Study On The Correction Method Of Infrared Sea Surface Temperature Measurement Based On Atmospheric Transmittance And Sea Surface Emissivity

As the interface between ocean and atmosphere,sea surface is an important place for energy exchange between them.Sea surface temperature plays an important role in the process of air-sea.Obtaining accurate sea surface temperature is significant to the study of air-sea effect and climate change.At present,satellite sensors are mainly used to measure sea surface temperature,but their accuracy is mostly limited to ±0.3?.It's resolution is low,and it is easy to be affected by weather conditions.The shipboard sea surface thermal radiation meter can overcome the above shortcomings.However,due to the unique electrical characteristics of the infrared detector and non-contact measurement,in order to ensure the accuracy of the measurement results,the measurement results need to be corrected.There are mainly two aspects of error correction: the error caused by the instrument and the impact of the marine environment.As for instrument error,under different working temperature,the accuracy of infrared detector will be greatly affected.The temperature stability system is used to ensure that the infrared detector works at the optimum temperature.In order to prevent the equipment from being eroded,the infrared detector needs to be protected.The infrared detector measures the external target temperature through the window lens,so it is necessary to correct the impact of the window lens,analyze the infrared radiation transmission of the window lens,and correct it through experiments.For the marine environment,because it is non-contact measurement,sky radiation,atmospheric thermal radiation,atmospheric transmittance and sea surface emissivity will affect the temperature measurement,among which the most important is the influence of atmospheric transmittance and sea surface emissivity.The influence of these two factors is quantified.Atmospheric transmittance is calculated by empirical model and sea surface emissivity is calculated by Masuda theoretical model.The influence of ocean environment is corrected by the infrared radiation transfer model.The effect of emissivity was verified under laboratory conditions,and the experimental results were in accordance with the theoretical analysis.The data collected by the field offshore experiment are used to verify the error correction method.In most cases,the experimental results can meet the requirements,but in some cases,the corrected error even exceeds ±1?.Using the method of data regression,the back propagation neural network is constructed according to the radiation transmission model,and the error results corrected by BP neural network are all in the range of ±0.3?,which meet the accuracy requirements.