I Workshop on Lidar Inversion Algorithms

Location

UNIVERSIDAD DE CONCEPCIÓN, CENTRO DE ÓPTICA Y FOTÓNICA
Programa de Financiamiento Basal para Centros Científicos y Tecnológicos de Excelencia -CONICYT
March, 10 to 13, 2014

• Program Workshop Lidar Inversion March 2014 - (PDF or HTML)

• REPORT of activities during this workshop - (PDF)

Participants

• Antonieta Silva, CEFOP - Chile

• Daniel Nisperuza, UNAL - Colombia

• Fabio Lopes, IPEN - Brasil

• Henrique Barbosa, USP - Brasil

• Pablo Ristori, CEILAP - Argentina

Data files

Our friends from Earlinet have provided datasets for us to try our algorithms.

• Profiles used for the first Earlinet paper on the elastic retrievals (Bockmann et al, App. Opt. 2004)

  • Signals for 355, 532 and 1064 nm and input P, T download

  • P and T with better format download

    • pressure = hPa

    • temperature e dew point = degC

    • lidar ratio = sr

    • altitude = m

  • Noise and background added to the input signal as Signal -> Poissrnd(Signal + BG). Columns are: alt, 355, 532 and 1064. This is wrong because the original signal is very small at the end and the Poisson() will result in single values.

    • Bg=1e6, 1e4, 1e2, and 1e0 -- download holger-bg.zip

  • Noise and background added to the input signal as Signal -> Poissrnd(1000*(Signal + BG)). Columns are: alt, 355, 532 and 1064.

    • Bg=1e8, 1e7, 1e6, 1e5, 1e4, 1e3, 1e2, 1e1, and 1e0 -- download holger-poisson.zip

  • Expected output 355nm, 532nm, and 1064nm

• Profiles used for the raman Earlinet paper (Pappalardo et al, 2004)

  • Signals for 355, 387, 532, 608, 1064 download input and output

Pablo Ristori (Ceilap - Argetina) also provided a simulated dataset. This is more complicated as it includes clouds and aerosols. Input is only available for 355nm at the moment. Two versions are available:

• Strong cloud - download input

  • Input is: 1- altitude and 2 - signal at 355nm

  • Cloud at 6km LR=28 and beta ~= 280 Mm^-1 sr^-1 (Cloud optical depth = 1.0)

  • BL Aerosols up to 1.5km, LR=28, beta = 5 Mm^-1 sr^-1

  • Residual aerosols, LR=28, beta = 0.5 Mm^-1 sr^-1

• Weak cloud - download input and output

  • Input is: 1- altitude and 2 - signal at 355nm

  • Cloud at 6km LR=28 and beta ~= 57 Mm^-1 sr^-1 (Cloud optical depth = 0.2)

  • BL Aerosols up to 1.5km, LR=28, beta = 5 Mm^-1 sr^-1

  • No residual aerosols

  • Alternative input with extra noise and background

    • Bg=1e6, 1e4, 1e2, 1e0 -- download ristori-bg.zip

Data format

The output of your algorithms should produce simple ascii files without header and columns separated by TAB with the following columns:

1. height = m

2. backscatter = 1/Mm / sr

3. extinction = 1/Mm

4. lidar ratio = sr

5. molecular backscatter = 1/Mm / sr (step 3)

6. molecular extinction = 1/Mm (step 3)

7. synthetic molecular signal = 1/m / sr /m2 (step 3)

References

• Böckmann, C., Wandinger, U., Ansmann, A., Bösenberg, J., Amiridis, V., Boselli, A., … Wiegner, M. (2004). Aerosol Lidar Intercomparison in the Framework of the EARLINET Project. 2. Aerosol Backscatter Algorithms. Applied Optics, 43(4), 977. doi:10.1364/AO.43.000977

• Pappalardo, G., Amodeo, A., Pandolfi, M., Wandinger, U., Ansmann, A., Bösenberg, J., … Wang, X. (2004). Aerosol Lidar Intercomparison in the Framework of the EARLINET Project. 3. Raman Lidar Algorithm for Aerosol Extinction, Backscatter, and Lidar Ratio. Applied Optics, 43(28), 5370. doi:10.1364/AO.43.005370