My first five months as a PhD student within the Dust-DN

Blog post from Francesco Moncada

I’m currently in my fifth month as a PhD candidate within the Dust-DN doctoral network, working on the DC12 project at the Barcelona Supercomputing Center (BSC). Time has flown by, and yet the world of atmospheric dust still feels very new to me: there’s so much to learn and explore, which makes the work both challenging and intriguing.

Although I come from a physics background—more specifically climate physics, having obtained my Master’s degree at Utrecht University—my previous research focused on a very different topic. During my Master’s thesis at IGE in Grenoble, I studied ice sheets, so transitioning into dust research has been a significant but exciting shift.

One of the main reasons I chose to explore this new field is the fascinating and complex objective of the DC12 project: improving the way mineral dust is represented in climate models. Dust particles vary in mineral composition, and these differences affect critical climate processes like solar radiation absorption, scattering and cloud formation. Better understanding and implementation of these interactions in model is essential for improving climate predictions and developing more robust climate scenarios for the future.

To tackle this challenge, I work with high-resolution satellite data from NASA’s EMIT mission, which provides detailed information about the surface mineralogy of major dust source regions. The goal is to integrate this data into the MONARCH model (Multiscale Online Nonhydrostatic AtmospheRe CHemistry), the chemical weather model developed at BSC. This will help us better capture how different mineralogies and types of dust—natural, and anthropogenic (resulting from human-altered land surfaces)—interact with the climate system.

Figure 1: Marenostrum5 infrastructure 

A key asset in this work is Marenostrum5, the High-Performance Computer (HPC) located at BSC and reported in figure 1. One of the most powerful supercomputers in Europe, it spans 180 m²—roughly a third the size of a basketball court—and gives us the computing power needed to run complex simulations at high resolution.

So far, I’ve been learning how to work with the HPC environment, getting familiar with the structure of the MONARCH model, and running and validating simulations. Our current line of research focuses on investigating the contribution of anthropogenic and natural dust to climate impact.

Figure 2: DOD at 550 nm modeled by MONARCH model

In figure 2 is possible to observe a map of the annual average Dust Optical Depth (DOD) at 550 nm, as modeled by MONARCH, for both natural and anthropogenic dust sources. DOD is a dimensionless metric that indicates how much incoming solar radiation is absorbed and scattered by dust particles in the atmosphere, making it a key variable for assessing dust impacts on climate, air quality, and radiative forcing. In Figure 3, I reported the comparison between the modeled DOD at 550 nm over Northeast Africa—differentiating between natural and anthropogenic dust—and the observations available in that region, retrieved and appropriately filtered from AERONET, a global network of ground-based sun photometers that provides long-term, continuous aerosol optical measurements. . Since the focus is on dust aerosol, the AERONET dataset was filtered using the Ångström exponent, a parameter sensitive to particle size and thus effective in excluding non-dust aerosols, which typically consist of smaller particles. This validation step is essential to assess the model’s ability to realistically simulate dust processes and ensure the reliability of its outputs.

Figure 3: Validation of the DOD 550 nm modeled by MONARCH in North East Africa through comparison with AERONET observational data. 

Being part of the DUST-DN network has so far been a great opportunity to connect with experts of the dust field and to feel part of a collaborative and supportive community. I’m especially excited about the upcoming workshop and training school in Cyprus next April, where I’ll have the chance to present my first results and meet the other doctoral candidates in person.