Surface air pressure–based reconstruction of tropical cyclones affecting Hong Kong since the late nineteenth century

Abstract

By using instrumental surface air pressure (SAP) records for 1885–2017 at Hong Kong (HK) station, SAP data for 1951–2016 at 66 stations over mainland China and modern tropical cyclones (TCs) of 1951–2017 derived from the Best Tracks (BT) dataset over Northwest Pacific, an objective identification method (OIM) for TCs is developed. Taking HK station as an example, the general distance of detectable landing TCs and the SAP thresholds are determined by utilizing the correlation between the SAP metrics at HK station and those at each landing site for all 392 real modern TC processes during the period of 1951–2016. Then, a long series of TCs affecting HK station for 1885–2017 is reconstructed by applying the thresholds of daily mean SAP and 24-h SAP difference to observed SAP data during the whole stage. The misjudgment of this OIM is about 10%, and it provides a homogeneous series of TCs affecting HK area during 1885–2017. The reconstructed annual TC series shows a visible decreasing trend from 1885 to 2017, with a more obvious reduction occurring after the early 1960s. Specifically, the 10 years of 1994–2003 had the smallest number of TCs. This SAP-based method developed in this study is potentially applicable for other areas.