The development of statistical methods for retrieving MODIS missing data: Mean bias, regressions analysis and local variation method

MODIS 손실 자료 복원을 위한 통계적 방법 개발: 평균 편차 방법, 회귀 분석 방법과 지역 변동 방법

  • Received : 2016.12.23
  • Accepted : 2016.12.26
  • Published : 2016.12.31

Abstract

Satellite data for remote sensing technology has limitations, especially with visible range sensor, cloud and/or other environmental factors cause missing data. In this study, using land surface temperature data from the MODerate resolution Imaging Spectro-radiometer(MODIS), we developed retrieving methods for satellite missing data and developed three methods; mean bias, regression analysis and local variation method. These methods used the previous day data as reference data. In order to validate these methods, we selected a specific measurement ratio using artificial missing data from 2014 to 2015. The local variation method showed low accuracy with root mean square error(RMSE) more than 2 K in some cases, and the regression analysis method showed reliable results in most cases with small RMSE values, 1.13 K, approximately. RMSE with the mean bias method was similar to RMSE with the regression analysis method, 1.32 K, approximately.

원격 관측 자료인 위성 자료는 한계점이 있으며, 특히 광학 관측기를 활용하면 구름이나 기타 요인에 의해 손실 자료가 발생한다. 본 연구에서는 MODerate resolution Imaging Spectrometer(MODIS)의 관측 자료 중, 지표면 온도 자료를 대상으로 손실 자료를 복원하기 위한 방법인 평균 편차 방법, 회귀 분석 방법, 지역 변동 방법의 세 가지 복원 방법을 개발하였다. 검증을 위해 2014년과 2015년의 위성 자료에서 관측 비율을 근거로 사례를 선택하였다. 검증 자료에서 확인된 지역 변동 방법의 평균 제곱근 편차(RMSE)는 일부 사례에서 약 2 K 이상으로 다른 복원 방법에 비해 낮은 정확도를 보였으며, 회귀 분석 방법의 RMSE는 평균 약 1.13 K으로 대부분의 사례에서 가장 좋은 결과를 보였다. 평균 편차 방법 사용 시, RMSE는 회귀 분석 방법 시와 유사하게 약 1.32 K으로 나타났다.

Keywords

References

  1. I. Dyras, B. Bizzarri, L. De Leonibus, and P. Struzik, "EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management," Geophysical Research Abstracts, vol. 8, p. 09529, 2006.
  2. Z. Kugler and T. D. Groeve, The Global Flood Detection System, JRC Scientific and Technical Reports, Luxembourg: Off. Official Publ. Eur. Communities, 2007, pp. 4-5.
  3. K. Fukami, T. Sugiura, J. Magome, and T. Kawakami, Integrated Flood Analysis System (IFAS Version 1.2) User's Manual, Technical Note of PWRI, 2009, pp. 7-8.
  4. F. A. Al-Wassai and N. V. Kalyankar, "Major limitations of satellite images," Journal of Global Research in Computer Science, vol. 4, no. 5, pp. 51-59, 2013.
  5. E. B. Brooks, V. A. Thomas, R. H. Wynne, and J. W. Coulston, "Fitting the multitemporal curve: A Fourier series approach to the missing data problem in remote sensing analysis," IEEE Transactions on Geoscience and Remote Sensing, vol. 50, no. 9, pp. 3340-3353, 2012. https://doi.org/10.1109/TGRS.2012.2183137
  6. C. Zeng, H. Shen, M. Zhong, L. Zhang, and P. Wu, "Reconstructing MODIS LST based on multitemporal classification and robust regression," IEEE Geoscience and Remote Sensing Letters, vol. 12, no. 3, pp. 512-516, 2015. https://doi.org/10.1109/LGRS.2014.2348651
  7. H. Shen, X. Li, Q. Cheng, C. Zeng, G. Yang, H. Li, and L. Zhang, "Missing information reconstruction of remote sensing data: A technical review," IEEE Geoscience and Remote Sensing Magazine, vol. 3, no. 3, pp. 61-85, 2015. https://doi.org/10.1109/MGRS.2015.2441912
  8. B. A. Baum, and S. Platnick, Introduction to MODIS cloud products, In Earth science satellite remote sensing, Springer Berlin Heidelberg, 2006, pp. 74-91.
  9. E. Masuoka, A. Fleig, R. E. Wolfe, and F. Patt, "Key characteristics of MODIS data products." IEEE Transactions on Geoscience and Remote Sensing, vol. 36, no. 4, pp. 1313-1323, 1998 https://doi.org/10.1109/36.701081
  10. Z. Wan, MODIS land surface temperature products users' guide, Institute for Computational Earth System Science, University of California, Santa Barbara, CA, 2006, pp 5-15.
  11. L. Boschetti, D. Roy, and A. A. Hoffmann, MODIS Collection 5 Burned Area Product-MCD45, User's Guide, Ver, 2, 2009, pp 1-2.