Tropical Error Covariances for Data Assimilation

Tender description

This project aims to develop underpinning capability in modeling at the global, regional and local scale, and advance the understanding of high-impact weather events in order to support services to improve advice and mitigate the impacts of extreme weather and climate events. The specification of appropriate background error covariances is an essential component of the data assimilation process required to provide accurate analyses to initialize numerical weather prediction (NWP). Whilst significant work has been done in this area for global and mid-latitude applications, understanding of the nature of tropical forecast errors, and their modelling within data assimilation schemes, is a very new area. Numerous recent convective-scale applications of the Met Office’s Unified Model in Southeast Asia, combined with data from a mature global ensemble (MOGREPS-G) provide a wealth of training data to assess the nature of tropical forecast errors. Keywords: research