open-path and extractive FTIS measurements were performed using the high-resolution K300 spectrometer operating inside a mobile laboratory for air pollution detection. The analysis software is based on a line-by-line radiative transfer algorithm and least-square fit procedures. Quantitative results of several measurement campaigns to analyze compost emissions are described. Emission rates are calculated from the concentration values and simultaneously recorded meteorological data using a Gaussian plume dispersion model.

We developed a new concept for lidar measurements: multispectral differential absorption. Instead of comparing the backscattered light at one absorbed and one reference wavelengths (DIAL), a broadband laser is sent into the atmosphere and the returned signal is spectrally resolved on the receiver side (MULTIDIAL). Potential advantages are thoroughly reviewed. First, experiments combining lidar with a Fourier transform spectrometer are presented. Broadband nontracking OPOs are being built for this application in the infrared to detect VOCs. This idea was also applied to the UV range. A tunable (280-300 nm) broadband Ce:LiSAF laser has been developed for this purpose. The detection is performed by a gated amplified CCD-OMA.

A simple light scattering photometer has been developed for measuring the angular distribution of the intensity of polarized laser light scattered by micro- and sub-micron samples. The photometer uses an ellipsoidal reflector and simple optical components to collect the He-Ne laser scattered light and to focus it onto a 512 element photodiode array. Although the technique is inherently a static one, it is possible to monitor relatively slow kinetics at a rate of 25 ms per sampling period. Experimental data have been obtained for several monodisperse aqueous solutions of latex spheres of different sizes. The results have been satisfactorily interpreted on the basis of the Mie theory with suitable geometrical corrections.

A study aimed at defining an efficient monitoring technique for an industrial zone coverage, using available scanning remote sensors, is presented. Scanning IR LIDAR and FTIR systems are considered as remote sensors. An analysis to define an optimal positioning of a central system and designing optimal lines of sight distribution is described. The application of this procedure is illustrated as a case study of a typical industrial zone, located near a populated urban zone.

An optical hygrometer has been developed to make night-time measurements of water vapor in the upper atmosphere using the technique of photofragment fluorescence. Hygrometer uses the fluoroscence at wavelengths near 310 nm of excited OH molecules produced in the photodissociation of water molecules by UV radiation with hydrogen lamp (121.6 nm). A coaxial optical scheme is used. A compact hydrogen lamp is aligned along the axis of the instrument, inside the annular fast focusing optics. It provides the effective convergence of fluorescence from a large angle and thus increase the sensitivity of this method. The modulation of the hydrogen lamp with the frequency 1 kHz is used. Specially designed laboratory technique for calibration of the hygrometer is described. Total weight of the optical hygrometer is about 3 kg. It can be easily integrated with the Vaisala RS-80 radio sonde. This allows us to obtain reliable real-time data on temperature, pressure, and water vapor concentrations in the upper troposphere and stratosphere.

Recently developed mid-IR LEDs and diode lasers can be used to fabricate spectroscopic analyzers of a new generation. The LED-based design is advantageous, especially in process and portable instruments and fiber optic applications. The benefits are smaller instrument size, rugged operation without any moving parts, and high output pulse power. The availability of new ambient temperature operating LEDs based on A³B⁵ alloys in the spectral range of 2.8-4.7 micrometers and cooled diode lasers has made it possible to monitor several gases having strong absorption bands in this region. This paper briefly describes the properties and fabrication of infrared light emitting diodes and diode lasers by liquid phase epitaxy for emission wavelengths from 3.8 micrometers to 4.85 micrometers and 3-3.6 micrometers respectively. Some of the first applications of these LEDs and diode lasers in spectroscopic instrumentation (nondispersive CO₂ and fiber optic C_nH_m analyzers) are described.

The concept of surface plasmon resonance chemical sensors based on Si substrate is developed which combines a number of advantages of both optical and microelectronical approaches to investigate gaseous media. This type of the gas sensor has been made and tested for the first time. A reversible response to ntirogen dioxide at the level of ppm concentrations has been recorded.

Trail in-the-field measurements of humidity fluctuations are taken by a laser-acoustic hygrometer which involves: a CO-laser, a CO-laser, a mechanical modulator of its radiation, a receiver, an audio-frequency amplifier, and a device for processing signals. The proposed method based on measuring the intensity of sound which is generated when pulse-modulated laser radiation is absorbed in the air. The CO-laser spectrum (5.0-5.7 mkm) is within the fundamental vapor absorption band (5.2-7 mkm). This allows a medium-powered CO-laser to be employed for measurements of humidity fluctuations. Radiation with a pulse recurrence frequency of 16 kHz, which is required for generation of sound waves, is generated by an obturator modulation of the CO-laser continuous radiation. At the point of sound measurement the laser-beam power is about 3 W. An acoustic signal carrying the information on humidity is received by a 1/2 inch microphone installed perpendicular to the beam (d equals 0.5 cm) at a distance of 3 cm from its axis. The spectra of absolute humidity fluctuations over the frequency range (2^*10^-3 - 10)Hz have been calculated from the obtained realizations for atmospheric conditions when relative humidity ranges between 27 and 47%. Noise characteristics of the laser-acoustic hygrometer have been studied.

An investigation of the characteristics of multiply-scattered lidar returns from homogeneous layres of nonspherical Chebyshev particles is presented. A Monte Carlo procedure has been employed to simulate lidar measurements in a ground-based configuration. Total detected power and depolarization of the return signal have been calculated for a variety of particle sizes and deformations, as well as for different fields of view of the instrument and optical thicknesses of the medium. As far as depolarization is concerned, particles characterized by a high backscattering depolarization ratio have shown a peculiar behavior in multiple scattering. Results have been checked, for double scattering, employing an analytical formula previously developed.

A flight qualified Fourier transform infrared spectrometer has been built by applying the MIROR principle (Michelson interferometer with rotating retroreflector) where an eccentrically rotating retroreflector generates optical path differences. The unique optical design is especially suiting the rough environment of airborne missions. The purely optical and passive method in no way influences the gases as sample collecting procedures are likely to do. It is able to deliver true space/time resolved spectra of several species in the exhaust plume simultaneously. First measurements aboard a civil jet aircraft have been performed, successfully collecting spectra of the infrared radiation emitted by the hot exhaust gases just behind the engine's nozzle. The spectra were radiometrically calibrated and column densities of trace gases in the plume and in the for- and background as well as gas temperatures were calculated applying inversion algorithms. From these then emission indices (mass pollutant per mass kerosene) of the engine for specific trace gases were determined.

The propagation of an explosion caused by an error can be successfully prevented by flameproof enclosures. This paper describes investigations applying different nonintrusive diagnostic methods to the combustion process in a turbulent jet emerging from the joint gap of a flameproof enclosure model. Experiments are carried out using a tuneable excimer laser operated either with ArF (193 nm) or KrF (248 nm). Rayleigh scattering is used to determine temperature distribution in the free jet, and laser-induced fluorescence of OH is used to observe the combustion and flow processes. The enclosure consists of two explosion vessels which are connected via a nozzle of variable dimensions. H₂/O₂/N₂ mixtures with a stoichiometric H₂/O₂ ratio are investigated. The position of the ignition source in the smaller vessel is made use of to vary the explosion pressure prevailing in the enclosure at the moment when the flame front enters the joint. Prevention of an outside explosion depends on both the nozzle dimensions and the upstream explosion pressure at the moment when the flame front enters the joint. In the range of application of flameproof enclosures, the outside ignition takes place as a result of a flame quenching process in the nozzle and subsequent re- ignition in the turbulent mixing zone of fresh and exhaust gas that has escaped from the joint gap into the environment.