ORSL The Optical Remote Sensing Laboratory of The City College of New York
138th St. & Convent Ave., New York, NY 10031

ORSL version 2.2.0
ORSL » Atmospheric Research » Research Areas

Index of Atmospheric Research Areas

The purpose of the Atmospheric Research ORS Lab is to monitor and investigate the optical properties of the atmosphere. This is accomplished with remotely-sensed data, received from operational and research satellites, and stationary instruments deployed at observation sites.

Some of our active development efforts to augment and expand our capabilities and understanding are outlined herein:

PageResearch Area
  1  Eye-safe Heterodyne Detection Coherent Doppler Lidar
  2  Mid-IR Quantum Cascade Laser for LIDAR Application

Eye-safe Heterodyne Detection Coherent Doppler Lidar

Sameh Abdelazim, David Santoro, Mark Arend, Fred Moshary, Sam Ahmed


Coherent pulsed LIDAR is receiving increasing attention as a method for detecting aerosol concentration in the air and detecting wind speed. Wind speed detection, in particular within urban areas, is essential to modeling air flow patterns to analyze pollution transmission and determine, for example, optimal locations for wind turbines. City College of New York is currently developing a mobile coherent Doppler LIDAR station to detect wind speed. In Doppler sounding with coherent LIDAR, pulses are transmitted into the atmosphere. These pulses reflect off aerosol particles in the sky and return to the system. The motion of these aerosols can be measured based on the Doppler shift of the wavelengths transmitted. With a mobile system, it is possible to point at the same location from three or more different directions, and thus to calculate an accurate vector wind speed for the area.

The station in development at City College of New York uses a 1.54 m near-infrared pulsed beam and polarization maintaining fiber optics. This wavelength is safe to the eye and provides an efficient balance of back-scattering and absorption. Signal analysis and detection is accomplished using heterodyne detection to supress RIN noise. In this process, two signals of slightly different frequency are created. Shifting can then be detected by comparing the initial modified frequency to the return frequency. A/D conversion is performed by using a data acquisition board with FPGA on-board. Finally, a basic wedge prism i0073c is used to control the zenith angle of the beam as it is transmitted to the atmosphere. This will enable data to be taken from a variety of directions. Real-time scanning will be accomplished by a computer-controlled dual-axis knuckle assembly to provide both azimuth rotation and elevation beam positioning.

The system consists of the following components: laser source, modulator, fiber amplifier, optical antenna, detector, and signal processor. These components and their relationships are shown in the second image.

Wavelength Selection

    Aerosol transmission rises with increasing wavelength, however, aerosol backscattering coefficient rises with shorter wavelengths.
    In the NIR band, spectral absorption takes place mainly due to CO2 and water vapor.
    Inexpensive key LIDAR components that operate at 1.5µm have been obtained easily due to their wide usage within the telecom industry.

Laser Power Analysis

    By increasing local oscillator power (PL), shot noise from local oscillator dominates thermal noise on load impedance
    Low level of PL causes thermal noise to dominate over shot noise, and optical efficiiency suffers.
    RIN noise can be reduced by a factor of RB (RIN suppression ratio) through the use of a balanced detector.
    The efficiency on power penalty can be plotted as a function of local power for RIN values of between -140 db/Hz and -170 db/Hz, shown in figure 3.
    As RIN is frequency dependent, we measure RIN versus frequency in the band of interest to us (50 to 110 MHz).
    Figure 4 shows that the measured Rin is less than -152 dBm/Hz in band, therefore operating with about 12 dBm LO power on each detector of our balanced detector units should be optimum and provides us with a few dB of margin in available power.

System Parameter Tests

    The reflected optical signal was simulated by using optical attenuators between the AOM and the 50/50 coupler representing the attenuation that the scattered laser suffers through the air before it is received by the telescope.
    The output signal of the optical receiver was measured using ADC.

Signal Processing

    In order to achieve high speed Doppler signal processing, we use an analog to digital board that runs at 400 MHz with FPGA (Field Programmable Gate Array) on-board.
    The FPGA is programmed to calculate autocorrelation which can be averaged from shot-to-shot.
    The modulator is triggered at 10-20 KKz and gated for 200 ns pulses, which results in a range resolution of ~30m. We anticipate to reach a range of 2-5 km.

In future work, we will compare our results at various locations around the New York City area, specifically at or near sites where we have other wind measurement vertical profilers by expanding the system to be mobile and multi-directional and using our mobile Lidar van. We can then fully analyze wind speed and aerosol density in a 3-D spacial manner.


  • Abdelazim, S., Santoro, D., Arend M., Moshary, F., Ahmed, S. "All-fiber Coherent Doppler LIDAR for Wind Sensing."
  • Abdelazim, S., Santoro, D., Arend M., Moshary, F., Ahmed, S. "Wind velocity estimate and signal to noise ratio analysis of an all fiber coherent Doppler Lidar system", 16th Coherent Laser Radar Conference.
  • Cariou, J., Boquet, M. "LEOSPHERE Pulsed Lidar Principles", UpWind WP6 on Remote Sensing Devices, http://www.upwind.eu/media/576/D6.1.1.pdf.
  • Ostaszewski, M., Harford, S., Doughty, N., Hoffman, C., Sanchez, M., Gutow, D., Pierce, R. "Risley Prism Beam Pointer", Free-Space Laser Communications VI, 2006.

Mid-IR Quantum Cascade Laser for LIDAR Application

Morann Dagan1, Ihor Sydoryk1, Fred Moshary1, Barry Gross1


LIght Detection and Ranging (LIDAR) instruments have proved to be powerful for atmospheric research and environmental monitoring [1]. LIDAR systems are used to build atmospheric profiles by collecting the backscatter signal from molecules and particles in the atmosphere. A LIDAR system consists of a transmitter which generates a pulsed signal at a specific wavelength, a receiver to collect the backscatter signal, and a computer system which digitizes the signal as a function of time or range [2]. The wavelength chosen for the transmitter depends on what you want to see in the atmosphere. To see the backscatter signal from the larger particles, a longer wavelength should be used and vice versa. Choosing a wavelength in the mid infrared (MIR) band would be beneficial for detecting aerosols and characterizing cloud drop size below the cloud [3].

Feasibility Study

The drawback of a MIR LIDAR system is the low signal to noise ratio (SNR). This is due to a low backscatter in the MIR band (3 m 8 m) compared to the visible (380 nm 750 nm) since SNR is inversely proportional to wavelength. Based on SNR estimates we explored the viability of a Mid-IR QCL LIDAR, including calibration issues. The measurement sensitivity typically depends on the amount of light that is backscattered, collected, and focused onto the detector. This is dictated primarily by the backscatter coefficient of the scattering source. Typical LIDAR systems detect elastic backscattering from atmospheric molecules (Rayleigh scattering) and aerosol particles (particle or Mie scattering). In the mid-IR elastic backscattering is dominated by coarse mode particulate scattering because of the 1/λ4 wavelength dependence of Rayleigh scattering. Particle backscatter efficiencies depend strongly on the atmospheric density of aerosols (1/ λ2); however, in the MIR, backscatter efficiencies tend to be very small. In order to improve the low SNR, averaging pulses over time helps reduce the noise and therefore increases our SNR. Figure 3 shows the SNR values of the potential LIDAR system, where an SNR of approximately .4 (should be 1) at 1km can be achieved for MIR channel (4.55 um) on a clear day after 30 minute averaging. Figure 4 shows the SNR of the system when looking at clouds, where the SNR is improved by more than a factor of ten for one minute averaging.

Figure 1a Figure 1b


At CCNY, we are building a MIR LIDAR to improve and better understand the atmospheric profile. By using the Mid-IR LIDAR data along with the visible (VIS) LIDAR data, it would make it possible to separate the fine from the coarse aerosols which is critical in interpreting anthropogenic aerosol sources. Further, using both data sets will help in characterizing aerosol-cloud interaction and cloud drop size below the cloud. Backscatter data in the VIS, NIR (near IR), and MIR (mid IR) is able to distinguish these modes much better than backscatter measurements in the VIS and NIR only.

Prototype Design

The MIR LIDAR is being designed to be mobile; therefore our system is small and compact to make it portable. Figure 2, below, shows our current system configuration. This LIDAR is capable of running anywhere with power available since the entire system is situated on a cart for easy transportation. We are able to take data at any angle, making our LIDAR functional for many applications. Our transmitter is a Quantum-Cascade laser (QCL) with a wavelength of 4.55 microns, a pulse width of 202 ns and a peak power of 4.5 W. We chose a QCL at that wavelength since it falls within an atmospheric window (wavelengths at which electromagnetic radiation can transmit through the atmosphere to the surface), see figure 1. Quantum-Cascade lasers are particularly suitable in this wavelength range, as they offer several Watts of pulsed optical power, while retaining a good far field pattern as required for laser remote sensing techniques in atmospheric research. Our receiver is an f/5 Newtonian telescope with a 10 inch primary mirror. Our data is collected using a detector with a spectral range of approximately 2.5 5 m and a spectral response, D*, of 9.25x1010. Once signal is collected, it is digitized using a 12 bit Gage digitizer and then analyzed using LabVIEW and MATLAB. At 100 KHz repetition rate, our system range is 1.5 km with 60 m resolution.

Preliminary Results

Currently, we are working on the final calibrations of our LIDAR system. We have so far collected hard target signal from distances up to 800 meters away. Figure 4 shows examples of backscatter signal off buildings 90 - 250 meters away. These figures average the signal 256 times to increase the SNR. In the top plot of figure 4 a clear signal can be seen from a building 250 meters away. The lower plot shows a signal coming from a building 90 meters and another 150 meters away. However, based on the SNR calculations above, we will probably need to average over 30 minutes to see a signal from aerosol particles.

Future Work

Our next crucial step to complete our system is to improve detector alignment. To do this, we plan on redesigning the layout of our detector by replacing our relay lens with a folding mirror and adding a xyz stage for the detector. Once we have better alignment we hope to start picking aerosol and cloud signal. With a working system, we can then consider revising it with a remote controlled Newtonian telescope for easy angle adjustments or coming up with a design to thermo-manage our system giving us the possibilities of running our system at lower temperatures.


  • [1] Hinkley, E.D., Kelley, P.L. "Detection of Air Pollutants with Tunable Diode Lasers. Science 171, 635-639, (1971)
  • [2] Kovalev, Vladimir A., and William E. Eichinger. Elastic Lidar: Theory, Practice, and Analysis Methods. Wiley Interscience, 2004. Print.
  • [3] Thrush, E., Salciccioli, N., Brown, D.M., Siegrist, K., Brown, A.M., Thomas, M.E., Boggs, N., Carter, C.C. "Backscatter signatures of biological aerosols in the infrared." Applied Optics 51.12, 1836-1842, (2012)
  • [4] Vaicikauskas, V., Kuprionis, Z., Kaucikas, M., Svedas, V., Kabelka, V. "Mid-infrared all solid state DIAL for remote sensing of hazardous chemical agents." Proc. SPIE 6214, Laser Radar Technology and Applications XI, 62140E (May 19, 2006)
  • [5] Normand, E., McCulloch, M., Duxbury, and Langford, N. "Fast, real-time spectrometer based on a pulsed quantum-cascade laser." Optics Letters, 28.1, 16-18, (2003)
  • [6] Beyer, T., Braun, M., Lambrecht, A. "Fast gas spectroscopy using pulsed quantum cascade lasers." Journal of Applied Physics 93, 3158 (2003)

ORSL/CCNY/CUNY disclaim liability of any kind whatsoever, including, without limitation, liability for quality, performance,
merchantability and fitness for a particular purpose arising out of the use or inability to use the data presented herein.

Program Code Copyright © Thomas Legbandt 2010 - 2018

Locations of visitors to this page