|
|
- 2016
利用MODIS温度产品进行秩修正滤波FRF时空插值
|
Abstract:
普通克里金方法具有空间结构探索及插值分析（给出插值结果及其精度）的功能。但是，绝大部分的克里金方法主要用于空间插值，在时空插值方面的研究还较少。综合考虑具体的实验区域和观测数据构建基函数，使用秩修正滤波（fixed rank filtering， FRF）方法对MODIS平流层温度数据进行时空插值预测并将其结果与秩修正克里金（Fixed rank Kriging， FRK）方法的插值结果进行对比分析。实验结果表明，在空间数据（空间点）整体分布均匀且有已知点的情况下，FRK方法预测的数据精度更高，略优于FRF；而对于较大空间范围内缺失数据的情况，考虑温度在时间维上具有一定的相关性，FRF方法在缺失空间信息时能够引入更多时空信息从而获得较其他方法更高质量的插值结果
| [1] | Cressie N A C. Statistics for Spatial Data[M]. US: John Wileg & Sons. Inc, 1993 |
| [2] | Kammann E E, Wand M P. Geoadditive Models[J]. <em>Journal of the Royal Statistical Society: Series C (Applied Statistics)</em>, 2003, 52(1): 1-18 |
| [3] | Banerjee S, Gelfand A E, Finley A O, et al. Gaussian Predictive Process Models for Large Spatial Data Sets [J]. <em>Journal of the Royal Statistical Society: Series B (Statistical Methodology)</em>, 2008, 70(4): 825-848 |
| [4] | Cressie N, Johannesson G. Spatial Prediction for Massive Datasets[C].Australian Academy of Science Elizabeth and Frederick White Conference, Australia, 2006 |
| [5] | Kang E L, Cressie N, Shi T. Using Temporal Variability to Improve Spatial Mapping with Application to Satellite Data [J]. <em>Canadian Journal of Statistics</em>, 2010, 38(2): 271-289 |
| [6] | Antoulas A C. Approximation of Large-scale Dynamical Systems [M]. Philadelphiau: Society for Industral and Applied Mathematics, 2005 |
| [7] | Shi T,Cressie N. Global Statistical Analysis of MISR Aerosol Data: A Massive Data Product from NASA's Terra Satellite[J]. <em>Environmetrics</em>, 2007, 18(7): 665-680 |
| [8] | Cressie N, Kang E L. High-resolution Digital Soil Mapping: Kriging for very Large Datasets[M]. Netherlands: Springer, 2010: 49-63 |
| [9] | Katzfuss M, Cressie N. Tutorial on Fixed Rank Kriging (FRK) of CO2 Data[D]. Columbus: The Ohio State University, 2011 |
| [10] | Henderson H V, Searle S R. On Deriving the Inverse of a Sum of Matrices[J]. <em>Siam Review</em>, 1981, 23(1): 53-60 |
| [11] | Nychka D, Haaland P D, O'Connell M A, et al. Appendix A: FUNFITS, Data Analysis and Statistical Tools for Estimating Functions[J]. <em>Case Studies in Environmental Statistics</em>, 1998, 132: 159-179 |
| [12] | Nychka D, Wikle C, Royle J A. Multi-resolution Models for Nonstationary Spatial Covariance Functions[J]. <em>Statistical Modelling</em>, 2002, 2(4): 315-331 |
| [13] | Cressie N, Johannesson G. Fixed Rank Kriging for very Large Spatial Data Sets [J]. <em>Journal of the Royal Statistical Society: Series B (Statistical Methodology)</em>, 2008, 70(1): 209-226 |
| [14] | Wikle C K. Low-rank Representations for Spatial Processes[J]. <em>Handbook of Spatial Statistics</em>, 2010: 107-118 |
| [15] | Chen Yuejuan, Bi Yun, Ling Jian, et al. The Temperature of the Stratosphere Variation Trendency and Its Relationship with Tropospheric Temperature[C]. Chinese Meteorological Society Annual Conference, Beijing, 2004(陈月娟,毕云,凌健,等.平流层的温度变化趋势及其与对流层温度变化的关系[C].中国气象学会2004年年会,北京,2004) |
| [16] | Nguyen H,Cressie N, Braverman A. Spatial Statistical Data Fusion for Remote Sensing Applications[J]. <em>Journal of the American Statistical Association</em>, 2012, 107(499): 1 004-1 018 |
| [17] | Cressie N, Shi T, Kang E L. Fixed Rank Filtering for Spatio-temporal Data [J]. <em>Journal of Computational and Graphical Statistics</em>, 2010, 19(3): 724-745 |
