A Sensitivity Retrospective Method and System for Identifying Observational Sensitive Areas of Rapidly Intensifying Typhoons

Abstract

This application discloses a sensitivity backtracking method and system for identifying observational sensitive areas of rapidly intensifying typhoons, belonging to the field of meteorological forecasting technology. To address the problem of inaccurate sensitive area estimation caused by the strong nonlinear characteristics of rapidly intensifying typhoons in existing technologies, this application proposes the following: First, based on the ensemble transformation sensitivity method, an initial sensitive area is estimated at the verification moment of the rapidly intensifying typhoon. Then, a sensitivity backtracking experiment is conducted, using this initial sensitive area as a new verification area. Starting from the verification moment, the sensitive areas at different historical moments are iteratively calculated at preset time intervals until the initial analysis moment is reached. This application reduces the accumulation of nonlinear errors by decomposing long-term nonlinear estimation into multi-step short-term approximate linear estimation, improving the accuracy of sensitive area identification and providing dynamic guidance for adaptive observation of rapidly intensifying typhoons.

Images