| 摘要: |
| 根据1995年9月中日合作南黄海现场观测资料，对南黄海生物-光学算法进行研究。首先对南黄海现场多光谱反射比数据进行因子分析，解反射比协方差矩阵的特征问题；解出特征值和对应的特征向量，求出主因子，然后建立叶绿素a浓度对主因子的多元线性回归方程，并对算法进行回报和预报检验。计算结果表明，该算法具有较高精度，同时也显示出天气条件对估算精度产生较大影响。 |
| 关键词: 因子分析 多光谱反射比 叶绿素a浓度 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金资助项目，495.46004号;东京水产大学创立百周年纪念学术研究奖励基金 |
|
| A STUDY ON MARINE BIO-OPTICAL ALGORITHM — ESTIMATION OF CHLOROPHYLL CONCENTRATION IN THE SOUTH YELLOW SEA |
|
Fu Kecun1, Zeng Xianmo1, Li Baohua1, Zeng Xiaoqi2, Hisayuki Arakawa3
|
|
1.First Institute of Oceanography.SOA. Qingdao 266003;2.Department of Fisheries, Ocean Uniuersity of Qingdao, Qingdao 266003;3.Tokyo University of Fisheries, Tokyo l08
|
| Abstract: |
| On the basis of the multi-spectral reflectance data in the South Yellow Sea and the chlorophyll a concentration data determined in situ with a fluorometer during the “Sino-Japanese Joint Basic Research on visible light remote sensing in the Yellow Sea” in September, 1995, a marine bio-optical algorithm has been studied. Factor analysis has been undertaken for the multi-spectral reflectance data. The reflectance data covariant matrix is solved for eigenvalues and their corresponding eigenvectors; the principal factors are obtained. The contributions of the first three principal factors to the variance of reflectance are, respectively, 66.37%, 31.8% and 0.97%. From the first three principal factors 99.19% of the total information on the original variables can be retrieved. A multivariate linear regression equation has been established between the natural logarithm of the chlorophyll a concentration and the first three principal factors. The result shows that the overall correlation coefficient of the subsample reaches 0.9713 and the root-mean-square error of the hindcasting test estimate of chlorophyll a concentration is 0.0262 mg/m3 (the range of the chlorophyll a: 0.130 - 0.840 mg/m3), with a relative error of 10.06%. The relative error for the forecasting test is 17.79%. Predicted values of chlorophyll a concentration are greatly affected in cloudy and rainy days. |
| Key words: Factor analysis, Multi-spectral reflectance, Chlorophyll a concentration |
