2008 
95 
V ALIDATION OF MODIS AEROSOL OPTICAL THICKNESS PRODUCT DISTRIBUTED 
BY NSMC OVER SEAS AROUND CHINA AND ITS ADJACENT AREA 
L. Sun"’, X. Li a , M. Guo b 
a National Satellite Meteorological Center, China Meteorological Administration, No.46 South Avenue of Zhong-Guan- 
Cun, Beijing 100081, China - (sunling, lixjing)@nsmc.cma.gov.cn 
b National Satellite Ocean Application Service, National Oceanic Administration, No.8 Da-Hui-Si Road, Beijing 
100081, China - gmh@mail.nsoas.gov.cn 
KEY WORDS: Atmosphere, Sea, Matching, Accuracy, Analysis, Satellite, Observation 
ABSTRACT: 
MODIS (Terra) aerosol optical thickness product distributed by NSMC were validated over seas around China and its adjacent area 
using AERONET data from Jan. 2005 to May 2007. Two spatio-temporal matching up schemes were adopted and the results were 
not quite the same. Using 25km radius and ±0.5h as the sampling window, there were 32 match-ups. The linear fitting results 
between MODIS and AERONET were y=0.672075x+0.132479(R 2 =0.68), y=0.707179x+0.112427(R 2 =0.65) and 
y=0.650986x+0.089910(R 2 =0.56) for MODIS 550, 658 and 860nm channels respectively. The RMSE was 0.15, 0.13 and 0.10 
respectively. The percentages of retrievals within the expected uncertainties were 62.5%, 46.9% and 46.9% respectively. Using data 
in spring (Mar. to May, 22 match-ups), the linear fitting results were y=0.804612x+0.050988 (R 2 =0.77), y=0.862697x+0.034432 
(R 2 =0.75) and y=0.851294x+0.012003 (R 2 =0.71) respectively. Using 45km radius and ±1.0h as the sampling window, there were 
120 match-ups. The validation results were relatively poor with large data dispersion and R 2 of 3 channels were all less than 0.3 
while exceeding 0.6 without data from “Taichung” in the linear fitting. The RMSE was 0.23, 0.20 and 0.17 for 550, 658 and 860nm 
respectively with the best result 0.11, 0.09 and 0.07 at Hong Kong PolyU. The percentages of retrievals within the expected 
uncertainties were all less than 50% with the best result 61.5% at “NCU Taiwan”. Generally speaking, for the research area, the 
sampling window of 25km radius and ±0.5h was more suitable than 45km radius and ±1.0h. Although the MODIS AOT products 
over ocean produced at NSMC can assure certain precision, the validation results are not quite satisfying. Besides the influences of 
turbid water reflectance near shore, cloud contamination, aerosol model suitability and representative ability of aerosols over ocean 
by nearby land-based observation, we conclude that calibration of LIB data may be the important error source. 
1. INTRUDUCTION 
With intensive investigations for many years, the National 
Satellite Meteorological Center (NSMC) has constructed the 
MODIS products processing software based on the operational 
framework of acquisition, collection, preprocessing and 
distribution of MODIS data. It can operate in real-time or batch 
processing mode and aerosol product is one of its products. 
The MODIS aerosol retrieval algorithm over ocean adopted by 
NSMC is a multi-wavelength optimization algorithm based on 
look-up tables, coming from the IMAPP software package 
developed by CIMSS Institute of Wisconsin University. The 
product includes aerosol optical thicknesses (AOT) of 7 
channels over ocean (0.47-2.13pm), Angstrom exponent, 
aerosol particle effective radius and ratio of small mode optical 
thickness with a resolution of 10km,. 
From the AOT distribution over China averaged for each 
season from 2005 to 2007 shown in figure 1, it is clear that 
AOT over ocean is maximum in spring, smaller in autumn and 
winter, and minimum in summer over southern waters, but 
influenced by high turbid water, AOT in coastal regions around 
China is overestimated, especially at Subei and Hangzhou Bay 
in winter and spring. 
In this paper, the operational MODIS (Terra) AOT product over 
ocean distributed by NSMC is validated with ground-based 
measurements. 
Figure 1. AOT over China averaged for each season from 2005 
to 2007 
2. VALIDATION METHOD 
2.1 Data Description 
The level 2 sunphotometer product from AERONET (Aerosol 
Robotic NETwork) with cloud clearing and quality insurance 
was adopted as the true value for validation.