|
Oceanography/Meteorology
|
Correction of non-uniform aerosol effect of Landsat TM image by using blockwise approached atmospheric correction model (BACM)
The major limitation of BACM is the requirement of DDV target in scene and a priori knowledge of blocksize to perform correction of non-uniform aerosol effect. The errors of retrieved aerosol optical depth and surface reflectance will be increased, if the image covers less dense and dark target, such as in semi-arid area. Blockwise approach will be less satisfactory, if the aerosol distribution on image is highly non-uniform, such as plume. Therefore, more studies should be done to assess the algorithm.
Acknowledgement
This work is supported by both post-doctoral scholarship from Ministry of Education of Taiwan and NSF LTER project of US.
Reference
-
Hill, J., and Sturm, B., 1991, Radiometric correction multitemporal thematic mapper data for use in agricultural land-cover classification and vegetation monitoring, International Journal of Remote Sensing, 12, 1471-1491.
-
Holben, B.N., Vermote, E., Kaufman, Y. J., Tanre, D., and kalb, V., 1992, Aerosol retrieval over land from AVHRR data-application for atmospheric correction. IEEE .Transactions on Geoscience and Remote Sensing, 30, 212-222.
-
Holben, B.N., Eck, T.F., Slutsker, I., Tanre, D., Buis, J.P., Setzer, A., Vermote, E., Reagan, J. A., Kaufman, Y.J., Nakajima, T., and Lavenu, 1997, Multi-band authomatic sun and sky scanning radiometer system for measurement of aerosols. Remote Sensing of Environment, forthcoming.
-
Kaufman, Y.J., 1994, Atmospheric effects on remote sensing of surface reflectance. SPIE Society of Photo-Optical Instrumentation Engineers, 475, 20-33.
-
Kaufman, Y.J., and Tanre, D., 1992, Atmospheric resistant vegetation index (ARVI) for EOS MODIS. IEEE. Transaction on Geoscience and Remote Sensing, 30, 261-270.
-
Kaufman, Y.J., Tanre, D., Remer, L.A., Vermote, E. F., Chu, A., and Holben, B.N., 1997, Operational remote sensing of troposphericaerosol over the land from EOS-MODIS. Accepted by Journal of Geophysical Research-Atmosphere.
-
Liu C.H., 1995, Radiometric correction of SPOT satellite imageries. Ph. D. dissertation, National Central Univerisity., Taiwan, 172pp.
-
Liu, C.H., and Chen, A.J., 1995, An improved spectral knowledge for multi-temporal image classification a case study of urban area. Proceedings of the International Geoscience and Remote Sensing Symposium, Firenze, Italy, 10-14 July 1995 (Piscataway, NJ:IEEE), pp. 1279-1281.
-
Liu, C.H. Chen, A.J., and Liu, G.R., 1996, An image-based retrieval algorithm of aerosol characteristics and surface reflectance for satellite images. International Journal of Remote Sensing, 17, 3477-3500.
-
Moran, M.S., Jackson, R.D., Slater, P.N., and Teillet, P.M., 1992, Evaluation of simplified procedures for retrieval of land surface reflectance factors from satellite sensor output. Remote Sensing of Environement, 41, 169-184.
-
Richter, R., 1996, A spatially adaptive fast atmospheric correction algorithm,. International Journal or Remote Sensing, 17, 1201-1214.
-
Vermote, E., 1995, Atmospheric correction software for Landsat 5 Thematic Mapper data set. NASA/GSFC, Technical Report, 89 p.
-
Vermote, E.F., Tanre, D., Deuze, J. L., Herman, M. and Morcette, J. 1997, Second simulation of the satellite singnal in the solar spectrum, 6S: an overview. IEEE. Transaction on Geoscience and Remote Sensing, 35, 675-686.
Table 1. Comparison of image-derived aerosol optical depths (with/without iterated) and their errorsD t to the field -measurement (AERONET) in TM1 and TM3 bands. High D t in TM1 band from iterated method is due to the abnormally high image-derived n which is also listed. HI, MD are abbreviations of Hog-Island and Madison sites respectively. Mean errors of retrieved t by iterated method are 0.25 and 0.12 in TM1 and TM3 bands as well as 0.14 and 0.05 by non-iterated method, respectively.
|
| Site-Date | Field
| D t |
| |
|
| Measurement |
Iterationa | No Iterationb |
TMSACd
|
| TM1 | TM3 | TM1 | TM3 | u | TM1 | TM3 | TM1 | TM3 |
|
| HI-12jul93 | 0.69 | 0.43 | -0.25 | -0.19 | 4.0 | 0.09 | -0.02 | 0.08 | 0.06 |
| HI-28JUL93 | 0.25 | 0.15 | 0.36 | -0.03 | 7.2 | 0.19 | -0.03 | -0.11 | 0.01 |
| MD-22jun95c | 0.56 | 0.25 | -0.14 | -0.14 | 4.0 | 0.13 | 0.1 | 0.11 | 0.14 |
|
|
a is the iterative procedure by iterating the Junge parameter u with initial u =3.0 (Liu et al. 1996).
b u is set to 3.3 which corresponds to continental model.
c derived aerosol optical depth is the average value of the center four blocks. See table 2 and context.
d r-/r1=2.4, r7/r3=1.35 for pixels selected r7=[0.015, 0.05] (see context).
Table 2. Retrieved aerosol optical depths in TM1 and TM3 bands for every block of MD -22jun95 image which contains non-uniform aerosol effect. Values in parenthesis are (t (TM1), t(tm3)) respectively.
| (0.62,0.29 ) | ( 0.62,0.29) | ( 0.64,0.37) | ( 0.64,0.37) |
| ( 0.64,0.33) | ( 0.64,0.33) | ( 0.66,0.33) | ( 0.64,0.33) |
| ( 0.74,0.45) | ( 0.70,0.37) | ( 0.70,0.37) | ( 0.72,0.37) |
| ( 0.82,0.49) | ( 0.84.0.53) | ( 0.82,0.49) | ( 0.82,0.49) |
|
|
|
|
|
