Models and Observational Predictions of Dust Traps in Protoplanetary Discs

TL;DR

This paper investigates how key dust-evolution parameters—disc viscosity $α$ and fragmentation velocity $v_{\rm frag}$—together with pressure bumps, influence dust retention and trapping in protoplanetary discs. Using Dustpy to model gas+dust evolution and RADMC-3D for radiative transfer, the study predicts millimeter continuum observables across multiple wavelengths, revealing strong degeneracies among parameters. It shows that pressure bumps can dramatically boost dust masses and create ring-like structures, with ring sharpness or shoulders depending on $α$ and $v_{\rm frag}$, while smooth discs require very low $v_{\rm frag}$ and moderate $α$ to retain substantial dust over Myr timescales. The work also links trapping to the pebble flux into the inner disc and its potential impact on volatile delivery, highlighting the need for multi-wavelength, high-resolution observations to robustly interpret disc substructures and their planet-forming implications.

Abstract

This manuscript investigates the impact of key dust evolution parameters on dust retention and trapping in protoplanetary discs. Using models with and without pressure bumps, combined with radiative transfer simulations, images of the dust continuum emission at (sub-)millimeter wavelengths, their fluxes and observed disc sizes are presented. For discs without pressure bumps (smooth discs), significant dust mass can only be retained over Myr timescales when dust fragmentation velocities are low (1m/s) and with viscosity values of $α=10^{-3}$. For such a combination of fragmentation velocity and viscosity, the synthetic images show a bright inner emission follow by a shallow emission with potential gaps if they are present in the gas profile as well. At higher fragmentation velocities (5-10m/s), most dust is lost due to radial drift at million-year timescales unless pressure traps are present, in which case dust masses can increase by orders of magnitude and structures are observed in synthetic images. The viscosity parameter strongly shapes observable features, with low $α$ producing sharper, potentially asymmetric inner wall cavities in inclined discs due to optically thick emission. High $α$ favors the appearance of shoulders around the predominant rings that dust trapping produces. However, distinguishing between different fragmentation velocities observationally remains challenging. The inferred dust disc sizes from synthetic observations do not always correspond directly to dust model sizes or to the location of pressure bumps. Finally, we discuss implications for pebble fluxes and the delivery of volatiles to the inner disc. These results emphasize the strong degeneracies among dust evolution parameters and highlight the need for multi-wavelength, high-resolution observations to disentangle the processes shaping the formation of planets in protoplanetary discs.

Figures (10)

• Figure 1: Sketch of the radial drift of particles in a smooth disc where $\partial_r P <0$ in the entire disc (left panel) vs. a disc where there is a pressure bump (right panel).
• Figure 2: Examples of ALMA observations of protoplanetary discs with structures from different observing programs, including andrews2018, long2019, Oberg2021, Teague2025, guerra2025, and Vioque2026. In the upper right of each panel, a bar of 0.1" is shown as a reference for the scale.
• Figure 3: Vertically integrated dust density distribution after 1 Myr of evolution, when assuming $R_c=20$ au for smooth disc models (top panels) and for models assuming pressure bumps (bottom panels). The columns correspond to different values of the fragmentation velocity ($v_{\rm{frag}}=1, 5, 10\,\rm{m\,s}^{-1}$, from left to right, respectively), and the rows correspond to different values of $\alpha$ ($\alpha=10^{-4}, 10^{-3}$, top and bottom, respectively). The corresponding $M_{\rm{dust}}$ and $R_{90, M}$ are given for each panel.
• Figure 4: As Fig. \ref{['fig:dust_distribution']}, but after 5 Myr of evolution.
• Figure 5: Dust volume density distribution $\rho_{\rm{dust}}$ as a function of height ($z$) and distance from the star. This is the case of $v_{\rm{frag}}=5$m s$^{-1}$ and two values of $\alpha$ ($\alpha=10^{-4}$ in the top panels and $\alpha=10^{-3}$ in the bottom panels). The different columns represent the grain size range that is considered: left ($<1\mu$m), middle($>0.1$mm), and right is when all grain sizes are considered. Results are shown for the case when $R_c=20$ au and after 1 Myr of evolution.