总有机碳(TOC)是指溶解和悬浮在水中的有机物的总含碳量, 是以含碳量表示水体中有机物总量的综合指标。 TOC的测定通常采用的是国标燃烧法和湿化学法, 但这两种传统方法都具有测试方法复杂、测量时间长、速度慢等缺点, 对大气环境会产生一定污染且仅能在实验室内完成, 无法进行海水的原位在线测量。 而该研究采用紫外吸收光谱法测量TOC, 通过燕山大学自主研发的光谱技术与集成电路相结合研制的TOC光学原位传感器, 能够快速、不添加试剂、不产生二次污染的测量海水中总有机碳的浓度, 且可以不受实验室环境的制约实现海水TOC的在线原位测量。 将基于该方法研制的传感器进行实地下海实验, 测量不同海域（河北沧州黄骅港周边海域、秦皇岛市周边海域）的TOC值, 将传感器测量的结果与国标法测量结果的相关性、一致性、误差等方面进行对比。 结果表明: 在黄骅港海域采取的13个不同水样和秦皇岛周边海域采取的14个不同水样用国标法和TOC光学原位传感器测量的浓度值变化趋势基本一致, 有较好的相关性和一致性, 存在极个别水样产生偏离整体样本曲线的情况。 实验数据经过线性拟合结果显示相关性较好, 对两处不同海域的数据拟合曲线和常规残差分析, 显示黄骅港海域水样的拟合相关系数r为0.859 0, 残差平方和数值为0.165 4, 秦皇岛周边海域水样的拟合相关系数r为0.939 9, 残差平方和为3.513 1。 由于相关系数r=0.939 9＞r=0.859 0, 所以秦皇岛周边海域的水样线性拟合效果好于黄骅港实验。 常规残差0.165 4<3.513 1, 是因为秦皇岛周边一些样本如河流入海口是污染的重灾区, 生活污水和工业污水使水质存在更多干扰因素, 这些干扰因素对海洋TOC光学原位传感器的准确性和稳定性造成了一定的影响。 由于基于紫外吸收光谱法研制的TOC光学原位传感器与传统的国标法测量原理不同, 样本集比较少, 样本浓度覆盖面不大, 海洋环境复杂多变, 存在多种影响因素, 如浊度、 温度、 pH、 浮游生物等, 传感器无法彻底消除所有影响因素引起的误差, 所以测量结果与国标法结果存在一定的误差。 后续工作如何消除海洋环境复杂干扰因素的影响, 减小传感器测量值的误差, 使测量结果更加精确和真实, 需要进一步讨论和更加详细的分析和研究。

Total Organic Carbon（TOC) refers to the total carbon content of the organic substances dissolved and suspended in water, which is a comprehensive indicator for the total amount of organic substances in water. The traditional measurement technologies of TOC arethe national standard combustion method and the wet chemical method, which always involves complicated test methods, long measurement time, slow speed, some atmospheric environment, and can only be completed in the laboratory and not available to the seawater online in-situ measurement. The concentration of TOC in seawater is measured with ultraviolet absorption spectroscopy technology using the optical in-situ sensor developed by our research team, which achieves online in-situ quick measurement of TOC without adding reagents, doesn’t induce secondary pollution, and is not restricted by the laboratory environment. In this paper, the sensor developed by our research team is used to real underground seawater TOC measurements for different sea areas (the sea area around Huanghua Portand Qinhuangdao City, in Hebei Province), and comparison of correlation, consistent and measurement error between sensor measurement method andnational standard method is conducted, which shows that: The evolution trend of the concentration of TOC from the 13 different seawater samples in the Huanghua Port area and the 14 different seawater samples in the area around Qinhuangdao obtained using TOC optical in-situ sensor are basically consistent with that obtainedfrom the national standard method. Better linear correlation and consistency is demonstrated, and there are very few water samples that deviate from the overall sample curve. The experimental data shows a good correlation through the linear fitting, and the data fitting curves and conventional error are analyzed for two different sea areas. The correlation coefficient of the linear fitting for the water sample of Huanghua Port is r=0.859 0, and the sum of squared residuals is 0.165 4. The correlation coefficient of the linear fitting for the water sample around Qinhuangdao is r=0.939 9, and the sum of squared residuals is 3. 5131. Because the correlation coefficient r=0.939 9>r=0.859 0, which means that the linear fitting effect of water samples around Qinhuangdao is better than that of Huanghua Port experiment. The conventional error is 0.165 4<3.513 1, which is caused bymore serious pollution induced by some samples around river estuaries in Qinhuangdao city. Domestic sewage and industrial wastewater induce more interfering factors in water quality, which cause certain influence to the accuracy and stability of marine TOC optical in-situ sensors; TOC in-situ optical sensors based on ultraviolet absorption spectroscopy technology and traditional methods are different from the national standard method, which uses smaller sample set and sample concentration coverage, however, the marine environment is complicated and changeable, and the sensor cannot completely avoid the influence of all other factors, such as turbidity, temperature, PH, plankton, etc., which is the main source of the measurement error. The future exploration is how to eliminate the influence of complicate interference factors in the marine environment, reduce the error of sensor measurement values, make the measurement results more accurate and true.