An neural network model of data mining is used to identify error sources in satellite-derived tropical sea surface temperature (SST) estimates from thermal infrared sensors onboard the Geostationary Operational Environmental Satellite (GOES). By using the Back Propagation Network (BPN) algorithm, it is found that air temperature, relative humidity, and wind speed variation are the major factors causing the errors of GOES SST products in the tropical Pacific. The accuracy of SST estimates is also improved by the model. The root mean square error (RMSE) for the daily SST estimate is reduced from 0.58 K to 0.38 K and mean absolute percentage error (MAPE) is 1.03%. For the hourly mean SST estimate, its RMSE is also reduced from 0.66 K to 0.44 K and the MAPE is 1.3%.
References
[1]
Merchant, CJ; Simpson, JJ; Harris, AR. A cross-calibration of GMS-5 thermal channels against ATSR-2. Remote Sens. Environ 2003, 84, 268–282, doi:10.1016/S0034-4257(02)00112-8.
[2]
McClain, EP; Pichel, WG; Walton, CC; Ahmad, Z; Sutton, J. Multichannel improvements to satellite-derived global sea surface temperatures. Adv. Space Res 1983, 2, 43–47.
[3]
Shoichi, K; FutoKi, S. A new set of MCSST equations for NOAA-9/AVHRR. J. Oceanogr 1996, 52, 235–249, doi:10.1007/BF02235672.
McClain, EP. Global sea surface temperatures and cloud clearing for aerosol optical depth estimates. Int. J. Remote Sens 1989, 10, 763–769, doi:10.1080/01431168908903917.
[6]
May, D; Stowe, L; Hawkins, J; McClain, EP. A correction for Saharan dust effects on satellite sea surface temperature measurements. J. Geophys. Res 1992, 97, 3611–3619, doi:10.1029/91JC02987.
[7]
Emery, WJ; Yu, Y; Wick, GA; Schluessel, P; Reynolds, RW. Correcting infrared satellite estimates of sea surface temperature for atmospheric water vapor contamination. J. Geophys. Res 1994, 99, 5219–5236, doi:10.1029/93JC03215.
[8]
Walton, CC. Nonlinear multichannel algorithms for estimating sea surface temperature with AVHRR satellite data. J. Appl. Meteorol 1988, 27, 115–124, doi:10.1175/1520-0450(1988)027<0115:NMAFES>2.0.CO;2.
[9]
Walton, CC; Pichel, WG; Sapper, JF. The development and operational applications of nonlinear algorithm for the measurement of sea surface temperatures with the NOAA polar-orbiting environmental satellites. J. Geophys. Res 1998, 103, 27999–28012, doi:10.1029/98JC02370.
[10]
Martin, S. An Introduction to Ocean Remote Sensing; Cambridge University Press: Cambridge, UK, 2004.
[11]
Richard, L; Tong, Z. Sea surface temperatures from the GOES-8 geostationary satellite. Bull. Am. Meteorol. Soc 1997, 78, 1971–1983, doi:10.1175/1520-0477(1997)078<1971:SSTFTG>2.0.CO;2.
[12]
Murray, MJ; Allen, MR; Merchant, CJ; Harris, AR; Donlon, CJ. Direct observations of skin-bulk SST variability. Geophys. Res. Lett 2000, 27, 1171–1174, doi:10.1029/1999GL011133.
[13]
Barton, IJ. Interpretation of satellite-derived sea surface temperature. Adv. Space Res 2001, 28, 165–170, doi:10.1016/S0273-1177(01)00337-4.
[14]
Wick, GA; Bates, JJ; Scott, DJ. Satellite and skin-layer effects on the accuracy of sea surface temperature measurements from the GOES satellites. J. Atmos. Ocean. Technol 2002, 19, 1834–1848, doi:10.1175/1520-0426(2002)019<1834:SASLEO>2.0.CO;2.
[15]
Berry, MJA; Linoff, G. Data Mining Technique for Marketing, Sale, and Customer Support; John Wiley & Sons: Hoboken, NJ, USA, 1997; pp. 243–285.
[16]
Frawley, WJ; Piatetsky-Shapiro, G; Matheus, CJ. Knowledge discovery in database: An overview. AI Mag 1992, 13, 57–70.
[17]
Berson, A; Smith, S; Thearling, K. Building Data Mining Applications for CRM; McGraw-Hill: New York, NY, USA, 1999; pp. 45–46.
[18]
Ronald, SS. Accelerating Customer Relationships-Using CRM and Relationship Technologies; Prentice-Hall: Upper Saddle River, NJ, USA, 2001; pp. 51–56.
[19]
Cheng, CT; Ou, CP; Chau, KW. Combining a fuzzy optimal model with a genetic algorithm to solve multiobjective rainfall-runoff model calibration. J. Hydrol 2002, 268, 72–86, doi:10.1016/S0022-1694(02)00122-1.
[20]
Muttil, N; Chau, KW. Neural network and genetic programming for modelling coastal algal blooms. Int. J. Environ. Pollut 2006, 28, 223–238, doi:10.1504/IJEP.2006.011208.
[21]
Xie, JX; Cheng, CT; Chau, KW; Pei, YZ. A hybrid adaptive time-delay neural network model for multi-step-ahead prediction of sunspot activity. Int. J. Environ. Pollut 2006, 28, 364–381, doi:10.1504/IJEP.2006.011217.
[22]
Lin, JY; Cheng, CT; Chau, KW. Using support vector machines for long-term discharge prediction. Hydrol. Sci. J 2006, 51, 599–612, doi:10.1623/hysj.51.4.599.
[23]
Zhang, J; Chau, KW. Multilayer ensemble pruning via novel multi-sub-swarm particle swarm optimization. J. Univers. Comput. Sci 2009, 15, 840–858.
[24]
Chau, KW. Particle swarm optimization training algorithm for ANNs in stage prediction of Shing Mun River. J. Hydrol 2006, 329, 363–367, doi:10.1016/j.jhydrol.2006.02.025.
[25]
Chau, KW; Muttil, N. Data mining and multivariate statistical analysis for ecological system in coastal waters. J. Hydroinform 2007, 9, 305–317, doi:10.2166/hydro.2007.003.
[26]
Elisa, GG; Joan, GS. Prediction of sea surface temperatures in the western Mediterranean Sea by neural networks using satellite observations. Geophys. Res. Lett 2007, 34, L11603, doi:10.1029/2007GL029888..
[27]
McCulloch, WS; Pitts, W. A logical calculus of the ideas imminent in nervous activity. Bull. Math. Biophys 1943, 5, 115–133, doi:10.1007/BF02478259.
[28]
Rumelhart, DE; Hinton, GE; Williams, RJ. Learning internal representations by error propagation. In Parallel Distributed Processing Explorations in the Microstructure of Cognition; Rumelhart, DE, McClelland, JL, Eds.; MIT Press: Cambridge, MA, USA, 1986; Volume 1, pp. 318–362.
[29]
Miller, SW; Emery, WJ. An automated neural network cloud classifier for use over land and ocean surfaces. J. Appl. Meteorol 1997, 36, 1346–1362, doi:10.1175/1520-0450(1997)036<1346:AANNCC>2.0.CO;2.
[30]
Neeraj, A; Rashmi, S; Sujit, BK; Abhijit, S; Vijay, AK. Evaluation of relative performance of QuikSCAT and NCEP re-analysis winds through simulations by an OGCM. Deep-Sea Res. Part I 2007, 54, 1311–1328, doi:10.1016/j.dsr.2007.04.006.
[31]
Wu, J. Effects of atmospheric stability on ocean ripples: A comparison between optical and microwave measurements. J. Geophys. Res 1991, 96, 7265–7269, doi:10.1029/91JC00359.
[32]
Legeckis, R; Zhu, T. Sea surface temperatures from the GOES-8 geostationary satellite. Bull. Am. Meteorol. Soc 1997, 78, 1971–1983, doi:10.1175/1520-0477(1997)078<1971:SSTFTG>2.0.CO;2.
[33]
Chambers, DP; Tapley, BD; Stewart, RH. Measuring heat storage changes in the Equatorial Pacific: A comparison between TOPEX altimetry and Tropical Atmosphere-Ocean buoys. J. Geophys. Res 1998, 103, 18591–18597, doi:10.1029/98JC01683.
[34]
Zeng, X; Zhao, M; Dickinson, RE; He, Y. A multiyear hourly sea surface skin temperature data set derived from the TOGA TAO bulk temperature and wind speed over the tropical Pacific. J. Geophys. Res 1999, 104, 1525–1536, doi:10.1029/1998JC900060.
[35]
Fairall, CW; Bradley, EF; Godfrey, JS; Wick, GA; Edson, JB. Cool-skin and warm-layer effects on sea surface temperature. J. Geophys. Res 1996, 101, 1295–1308, doi:10.1029/95JC03190.
[36]
Wu, X; Menzel, WP; Wade, GS. Estimation of sea surface temperatures using GOES-8/9 radiance measurements. Bull. Am. Meteorol. Soc 1999, 80, 1127–1138, doi:10.1175/1520-0477(1999)080<1127:EOSSTU>2.0.CO;2.
[37]
Haykin, S. Neural Networks-A Comprehensive Foundation, 2nd ed ed.; Pearson Education Asia: Singapore, 2002.
[38]
Fun, MH; Hagan, MT. Levenberg-Marquardt training for modular networks. Proceedings of IEEE International Conference on Neural Networks, Washington, DC, USA, 3–6 June 1996; 1, pp. 468–473.
[39]
Hagan, MT; Menhaj, MB. Training feedforward networks with the Marquardt algorithm. IEEE Trans. Neural Networks 1994, 5, 989–993, doi:10.1109/72.329697.
[40]
Makridakis, S. Accuracy measures: Theoretical and practical concerns. Int. J. Forecasting 1993, 9, 527–529, doi:10.1016/0169-2070(93)90079-3.
[41]
Lewis, CD. Industrial and Business Forecasting Methods; Butterworth Scientific: London, UK, 1982.
[42]
Ueyama, R; Clara, D. A climatology of diurnal and semidiurnal surface wind variations over the tropical Pacific Ocean based on the tropical atmosphere ocean moored buoy array. J. Clim 2008, 21, 593–607, doi:10.1175/JCLI1666.1.
[43]
Deser, C. Daily surface wind variations over the equatorial Pacific Ocean. J. Geophys. Res 1994, 99, 23. 071-23,078.
[44]
May, DA; Osterman, WO. Satellite-derived sea surface temperature: Evaluation of GOES-8 and GOES-9 multispectral imager retrieval accuracy. J. Atmos. Ocean. Technol 1997, 15, 788–797.
[45]
Liu, CT; Nan, CH; Chen, CM. The diurnal variation of sea surface temperature bias of geostationary satellite. J. Photogramm. Remote Sens 2006, 11, 237–247.
[46]
Large, WG; McWilliams, JC; Doney, SC. A review and model with a nonlocal boundary layer parameterization. Rev. Geophys 1994, 32, 363–403, doi:10.1029/94RG01872.
[47]
Bradley, EF; Weller, RA. Joint Workshop of the TOGA COARE Flux and Atmospheric Working Groups; TOGA CORE International Project Office, University Corporation for Atmospheric Research: Boulder, CO, USA, 1995.
[48]
Price, JF; Weller, RA; Pinkle, R. Diurnal cycling: Observation and models of the upper ocean response to diurnal heating,cooling and wind mixing. J. Geophys. Res 1986, 91, 8411–8427, doi:10.1029/JC091iC07p08411.
[49]
Donlon, CJ; Minnett, PJ; Gentemann, C; Nightingale, TJ; Barton, IJ; Ward, B; Murray, MJ. Toward improved validation of satellite sea surface skin temperature measurements for climate research. J. Clim 2002, 15, 353–369, doi:10.1175/1520-0442(2002)015<0353:TIVOSS>2.0.CO;2.
