Convective Mass Flux and Surface Precipitation Across Tropical Convective Life Cycles

With the advent of high-resolution operational weather models and km-scale global models, atmospheric modeling is producing unprecedented amounts of data. However, improving our understanding of convective processes from these simulations remains challenging using traditional analysis methods, particularly when it comes to comparing convection across regions, model frameworks, and convective types.

Here, we apply an object-based approach to explore how key convective properties—particularly convective mass flux (CMF) and surface precipitation rates—vary across a broad range of convective storm morphologies captured in the NASA INvestigation of Convective UpdraftS (INCUS) simulation dataset. The current INCUS simulation dataset comprises over 30 model runs that capture a range of convective environments across the tropics and subtropics, simulated using multiple models and microphysical parameterizations. This approach allows us to explore the relationship between grid spacing, mass flux, and surface precipitation across a diverse sample of storm environments, developmental stages, and morphologies. Simulations are run using the Regional Atmospheric Modeling System (RAMS) and Weather Research and Forecasting (WRF) models. We examine the properties of convective systems across three nested grids (Δx = 100m, 400m, and 1.6km). We then use the tobac (tracking and object-based analysis of clouds) package to identify and track individual clouds at high temporal resolution (30s output) over the course of their life cycles.

In this presentation, we compare cloud properties across the broad range of environments and convective types represented in the INCUS dataset. Furthermore, we explore how convective mass flux and surface precipitation evolve across stages in the cloud life cycle, and how this development varies according to factors such as cloud top height and convective organization. Finally, we will discuss how the representation of this convective life cycle varies as a function of grid spacing. We show that as the grid spacing is made finer, the cloud population distribution shifts towards more numerous but shallower and weaker updrafts that do not develop as rapidly compared to those in simulations with coarse grid spacings. Our results demonstrate the utility of object-based analysis in developing process-level understanding from large cloud-resolving simulation datasets.