一个基于固有光学特性的透明度半分析算法适用性探讨  

Evaluation of the Applicability of an IOP-based Algorithm to Derive Secchi Depth

作　　者：张居诗 魏国妹[3] 林供[1,2] 商少凌[1,2] 

机构地区：[1]厦门大学海洋与地球学院 [2]厦门大学水声通信与海洋信息技术教育部重点实验室,福建厦门361102 [3]厦门大学海洋观测技术研发中心,福建厦门361005

出　　处：《厦门大学学报（自然科学版）》2014年第4期549-554,共6页Journal of Xiamen University：Natural Science

基　　金：国家科技支撑计划项目(2013BAB04B00)

摘　　要：利用52组中国海实测透明度(Zsd)与遥感反射率(Rrs)数据,检验一个最新的基于固有光学特性的透明度算法(简称IOP算法)的适用性.结果表明,该IOP算法在Zsd实测值<6m的站位,反演Zsd与实测值相对一致,偏差(Bias)为-0.67m,相对误差(ε)为28.6%,对数均方根误差(RMSE)为0.197,仅存在一定程度的低估现象;但在实测值≥6m的站位存在明显的高估,Bias为7.95m,ε为53.7%,RMSE为0.212.进一步的分析发现该IOP算法对颗粒后向散射系数(bbp)和颗粒散射系数(bp)二者间的关系十分敏感,算法中所采用的bbp-bp经验关系,在中国海并不完全适用.Secchi depth （Zsd） is a classic parameter in monitoring water quality, which has been documented for more than 100 years.A lot of efforts have been paid to develop a solid approach to derive the Zsd from space.Recently a novel IOP-based algorithm to derive Zsdwas proposed by Doron et al （2011）.Yet it is not known if it is globally applicable. Here in this study,we used an in situ dataset of Zsd and remote sensing reflectance collected in the China Sea to evaluate its applicability.It was found that the algorithm performed well in turbid waters of Zsd〈6 m （Bias=-0.67 m,average percentage error （ε）：28.6% ,root mean square error at log scale （RMSE）=0. 197）.However,apparent overestimation was found in clear water （Zsd≥6 m）,with Bias of 7.95 m,ε of 53.7% and RMSE of 0. 212.Further analysis indicated that the algorithm was very sensitive to the empirical function of bbp-bp.The current function used in the algorithm was probably valid solely in the turbid water （Zsd〈6 m） ,and thus the algorithm under evaluation was not suggested to be generally applicable to the entire China Sea.

关 键 词：透明度 固有光学特性 算法 评估 

分 类 号：P76[天文地球—海洋科学]