Publications

Detailed Information

Retrieving Aerosol Optical Thickness and Altitude of Asian Dust from AIRS Measurements

DC Field Value Language
dc.contributor.advisor손병주-
dc.contributor.author한효진-
dc.date.accessioned2017-07-14T00:33:58Z-
dc.date.available2017-07-14T00:33:58Z-
dc.date.issued2013-02-
dc.identifier.other000000009737-
dc.identifier.urihttps://hdl.handle.net/10371/121184-
dc.description학위논문 (박사)-- 서울대학교 대학원 : 지구환경과학부, 2013. 2. 손병주.-
dc.description.abstractAerosol optical thickness (AOT) and effective height (zdust) of Asian dust were retrieved from infrared (IR) hyperspectral measurements. In order to examine how accurately dust properties can be retrieved from satellite-borne IR hyperspectral measurements, we first employed a statistical artificial neural network (ANN) method to retrieve the pixel-level AOT and zdust for Asian dust. This was done by relating the Atmospheric Infrared Sounder (AIRS) brightness temperatures for 234 channels, surface elevation, and relative air mass to target parameters (here, AOT and zdust) through an ANN model. Target data collocated with AIRS brightness temperature (TBs) were from the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived AOT and the Cloud Aerosol Lidar Infrared Pathfinder Satellite Observation (CALIPSO)-derived zdust. Results show a correlation coefficient of 0.84 and mean bias of 0.03 for AOT, and a correlation coefficient of 0.79 and mean bias of -0.01 km for zdust, suggesting that dust retrievals from IR hyperspectral measurements are comparable to MODIS-derived AOTs and CALIPSO-estimated dust heights.
The potential applicability of IR hyperspectral measurements to dust monitoring, demonstrated and substantiated by the ANN results, brings in a possibility of developing a physical retrieval algorithm for Asian dust. In doing so, the fast Radiative Transfer for Television Infrared Observation Satellite (TIROS) Operational Vertical Sounder (RTTOV) model (version 9.3) was implemented with an updated dust-size distribution and surface emissivity for better simulating the dust effect on top of atmosphere radiances. Also included were the refractive indices giving better Asian-dust simulation results, in comparison to AIRS-measured TBs. Adopting a one-dimensional variational (1DVAR) approach embedded with the implemented RTTOV model, a physical model has been developed for retrievals of Asian dust AOT and effective height. Results show that AIRS-derived IR AOTs are well correlated with the MODIS visible (VIS) AOTs, with correlation coefficients of 0.77–0.96. Effective dust heights also agree well with dust heights derived from CALIPSO backscattered reflectivity profiles. In conclusion, AOT and the effective height of Asian dust can be retrieved from IR hyperspectral measurements with accuracies comparable to MODIS AOT and CALIPSO dust height. Since those parameters can be retrieved on a two-dimensional basis in both day and night, IR hyperspectral sounder measurements can significantly improve dust forecasting through dust monitoring and data assimilation.
-
dc.description.tableofcontents1. Introduction 1
2. Review of previous IR-based dust retrievals 5
3. IR hyperspectral measurements 16
4. Examining dust signal in the AIRS hyperspectral measurements 24
4.1 Sensitivity of IR TB to dust amount and height 24
4.2 Construction of training data sets for ANN model 28
4.2.1 Detecting FOVS containing dust 32
4.2.2 Construction of collocated AIRS vs. MODIS data 35
4.2.3 Construction of collocated CALIPSO vs. AIRS data 36
4.3 ANN model development 39
4.4 Validation 46
4.5 Case studies 49
4.5.1 March 1–2, 2008 case 50
4.5.2 April 17, 2009 case 55
4.5.3 April 23, 2009 case 59
5. Simulation of dust impact on AIRS spectrum 63
5.1 Dust modeling for radiative transfer calculation 64
5.2. Radiative transfer modeling 65
5.2.1 Radiative transfer model 66
5.2.2 Dust modeling and improved Lorentz-Mie calculation 67
5.2.3 Surface emissivity 75
5.3 AIRS spectral radiance simulations 76
5.4. Comparisons with observations 77
6. Dust retrieval using a physical method 90
6.1 Retrieval algorithm 90
6.2 Channel selection 92
6.3 Meteorological model 95
6.4 Results 97
6.4.1 February 29, 2008 case 97
6.4.2 March 1, 2008 case 99
6.4.3 May 11, 2011 case 102
6.4.4 April 23, 2009 case 104
6.4.5 April 24, 2009 case 107
6.4.6 Evolutionary feature of April 23–24, 2009 dust event 112
6.4.7 April 23 and 24, 2009 cases with initial guess fields from ANN retrievals 116
6.4.8 Statistical summary 124
7. Summary and conclusions 126
References 132
국문초록 145
-
dc.formatapplication/pdf-
dc.format.extent9457126 bytes-
dc.format.mediumapplication/pdf-
dc.language.isoen-
dc.publisher서울대학교 대학원-
dc.subject.ddc550-
dc.titleRetrieving Aerosol Optical Thickness and Altitude of Asian Dust from AIRS Measurements-
dc.typeThesis-
dc.description.degreeDoctor-
dc.citation.pages147-
dc.contributor.affiliation자연과학대학 지구환경과학부-
dc.date.awarded2013-02-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share