The Reliable Sampling Interval For Monitoring Interannual Variability Of The Kuroshio Transport At 18°N

The interannual variation of Kuroshio transport in the East China Sea directly affects the atmospheric circulation and precipitation in the southeast coastal zone in summer.It is of great academic significance to further study the variation mechanism of Kuroshio transport.Exploring reasonable sampling interval and reliably monitoring the interannual variation of Kuroshio transport is of great importance for observing and studying the multi-scale variation of Kuroshio.Most of the data in current international oceanographic research projects comes from satellite altimeters,model simulations and hydrological observations.However,the satellite altimeter only provides the data of the ocean surface,and the vertical direction of the ocean is short of observation.Due to the hydrology observation would consumption a lot of manpower and material resources,which couldn’t achieve long timescale and high frequency observe and is relatively scarce.In addition,most of the current numerical simulation has various deficiencies,especially the velocity and transport in numerical simulation are often weaker than the actual observation,which underestimates the influence of the interannual variation on the Kuroshio.In order to accurately monitor the interannual variability of Kuroshio transport and to quantify the errors at different sampling intervals,the study explore a reasonable and reliable sampling interval that can provide a reasonable reference for other researchers to make observations,especially hydrological observations in the future.The main work accomplished in this paper is as follows:(1)The surface flow velocity and the mean amount of transport in the vertical direction in the Kuroshio source area are calculated.In this paper,the annual mean of surface velocity and transport in the Kuroshio source area were compared and analyzed by using satellite altimeter and HYCOM outputs from 1998-2021.The annual mean of Kuroshio surface velocity calculated by satellite altimeter is 37.0 cm/s,the annual mean of Kuroshio surface velocity calculated by HYCOM outputs is 42.5 cm/s,and the annual mean of Kuroshio transport is about 24.69 Sv.The results indicate that there is a significant interannual variation of Kuroshio transport and velocity.(2)Reliable sampling interval for monitoring interannual variation of Kuroshio.In this paper,we set the daily data of surface velocity and transport observed by satellite altimeter and HYCOM outputs as the "real" time series.On this basis,we investigate the sampling interval required to monitor the interannual variation of Kuroshio transport reasonably.This paper estimated the deviations between the new series of transport and the "true" time series with different sampling intervals from the 18°N section.The results show that for all practical purposes,a sampling interval significantly shorter than 20 days is required to obtain reliable interannual signals,i.e.,sampling at intervals of ～20 days can resolve the intraseasonal fluctuations in energy that significantly affect the interannual signals.The Root Mean Square Error of the error is less than 50%of the Standard Deviation of the interannual signals.In this paper,we quantify the reasonable sampling interval required to monitor the interannual variability of the Kuroshio,and fit the function between sampling interval and error by polynomial,which can provide a reference for how to select the observation frequency for subsequent hydrological observations.(3)The confirmatory analysis for the reliable sampling interval with 20 days required to monitor the interannual variation of the Kuroshio at 18°N section.First,by analyzing the correlation between the velocity and transport observed from satellite altimeter and HYCOM outputs and the new sequence generated after sampling.The result shows that when the sampling interval is 20 days,the correlation coefficient between them is 0.86,which both pass the significance test.Secondly,the transport and velocity of Kuroshio derived from the observed data and numerical simulation results are used to power spectrum analysis.The results show that the Kuroshio transport at 18°N section has significant energy fluctuations with intraseasonal(50-100 days),seasonal(～180 days),annual(～360 days)and interannual(～3 years)period,and the intraseasonal and seasonal signals are particularly prominent.The long sampling interval would miss these strong high-frequency signals,such as intraseasonal and seasonal signals.Therefore,it is reasonable to guess that these high-frequency signals should be accurately captured if the interannual variation of Kuroshio transport is to be accurately grasped.Finally,in order to verify the rationality of this guess,an ideal experiment is established.The time series with only one signal is obtained by filtering the velocity observed by satellite altimeter with different bandwidths through Butterworth filtering method.Based on the sine function,an ideal sequence with intraseasonal(～80 days),seasonal(～180 days)and interannual(2-7 years)time scales is constructed.The amplitude of time series with different periods is proportional with reference to the time series standard deviation ratio after filtering analysis.The same method is used to analyze the relationship between sampling interval and error of the ideal sequence.The results confirm that the reliable sampling interval is about 20 days,which can better capture the high-frequency signal,and then can reliably monitor the interannual variation of the Kuroshio at 18°N section.This study provides a basis for understanding the uncertainty caused by estimating interannual variability of the Kuroshio with sparse observational data,and provides a reference for the design of hydrological observations such as CTD or Glider transects.Nevertheless,it also shows that previous sample strip observations of one or a few times per year have non-negligible errors in resolving interannual variability.Therefore,integrated observations and analyses are necessary for reliable estimation of interannual variability of Kuroshio in the future.