Qualitative analysis of early warning: A case study from Oman

Abstract

This case study compares early warning in three weather-related emergencies that recently impacted Oman and were triggered by three different hazards: a tropical cyclone, a tropical depression, and a low-pressure trough. The paper examines the relationship between the activation of an early warning system and the underlying group of assumptions in current early warning procedures. In this work, we used data from two sources: (a) real-time Twitter data that were extracted when events were taking place and (b) formal weather forecasts, advisories, alerts, and warnings by formal agencies. First, we conducted an events sequence analysis using Excel in order to trace back warnings and the main events. Secondly, we analysed data qualitatively using NVivo qualitative data analysis software to identify themes in relation to early warning. Findings showed that while the three emergencies were not significantly different from each other in terms of reported casualties and property and infrastructure damages, the warnings against the depression and the trough were delayed and brief compared to the cyclones’ adequate warning. The outcomes were found associated with how weather systems are classified in the case study, primarily based on wind intensity.This article advocates for integrating rainfall, storm surge, and flood data when categorising weather hazards. Additionally, it recommends standardising warning terminology for all types of weather hazards, diversifying expertise within EW centres, utilising multiple methods, and simplifying warning language.