Views
3 years ago

Predicted errors of tropospheric emission spectrometer ... - TES - Nasa

Predicted errors of tropospheric emission spectrometer ... - TES - Nasa

Predicted errors of tropospheric emission spectrometer ... - TES -

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 109, D09308, doi:10.1029/2004JD004522, 2004Predicted errors of tropospheric emission spectrometer nadirretrievals from spectral window selectionJohn Worden, 1 Susan S. Kulawik, 1 Mark W. Shephard, 2 Shepard A. Clough, 2Helen Worden, 1 Kevin Bowman, 1 and Aaron Goldman 3Received 8 January 2004; revised 1 March 2004; accepted 11 March 2004; published 15 May 2004.[1] Error covariances and vertical resolutions are reported for Tropospheric EmissionSpectrometer (TES) nadir-view retrievals of surface temperature, atmospherictemperature, H 2 O, O 3 , CO, and CH 4 . These error covariances are computed as a result ofselecting spectral windows that maximize the information content of simulated, TESnadir-view atmospheric retrievals of four regions representative of northern midlatitude,southern midlatitude, tropical, and polar climates. The information content of a retrieval isa function of an a priori and an a posteriori covariance matrix where the a posterioricovariance depends on an estimated smoothing error, measurement error, and systematicerrors from interfering species, surface emissivity, atmospheric and surface temperature,and line parameter uncertainties. For conditions representative of northern midlatitudes,we can expect about 3 degrees of freedom (DOF) for retrievals of H 2 O, 5 DOF for O 3 withabout 2.4 DOF in the troposphere, and 0.8 DOF for CO. These measures for the verticalresolution and the predicted errors can be used to assess which atmospheric sciencequestions can be addressed with TES atmospheric retrievals. Proper characterization ofTES retrievals is also critical for applications such as atmospheric data assimilation andinverse modeling. INDEX TERMS: 0365 Atmospheric Composition and Structure: Troposphere—composition and chemistry; 0368 Atmospheric Composition and Structure: Troposphere—constituenttransport and chemistry; 0394 Atmospheric Composition and Structure: Instruments and techniques; 1640Global Change: Remote sensing; KEYWORDS: remote sensing, troposphere, error characterizationCitation: Worden, J., S. S. Kulawik, M. W. Shephard, S. A. Clough, H. Worden, K. Bowman, and A. Goldman (2004), Predictederrors of tropospheric emission spectrometer nadir retrievals from spectral window selection, J. Geophys. Res., 109, D09308,doi:10.1029/2004JD004522.1. Introduction[2] The Tropospheric Emission Spectrometer [Beer et al.,2001] is one of four instruments on the EOS-Aura platformdesigned to study the Earth’s ozone, air quality, and climate.The TES is an infrared Fourier transform spectrometer(FTS), which measures the spectral infrared (IR) radiancesbetween 650 cm 1 and 3050 cm 1 in a limb-viewing and anadir (downward looking) mode. The observed IR radianceis imaged onto an array of 16 detectors, which have acombined horizontal footprint of 5.3 km, by 8.4 km in thenadir viewing mode. In the nadir view, TES retrievals willbe sensitive to the more abundant tropospheric species suchas H 2 O, O 3 , CO and CH 4 . However, because verticalinformation about trace gas concentrations is obtained onlyfrom spectral variations along the line of sight, sufficientspectral resolution and signal-to-noise ratio are required to1 Jet Propulsion Laboratory, California Institute of Technology,Pasadena, California, USA.2 Atmospheric and Environmental Research, Inc., Lexington, Massachusetts,USA.3 Department of Physics, University of Denver, Denver, Colorado, USA.Copyright 2004 by the American Geophysical Union.0148-0227/04/2004JD004522$09.00distinguish between stratospheric and tropospheric infraredsignatures. TES spectral resolution was chosen to match theaverage pressure-broadened widths of weak infrared moleculartransitions in the lower troposphere for nadir measurements(0.1 cm 1 apodized) [Beer et al., 2001]. This paperfocuses on the selection of spectral windows to be used forTES nadir retrievals. A future manuscript will describe thespectral window selection for TES limb retrievals.[3] Spectral windows are desired in order to optimizethe quality of a retrieval in part by reducing the effects ofknown systematic errors on the retrieval as well as bychoosing those spectral regions with the best sensitivity tothe atmospheric species of interest. Spectral windows arealso selected to reduce computational burden withoutsignificantly degrading the quality of a retrieval. Errorcharacterization is a necessary part of the spectral windowselection when the computed information content is usedas the metric for window selection [Rodgers, 1998;Dudhia et al., 2002]. For example, von Clarmann andEchle [1998] and Echle et al. [2000] have used informationcontent as a metric for selecting spectral windows,principally for the Michelson Interferometer for PassiveAtmospheric Sounding (MIPAS) instrument. Similarly,Chédin et al. [2003] describes the selection of channelsfor the retrieval of CO 2 profiles by computing the infor-D093081of12

Atmospheric retrievals with the Tropospheric Emission Spectrometer ...
1 First satellite observations of lower tropospheric ... - TES - Nasa
Estimating the summertime tropospheric ozone ... - TES - Nasa
(ACE) measurements of tropospheric chemistry in tr - TES - Nasa
Processing - Mars Global Surveyor Thermal Emission Spectrometer ...
(tes) level 2 algorithm theoretical basis document - Nasa
Validation and Error Modelling of GPS Tropospheric Slant ... - DMI
Extracting Temperature and Emissivity from High Spatial ... - NASA
Calibration of a mid-IR optical emission spectrometer with a 256 ...
Export of emission spectrometers (metal analyzers). Inquiry: the Firm ...
Surface Emissions and Prediction of Atmospheric Composition ...
Bred vectors and forecast errors in the NASA coupled general ...
The prediction error in CLS and PLS: the importance of feature ...
19 TES-GAP - HyspIRI Mission Study Website - NASA
kosek@cbk.waw.pl The prediction errors of the pole coor