On numerical broadening of particle-size spectra: a condensational growth study using PyMPDATA 1.0

Abstract. This work discusses the numerical aspects of representing condensational growth particles in models aerosol systems such as atmospheric clouds. It focuses on Eulerian modelling approach, which fixed-bin discretisation is used for probability density function describing particle-size spectrum. Numerical diffusion inherent to employment solving arising transport problem (advection equation size spectrum evolution). The focus this a technique reducing solutions based upwind scheme: multidimensional positive definite advection algorithm (MPDATA). Several MPDATA variants are explored including infinite-gauge, non-oscillatory, third-order terms and recursive antidiffusive correction (double-pass donor cell, DPDC) options. Methodologies handling coordinate transformations associated with both choice grid layout expounded. Analysis performance scheme different parameters settings performed using (i) an analytically solvable box-model test case (ii) single-column kinematic driver (“KiD”) size-spectral due condensation solved simultaneously vertical spatial coordinate, supersaturation evolution coupled droplet through water mass budget. covers dynamics only; no dimension considered. involves solution two-dimensional (spectral dimensions). discussion presented paper size-spectral, temporal convergence well computational cost, conservativeness quantification broadening simulations demonstrate that, compared solutions, even 10-fold decrease spurious spectral can be obtained by apt variant (maintaining same resolution), yet at increased cost. Analyses reveal that width affected pertinent advection. Application single corrective iteration robustly decreases relative dispersion spectrum, roughly factor 2 levels maximal liquid content. Presented carried out PyMPDATA – new open-source Python implementation Numba just-in-time compilation infrastructure.