Upper Ocean Response And Feedback Mechanisms To Typhoons In The Western North Pacific Ocean

A typhoon refers to a tropical cyclone(TC)with a maximum sustained wind speed greater than 64 kts near the center in the Northwest Pacific Ocean(NWP).Typhoonocean interaction has received increasing attention in recent decades.On the one hand,typhoon can cause notable impact on the upper marine ecological environment along the typhoon track,including changes in ocean temperature and salinity,increases in ocean chlorophyll concentration,etc.The most obvious one might be the cold wake due to ocean surface enthalpy fluxes,vertical mixing,and upwelling.On the other hand,the marine environment,including sea surface temperature(SST),ocean eddies and other factors,are the key factors affecting the generation and development of typhoons.Although the intensity forecast of TCs has been improved over the recent decade,it remains to be a challenging issue relative to the track forecast and has drawn significant research efforts,especially difficult but important problem is to predict the extreme changes in of intensity,including both rapid intensification(RI)and rapid weakening(RW),are account for an important source of forecast errors.Past studies have shown that warm ocean is one of the most important environmental conditions to cause rapid intensification of TCs,and a sharp variation of SST gradient is one of the important factors that cause the typhoon weakened rapidly.Therefore,it is very important to research the typhoon-ocean interaction not only for typhoon intensity forecasting but also for marine environment prediction.Based on multi satellite remote sensing data,Argo floats,JTWC typhoon track data and reanalysis data,we make some special cases analysis and statistical analysis on typhoons in the NWP in recent 40 years.First,we study several typical cases of the upper ocean response to typhoons in the NWP under complex conditions,including the response scenarios of the upper ocean to sequential typhoons and sudden turning typhoons.Firstly,the impacts of binary typhoons Sarika and Haima in 2016 on the SST and SSS of the nearshore and offshore areas along the coast of Guangdong Province and Hainan Island in the northern South China Sea were studied.The results show that the offshore responses of SST and SSS of Guangdong Province and Hainan Island are very similar,but the nearshore responses are very different.Secondly,four sequential typhoons Neoguri,Matmo,Nakri and Halong generated and developed in the NWP in 2014 were studied.The response and feedback mechanisms of the marine environment caused by the first typhoon to subsequent typhoons with different intensities were compared through the modulation mechanism of ocean eddy.The results show that the ocean responses to sequential typhoons depended on not only typhoons intensity,but also typhoons track order and ocean mesoscale eddies.Thirdly,the sudden-turning typhoon Nida in 2009 caused much special ocean responses.The maximum SST cooling of-6.68℃ caused by Nida occurred at the maximum turning point(November 29),but not at the strongest intensity of typhoon(November 26).The cold wake caused by Nida occurs on the left of track,a phenomenon that has not been reported in previous studies.The reasons are that the typhoon winds were cyclonic during the sudden-turning stage,as well as the air-sea interaction time on the left side of typhoon track was longer.It is found for the first time due to the typhoon itself caused left-bias cold wake phenomenon.Then,we study the feedback mechanism of the upper ocean in the NWP to the specific typhoon case.Firstly,the super typhoon Nida in 2009 intensified rapidly along a straight track at a speed of 90kt/24h,with a maximum intensity of 155kts.Afterwards,typhoon Nida slowly experienced a suddenly left turning and caused a strong cold wake of-6.68℃.The enhanced cold wake and longer air-sea interaction resulted in a sharp decrease of ocean heat flux,which caused Nida to weaken rapidly.Within three days,Nida weakened from a Category 5 typhoon to a Category 2 typhoon,and then died out.The study of sudden turning typhoon and upper ocean shows that there is an important relationship between the extreme change of typhoon intensity and typhoon track.Secondly,the evolution of SST cooling during four sequential typhoons in 2014 and its feedback on subsequent typhoons were studied based on multi-platform satellite observations.The study shows that the intensity of subsequent typhoons varies in different ways:when the subsequent typhoon is a weak typhoon Matmo or Nakri,the cold wake caused by previous typhoon inhibits the intensification of the subsequent typhoon.When the subsequent typhoon is a strong typhoon(category 5 typhoon Halong),the cold wake can significantly weaken the typhoon intensity,similar to the results of typhoon Nida.On this basis,we make statistics of typhoons in the northwest Pacific in the past 40 years.The statistical results show that when typhoons turn suddenly,the probability of rapidly intensifying decreases significantly(about 1/6 of rectilinear-track typhoons),while the probability of rapidly weakening increases(about 2 times of rectilinear-track typhoons).The statistical result is closely related to the negative feedback effect of cold wake induced by typhoon itself.When the typhoon moves along a straight track,the cold wake caused by typhoon is weak,and the cold wake lags behind the typhoon center,which cannot induce a significant negative feedback effect on typhoon intensity.When the typhoon turns suddenly,the overlap area between the typhoon inner-core and the cold wake is larger and the interaction time between them is longer,and the ocean transport more energy to typhoon,thus inhibiting(leading)the typhoon strengthening(weakening).This paper analyzes the relationship between the intensity change and typhoon track by studying the negative feedback mechanism of cold wake,which provides a new idea for improving typhoon intensity rapid change prediction skills.Based on the results of statistical research,it is found that the typhoon track information is important factors affecting the extreme change of typhoon intensity.Considering the typhoon track factors(including four new characteristics,typhoon translation speed,24-hour equivalent translation speed,turning angle,absolute value of turning angle)were incorporated into the forecast of extreme change in typhoon intensity,and a prediction model for intensity change of major typhoons was constructed by using typical machine learning methods,decision tree,random forest,support vector machine and multi-layer perceptron.The results show that the prediction effect of the prediction model trained by adding 4 new features has been improved to varying degrees compared with the model trained by traditional features,especially for the rapid weakening and rapid strengthening of strong typhoon cases,the prediction accuracy is improved by 5%-12%.Among different machine learning prediction models,the prediction accuracy/precision of decision tree,random forest,support vector machine and multilayer perceptron are 0.591/0.536,0.691/0.712,0.676/0.649,0.695/0.666,respectively.On the whole,the prediction effect of random forest and multi-layer perceptron is better.The difference is that multi-layer perceptron performs better in predicting typhoon cases with weak intensity change,while random forest performs better in predicting typhoon cases with extreme intensity change.