Computer Science – Performance
Scientific paper
Dec 2007
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2007agufm.v41b0607s&link_type=abstract
American Geophysical Union, Fall Meeting 2007, abstract #V41B-0607
Computer Science
Performance
8419 Volcano Monitoring (7280), 8485 Remote Sensing Of Volcanoes
Scientific paper
The difficulty of applying a real-time measured reference spectrum represents the main issue while using automatic Differential Optical Absorption Spectroscopy (DOAS) UV-Scanner networks for monitoring active volcanoes. Here we present the performance of a DOAS retrieval using a modelled reference spectrum derived from a high- resolution solar spectrum. Data analyzed were collected by the five UV scanners installed on Mt. Etna using three calibration cells (LC: low cell 3.2 e17; MC: middle cell 8.46 e17; and HC: high cell 9.98 e17 molecules/cm2) in order to collect calibrated clear-sky spectra (CCSS). We evaluated the errors affecting the CCSS retrievals examining the effects of seasonal variations, time of the day, changes of the telescope-viewing angle, and the modelled and real-measured instrumental line-shape function (ILS). For these purposes, between July 2006 and July 2007, 51 CCSS were recorded in different times of the day and different weather conditions using the LC and the MC, whereas the error associated with the variations of the telescope-viewing angle was evaluated on data collected in May 2007 using the LC and HC. This was estimated as the mean of each of 100 CCSS collected for every scanning angle. The modelled ILS function resolution was found empirically, while the real was measured experimentally using a mercury lamp. The absolute difference retrieved for the CCSS recorded in 12 months respect the true amounts of the calibration cells varied between ~ 1.15 e15 - 8.39 e16 molecules/cm2 for the LC and ~ 2.78 e15 - 1.75 e17 molecules/cm2 for the MC. These results revealed that the modelled reference spectrum did not affect significantly the DOAS performance. This was consistent with the absolute differences estimated for each scanning-angle variations (~ 1.15 e15 - 8.39 e16 molecules/cm2 for the LC and ~ 1.44 e15 - 2.52 e17 molecules/cm2 for the HC) respect to the true amounts. These results prove that UV-Scanner DOAS networks can work efficiently using a modelled reference spectrum. Additionally, they highlight that the modelled ILS could be a source of errors, since it could be not representative of the real ILS of the spectrometer. Errors are mostly dependent on the signal to noise ratio of the spectra and therefore can vary as a factor of the time of the day and viewing angle.
Bruno Natalia
Burton Marcia
Caltabiano Tommaso
Longo V.
Oppenheimer Clive
No associations
LandOfFree
UV Scanner DOAS Data Retrieved Using A Modelled Reference Spectrum does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with UV Scanner DOAS Data Retrieved Using A Modelled Reference Spectrum, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and UV Scanner DOAS Data Retrieved Using A Modelled Reference Spectrum will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1408990