Introduction

This case study examines a major Saharan dust outbreak that crossed the tropical Atlantic and affected Barbados during August 2023. The event formed part of a strong Saharan Air Layer (SAL) episode and was investigated in detail by Rushley et al. (2026), who analysed its impact on the atmospheric environment over Barbados using observations, reanalyses and numerical modelling. The study identified a pronounced dust intrusion during 14–16 August 2023, associated with elevated aerosol optical depth (AOD), a deep Saharan Air Layer extending several kilometres above the surface, and substantial changes in temperature, humidity and wind profiles.

Here, the event is reproduced using OpenIFS 48r1 with the CAMS aerosol and atmospheric composition capability. A set of OpenIFS experiments is compared against CAMS operational analyses, which provide a useful reference representation of the evolving dust plume. The objective is to evaluate how well OpenIFS reproduces the timing, transport and magnitude of the trans-Atlantic dust outbreak before moving on to more detailed analysis of its vertical structure and meteorological impacts.

Literature

This table lists published references which directly describe relevant atmospheric processes for this case study. They can be used for background reading and provide further inspriration for the future expansion of this case study.

Experiment Description

Scientific Motivation

Rushley et al. (2026) examined an unusually strong Saharan Air Layer (SAL) event that reached Barbados during 14–16 August 2023. The study identified:

• 6–8 August as a relatively dust-free period.

• 14–16 August as the principal SAL event.

• A large trans-Atlantic dust plume approaching Barbados on 14 August.

• Peak AERONET aerosol optical depths approaching 0.8.

• Elevated dust layers extending approximately 2–6 km above the surface.

The objective in our worked example is not to reproduce all observed features but to establish in a first basic analysis whether OpenIFS simulations reproduce:

• the existence of the plume,

• its arrival time,

• its magnitude,

• and its westward propagation across the Atlantic.

The described experiments represent an OpenIFS–CAMS consistency study rather than an independent observational validation of the model results.

Methodology

Our worked example requires the following model outputs which are pre-included among the selected outputs in the supplied model namelist:

• Dust Aerosol Optical Depth (DU AOD550)

• Total Aerosol Optical Depth (AOD550)

The analysis focuses on generating global AOD maps to visualise the plume, AOD time series with focus on the Barbados region, and Atlantic longitude-time propagation diagrams. 

The worked example shown here can be completed entirely using OpenIFS and CAMS output. The user is free to build on this first analysis by expanding into a wider range of model outputs and diagnostics and/or by comparing against observational data as described in the relevant published literature (we do not include this here). 

All OpenIFS model experiments are initialised from CAMS analysis data and use a low-resolution horizontal grid to minimise computational requirements for the user.

We will consider the following simulations:

• CAMS Reference: We use 6-hourly CAMS analysis data from August 2023. 
• OpenIFS experiment 1: OpenIFS simulation nudged to ERA5-reanalysis data, initialised on 5 August until end date 24 August 2023. 
• OpenIFS experiments 2–6:  A lagged-start matrix of five free-running OpenIFS hindcasts, initialised every 24 hours between 10 August and 14 August at 00 UTC. All simulations run until at least 24 August. 

What does the Case Study data pack contain?

• We provide experiment data packs that allow the simulation of a time period within the described event by conducting OpenIFS forecast (hindcast) experiments with interactive atmospheric composition.

• The pack includes the global initial experiment data and boundary conditions for each OpenIFS experiment, CAMS analysis data, and relaxation files for nudged runs.

• The OpenIFS experiments use the N80 horizontal model grid (T159), which has approximately 125km / 1.125 degree grid spacing, on 137 vertical model levels.

• The case study includes both free-running model experiments, and one experiment with synoptic meteorology constrained by nudging towards six-hourly ECMWF ERA5 reanalyses.

• We provide plotting scripts using the ECMWF earthkit Python package to generate plots from the OpenIFS model outputs.

Available Data and Simulations

Worked Solution

A complete description of how to set up and run the case study experiments, including the further processing and example plotting of the OpenIFS model results can be found in the README.md file in the case study download directory.

After the OpenIFS model experiments have been completed, the provided Python scripts can be used as examples how to visualise both CAMS analysis data and OpenIFS model output. These scripts use the Earthkit software package to extract aerosol optical depth fields, compute regional averages, and generate diagnostic figures. The examples focus on dust aerosol optical depth (DU AOD) and total aerosol optical depth (AOD) at 550 nm.

Saharan Dust Plume Progression

The animation below shows the temporal evolution of Dust Aerosol Optical Depth (AOD) at 550 nm between 05 August and 24 August 2023 projected onto a global map. The westward transport of the Saharan dust plume across the Atlantic is clearly visible. The CAMS analysis in the panel on the left is compared against the nudged OpenIFS 48r1 simulation on the right. Both data fields were regridded to 1 x 1 deg before plotting. In spite of the much lower horizontal grid resolution used in OpenIFS the plumes compare reasonably well. 

(click on figure to enlarge)

Barbados Box Time Series

The next diagnostic averages AOD over the Barbados region (12–15°N, 60–57°W), following the regional box used by Rushley et al. (2026). The figure compares CAMS analyses with the nudged OpenIFS simulation and the four lagged-start hindcasts.

Figure: Barbados-box average dust AOD (top) and total AOD (bottom) at 550 nm during August 2023. The principal Saharan dust outbreak occurred during 14–16 August, when CAMS analyses show dust AOD exceeding 0.4 and total AOD approaching 0.7. All OpenIFS experiments reproduce the major event, although differences in peak magnitude and timing become apparent as forecast lead time increases.

Atlantic Dust Transport Hovmöller Diagram

The second diagnostic illustrates the westward propagation of Saharan dust across the tropical Atlantic. Dust AOD is averaged between 12°N and 15°N and plotted as a function of longitude and time.

Figure: Longitude–time (Hovmöller) diagram of dust AOD at 550 nm averaged between 12°N and 15°N. The left panel shows the CAMS analysis, the centre panel the ERA5-nudged OpenIFS simulation, and the right panel the free-running OpenIFS simulation initialised on 12 August 2023. The diagonal bands indicate westward transport of Saharan dust across the Atlantic towards Barbados (vertical dashed line). The OpenIFS simulations reproduce the large-scale propagation and timing of the main dust plume observed in the CAMS analyses.

Summary

This initial comparison demonstrates that OpenIFS 48r1 successfully reproduces even at a coarse horizontal grid resolution the large-scale evolution of the trans-Atlantic Saharan dust outbreak observed in CAMS analyses during mid-August 2023. The simulations capture the arrival of enhanced dust loading over the Barbados region during 14–16 August and reproduce the westward propagation of the plume across the tropical Atlantic. The Barbados time series show that all simulations reproduce the principal dust event during 14–16 August, while the Hovmöller diagrams confirm realistic westward transport across the tropical Atlantic. Differences between the experiments primarily reflect the influence of the forecast lead time and the evolution of the free-running aerosol fields. Because both OpenIFS and CAMS share the same composition framework and initial conditions, the comparison should be interpreted as a system-consistency assessment rather than an independent validation.