Non-global parameter estimation using local ensemble Kalman filtering

TL;DR

This paper introduces a localized ensemble Kalman filtering method to estimate spatially-varying parameters in chaotic models, enabling accurate parameter recovery even with limited observational data.

Contribution

It presents a novel approach for estimating non-global, spatially-varying parameters using localized data within the ensemble Kalman filter framework.

Findings

Accurately estimates spatially and temporally varying parameters.

Effective in recovering parameters representing unmodeled physics.

Demonstrates applicability in low-dimensional atmospheric models.

Abstract

We study parameter estimation for non-global parameters in a low-dimensional chaotic model using the local ensemble transform Kalman filter (LETKF). By modifying existing techniques for using observational data to estimate global parameters, we present a methodology whereby spatially-varying parameters can be estimated using observations only within a localized region of space. Taking a low-dimensional nonlinear chaotic conceptual model for atmospheric dynamics as our numerical testbed, we show that this parameter estimation methodology accurately estimates parameters which vary in both space and time, as well as parameters representing physics absent from the model.