Study On The Characteristics Of Hurricane Boundary Layer Height Based On Dropsonde Dat

In this paper,the thermal dynamic structure and height distribution characteristics of the hurricane boundary layer（HBL）were analyzed with the composite analysis method based on the 2674 dropsonde data of GPS dropped by the National Oceanic and Atmospheric Administration（NOAA）during 61 hurricanes from 2002 to 2020.The applicability of the bulk Richardson number（Ri_b）method to determine HBLH and its sensitivity to parameter selection were compared and analyzed.Results showed that the temperature and humidity distribution of HBL presented a typical"warm-wet core"structure,while the wind field had a significant asymmetric structure,i.e.,the wind speed was higher on the right side but lower on the left side in relation to the hurricane motion direction.The HBLH determined by theRi_bmethod depends on the selection of the critical value(Ri_bc)ofRi_b.As theRi_bc increases from0.25 to 0.5,HBLH increases systematically.Moreover,theRi_b method is more reasonable to calculate HBLH when the sea surface friction effect is considered.The HBLH determined by theRi_b method was the lowest in the eye region（0-0.5 RMW）,the highest in the gale region（0.5-1.5 RMW）.The HBLH decreases with the increase of the distance from the hurricane center outside the eye region,and presents an asymmetric structure similar to the wind field,which is‘low on the left and high on the right’.The ERA5 PBLH product generally overestimated the observed HBLH,especially for the lower HBLH in the eye area of the hurricane,but significantly underestimated the higher HBLH in the gale area of the hurricane.The HBLH revealed a slight regional difference on the whole.The HBLH deviation of ERA5may be due to its deviation of the wind field in the lower layer of the hurricane.In addition,ERA5’s underestimation of HBLH is also related to its underestimation of the stability of the lower troposphere（LTS）.The linear regression analysis results show that both HBLH values determined by theRi_b method and the PBLH products of ERA5 are highly correlated with the dynamic factors（e.g.,low-level wind field）of HBL,but have no significant relationship with the thermal factors（e.g.,LTS）of HBL.Compared with the Ri_b method,HBLH objective judgment methods（e.g.,maximum wind speed method,potential temperature gradient method and inflow layer height method）have certain limitations in their application scope because they depend on definite profile characteristics.Although there are some differences in the distribution characteristics of HBLH determined by different methods relative to the hurricane center,the HBLH determined by the above objective methods are consistently lower than that determined by theRi_b method,indicating that theRi_bc=0.25 or 0.5 commonly used by theRi_b method may be too high under the conditions of sea-surface hurricane.This in turn leads to systematic overestimation.In this paper,the optimumRi_bc=0.15 and 0.13 are determined by manually identified HBLH as the true value and determined the optimal Ri_bc under the sea surface hurricane conditions with the minimum error method.Although the results need to be further verified,they can provide a reference for the determination of the atmospheric boundary layer height under the condition of sea surface hurricane.