運用資料探勘技術分析熱帶海水表面溫度

Analysis of Tropical Sea Surface Temperature Using Data Mining Technique

10.6574/JPRS.2013.17(2).4

李永翔(Yung-Hsiang Lee)；郭南榮(Nan-Jung Kuo)

資料探勘 ； 倒傳遞類神經網路 ； 紅外線感測器 ； 海面溫度 ； 熱帶太平洋 ； Data mining ； Back Propagation Network ； Infrared sensor ； Sea surface temperature ； Tropical Pacific

航測及遙測學刊

17卷2期（2013 / 08 / 01）

135 - 148

繁體中文

本研究應用資料探勘技術提升地球同步作業環境衛星（Geostationary Operational Environmental Satellite, GOES）的紅外線感測器所量測導出的熱帶海面溫度資料的準確度，並探討影響誤差的主要因素。由倒傳遞類神經網路（Back Propagation Network, BPN）的演算，得到日平均的海面溫度均方根誤差（Root Mean Square Error, RMSE）從0.58 K降至0.37 K，平均絕對百分比誤差（Mean Absolute Percentage Error, MAPE）為1.03％；小時的海面溫度均方根誤差從0.66 K降至0.44 K，MAPE值為1.1％，顯示倒傳遞類神經網路演算法有效改善了海面溫度估計的準確度。對於不同比例的雲層遮蔽情況下，倒傳遞類神經網路對於衛星海面溫度資料修正後之RMSE均維持在0.38 K以下，展現倒傳遞類神經演算法對於海面溫度分析時之抗雜訊能力。另外，分析結果也顯示大氣溫度是影響誤差的主要因素，其次為風速與相對溼度。

Tropical sea surface temperature (SST) data derived from the Geostationary Operational Environmental Satellite (GOES) is analyzed by using data mining to explore the error sources of data and to further improve its accuracy. The SST data has been pre-processed into two kinds of data set, the daily mean and hourly. The root mean square error (RMSE) of daily SST estimate is reduced from 0.58 K to 0.37 K and the mean absolute percentage error (MAPE) is 1.03% by using the Back Propagation Network (BPN) algorithm. For the hourly SST estimate, its RMSE is also reduced from 0.66 K to 0.44 K and the MAPE is 1.1%. This indicates that the BPN algorithms improve the accuracy of the SST. While the proportion of cloud contamination is in different circumstances, the RMSE of retrieval satellite SST by using the BPN algorithm can be maintained below 0.38 K. This demonstrated the efficiency ability of anti-noise analysis of the neural algorithm. The factor analysis also shows that the errors are mainly caused by air temperature and then followed by wind speed and relative humidity.