Light detection and ranging (lidar) systems, which are based on the emission of a laser pulse to remotely sense physical phenomena in the atmosphere, have been used in various parts of the globe for the remote sensing of smoke emanating from plants, factory chimneys, ships' funnels, and others for a few decades.1, 2 In Africa, however, the technical means to monitor atmospheric pollutants and climatic changes are lacking at present. This motivated our team to cooperate with a lidar group in South Africa on research in this area. In this way, we are developing in Algeria an innovative mobile system containing two separate lidars. This dual system has several advantages relative to the use of a single lidar. For example, the two devices share some components, which makes the system cheaper, more portable, and versatile. Most importantly, it allows for the use of a single apparatus for more than one application.
In our dual system, one of the devices is intended for the early detection of fires in the continuous mountainous forests in northern Algeria, especially in the hilly and inaccessible zones where the risk factor of a forest fire is high in the summer. At present, fire fighters use ocular surveillance to detect fires in those regions, meaning intervention occurs only between one and three hours after the start of the fire. With the lidar technique, we are working to decrease this interval to a few minutes. The second lidar is planned for atmospheric studies of pollutant aerosols, especially those emanating from two cement plants in the Algiers metropolitan area and sand aerosols from the nearby Sahara desert. This is a serious public health concern in Algeria as the rate of respiratory allergies is increasing every year. Another planned environmental application of this second device is the analysis of water vapor in the tropopause.
Due to the distinct applications, the two lidars use two different lasers. The one for forest-fire detection is eye safe with 12mJ of energy at 1.57μm. The atmosphere-sensing device is more energetic to account for the weakness of the return pulses from the upper atmosphere: we use the second harmonic of a neodymium-doped yttrium-aluminum-garnet laser at 532nm. The two lidars use the same Newtonian telescope as it is a heavy component. The transmission and reception optics have parts shared between the two systems, though a scanning mirror is provided for the forest-fire lidar: see Figure 1. We also use different detectors for the two devices: an avalanche photodiode for the forest-fire lidar and a photomultiplier for the atmosphere device. The transient recording systems are different too. That for smoke sensing is rapid: it is based on one-pulse detection followed by pattern recognition.3 The atmospheric sensing, on the other hand, will be achieved by accumulating signals over time.4
Figure 1. Lidar setup for forest-fire remote sensing. The beam expander reduces the laser divergence for better sensing efficiency, while we use mirrors 1 and 2 for steering and alignment. In addition, the rotating reflector is scanned for horizontal sensing. f: Focal length. APD: Avalanche photodiode.
Though researchers have previously proved the effectiveness of the lidar technique for the detection of small forest fires,5 we conducted experiments to show our dual system would also be effective for this purpose. We used a ruby laser—which, compared to other lasers, has spikes that are temporally well spaced and less attenuated over the whole relaxation wave train—and carried out experiments in two regimes: free running and Q-switch. The former has not been used so far for lidar measurements because of known drawbacks.6, 7 However, we are looking into using the same laser in Q-switch to sense and in free running to check. Indeed, once we resolve the problems of the low-range resolution and the overlap between ongoing spikes and return signals inherent to the free-running regime, it will be more advantageous to check for smoke occurrence with a train of spikes (free running) rather than with just one pulse (Q-switch). During the sensing process, the Q-switch remains the preferred operation method due to better energy stability.
One experiment showed that it is not possible to detect tenuous fire smoke when the first lidar is in a transmission configuration due to the noisy reel environment, meaning it is not feasible to detect small forest fires using a rangefinder technique. In another experiment, with the lidar in a scattering configuration, we found the backscattered measured signal to be at least four times more sensitive than a right-angle scattered signal. In a further step, we established an equivalence relationship between the laboratory experiment and a real lidar experiment.8 More precisely, using a 0.3m-diameter receiving telescope and an indium-gallium-arsenide avalanche photodiode with a current gain (multiplication factor) of 12, we anticipate the measured signal to be as noticeable as that measured in the laboratory. These results show the effectiveness of the lidar technique for the detection of tenuous smoke emanating from small forest fires.
Given the applications planned for the atmospheric lidar, laboratory experiments are unfeasible. We will soon be testing this second device in the field, where we will sense the atmosphere for polluting particles by accumulating lidar signals during a long period to monitor temperature changes, extinction factors, and aerosol concentration and size over an extended time.
In summary, we are developing a dual lidar laser system to detect forest fires, monitor pollutant aerosols, and perform atmospheric analyses. The device is intended as a tool to reinforce collaborative work between Algerian and South African research teams. In the future, we plan on common use of the dual system in the field to fulfill its environmental-research purposes in the two countries.
This work is the Algerian contribution to a collaboration between our research team and V. Sivakumar's group at the University of Kwa-Zulu Natal, Durban (South Africa), within the framework of the African Laser Centre. We would like to thank V. Sivakumar for fruitful discussions and his suggestion to write this article.
Mohammed Traïche, Abdelkrim Kedadra
Centre for Development of Advanced Technologies (CDTA)
Mohammed Traïche graduated with Magister and PhD degrees in quantum electronics from Algeria's University of Sciences and Technology Houari Boumediene in 2000 and 2013. At CDTA, he has worked on laser resonators, solid-state lasers, and on a lidar development toward environmental applications.
Abdelkrim Kedadra graduated with a PhD degree in electrical engineering from the École Centrale de Lyon, France, in 1996. He is currently working on lidar development at CDTA.
1. P. M. Hamilton, The application of a pulsed-light rangefinder (lidar) to the study of chimney plumes, Phil. Trans. Roy. Soc. Lond. A 265, p. 153-172, 1969.
2. E. D. Hinkley, Laser Monitoring of the Atmosphere, Springer, 1976.
3. K. S. Fu, Pattern recognition: a potential tool for data analysis in atmospheric laser spectroscopy, Opt. Quantum Electron. 8, p. 169-183, 1976.
4. V. Sivakumar, M. Tesfaye, W. Alemu, A. Sharma, C. Bollig, G. Mengistu, Aerosol measurements over South Africa using LIDAR, satellite, and sun photometer, Advances in Geosciences: Atmospheric Science, p. 253-262, World Scientific Publishing Company, 2010.
5. A. B. Utkin, A. V. Lavrov, L. Costa, F. Simões, R. Vilar, Detection of small forest fires by lidar, Appl. Phys. B 73, p. 77-83, 2001.
6. C. A. Northend, R. C. Honey, W. E. Evans, Laser radar (lidar) for meteorological observations, Rev. Sci. Instrum. 37, p. 393-400, 1966.
7. B. R. Clemesha, G. S. Kent, R. W. H. Wright, A laser radar for atmospheric studies, J. Appl. Meteorol. 6, p. 386-395, 1967.
8. M. Traïche, K. Bouraï, N. Moussaoui, R. Beggar, F. Almabouadda, D. Louhibi, Lidar technique for early forest fire detection: design and development aspects, AIP Conf. Proc. 1047, p. 50-58, 2008.