1 [email protected]GOME2 Error Study WP 210: Spectral Aliassing PM2 10/2001 Spectral Aliassing I. “Bad Case Scenario” R. De Beek, M. Weber, R. Siddans (RAL), B. Latter (RAL) all albedo sequences were analysed to identify worst case scenario from LANDSAT/artificial albedo perturbation provided by RAL worst case scenario was defined by maximum variance of differential albedo error spectra definition of differential albedo error spectra: I() – P() = W a ()(a()/a) – P() I() relative intensity perturbation due to albedo variation P() fitted quadratic (cubic) polynomial W a () albedo weighting function (SCIATRAN) a() albedo sequence with mean albedo subtracted a mean albedo of sequence
9
Embed
Spectral Aliassing I. “Bad Case Scenario” R. De Beek, M. Weber, R. Siddans (RAL), B. Latter (RAL)
Spectral Aliassing I. “Bad Case Scenario” R. De Beek, M. Weber, R. Siddans (RAL), B. Latter (RAL) all albedo sequences were analysed to identify worst case scenario from LANDSAT/artificial albedo perturbation provided by RAL - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
R. De Beek, M. Weber, R. Siddans (RAL), B. Latter (RAL)
all albedo sequences were analysed to identify worst case scenario from LANDSAT/artificial albedo perturbation provided by RAL
worst case scenario was defined by maximum variance of differential albedo error spectra
definition of differential albedo error spectra:
I() – P() = Wa()(a()/a) – P()
I() relative intensity perturbation due to albedo variation P() fitted quadratic (cubic) polynomial Wa() albedo weighting function (SCIATRAN) a() albedo sequence with mean albedo subtracted a mean albedo of sequence
Preliminary Conclusion for IT=0.1875ms and readout of 46ms:
errors on the order of less than 0.1% (O3), 80% (NO2), 11% (BrO), 5% (ozone hole OClO) have been observed for albedo variations of less than 0.04 (very small albedo variations!)
spatial aliassing equivalent noise is a dominant error source for current IT and readout time settings
larger albedo variations leads to variability of reference slant column and air mass factor (additional error source). DOAS assumes unique slant column within the window
fits will be repeated for intensities with proper mean albedo (a=0.46 and a=0.5)
additional case studies and error checking on current approach is planned.