Research On Ocean Ambient Noise On The Arctic Chukchi Plateau

In the context of rapid changes in Arctic sea ice and environment due to global warming,it is necessary to study the noise under existing ocean ambient conditions in the Arctic sea areas.In this paper,a study on ocean ambient noise properties and its environmental correlation in different sea ice conditions,months and seasons on the Arctic Chukchi Plateau is carried out,which can provide a useful baseline for monitoring future changes in ambient noise levels due to sea ice changes and increased human activities in the region.To improve the detection performance of hydroacoustic equipment,such as sonar,when working in ice zones,this paper utilizesα-stable distribution theory for statistical modeling of under-ice noise,as well as analyzes the statistical characteristics of under-ice noise.The main findings of this paper are as follows:（1）The ambient noise level（ANL）is higher during the ice-free period due to the influence of wind and waves,and reaches its highest level in September.The ANL is lower during the ice-covered period due to the suppression effect of sea ice,and the ambient noise during this period is mainly influenced by the ice-generated noise induced by sea ice activities,and negatively correlated with sea surface temperature.In May and June,the ANL drops to the lowest level of the year as the sea ice activities weaken with the rising temperature in the middle and late spring.The ANL is intermediate during the ice transition period,during which ambient noise is jointly affected by the noise generated by sea ice activities and the interaction between wind and floating ice,as well as wind and sea surface.Sea ice is the most important environmental factor affecting Arctic ocean ambient noise,the ANL is expected to increase in the context of decreasing Arctic sea ice and increasing human activities due to global warming.（2）A new method for statistical modeling of under-ice noise with simple calculation and high estimation accuracy is given by extending the applicability of a parameter estimation method ofα-stable distribution which is used to estimate the parameters of SAR image signals and by solving the problem that the estimation accuracy of the skew parameterβis severely degraded when the characteristic exponentαis close to 2.In this method,two characteristic function points are used for parameter estimation,and the value range of the second characteristic function pointω₀is affected by the scale parameterγ.The results show that whenγ∈[0.00006,0.4],the estimated parameters exhibit better performance whenω₀>1;Whenγ∈(0.4,2],the estimated parameters exhibit better performance when 0<ω₀<0.5.The validation results of the under-ice noise data measured on the Arctic Chukchi Plateau indicate that theα-stable distribution shows a good modeling effect on the under-ice noise whenω₀>1.（3）Influenced by the ice-generated transient noise,the under-ice noise generally exhibits non-Gaussian statistics,and the number of ice-generated transient noise shows a strong negative correlation with the characteristic exponentαof the under-ice noise.Most of the under-ice noise exhibits weak non-Gaussian statistics,and the under-ice noise with full ice coverage has the strongest non-Gaussian statistics compared with the ice forming and melting process.The strength of the non-Gaussian statistics of the under-ice noise varies in different frequency bands.The non-Gaussian statistics of the under-ice noise in the middle frequency band are the strongest,followed by the high frequency band,while the non-Gaussian statistics of the under-ice noise in the low frequency band are the weakest.