柳瀬 友朗

Abstract Clouds play a central role in climate physics by interacting with precipitation, radiation, and circulation. Despite being a fundamental issue in convective organization, the self-aggregation of clouds lacks a theoretical explanation due to its complexity. In this study, we introduce an idealized mathematical model where the system’s state is represented solely by the vertically integrated water vapor content of atmospheric columns under the weak temperature gradient approximation. By analyzing the nonlinear dynamics of this simplified system, we mathematically elucidate the mechanisms that determine the onset of self-aggregation and the spatial scale of the self-aggregated state. Nonlocal coupling between atmospheric columns induces bistability, leading to dry and moist equilibria. This reflects the circulation effects driven by horizontal differential heating due to convection and radiation. The bistable self-aggregated state realizes when destabilization by nonlocal coupling, triggered by finite-amplitude disturbances in the uniform state, overcomes stabilization by diffusion. For globally coupled systems where all columns are equally coupled, perturbations with the maximum wavelength exhibit the highest growth rate. This results in a solution with an infinitely long wavelength, understood as the dynamical system’s heteroclinic trajectories describing the steady state’s spatial evolution. Conversely, for nonlocally coupled systems with finite filter lengths, perturbations with wavelengths close to the characteristic length of the coupling are preferred. The results reveal that the balance between nonlocal coupling and diffusion is essential for understanding convective self-aggregation. Moreover, this study suggests a physical similarity between convective self-aggregation and the moisture mode. Significance Statement Cloud self-organization is a longstanding fundamental problem in climate physics, and its representation in climate models may contribute to uncertainties in future climate projections. Clouds are complex phenomena, intricately connected to latent heat release during water phase changes, buoyancy-driven fluid motion, and interactions with radiation. As a result, their detailed modeling is ongoing. Efforts have also been undertaken to understand the macroscopic behavior of clouds through simple mathematical descriptions. In this study, we semianalytically reveal the mechanisms underlying the spontaneous clustering of clouds and the characteristic distances between clusters using an idealized mathematical model that describes the spatiotemporal variation of water vapor content in tropical atmospheric columns.

Precipitation droplets are influenced by environmental fields and transform in time and space, following cloud microphysical processes. Accordingly, a raindrop size distribution (DSD) changes shape in a various form. However, DSDs cannot be calculated directly in radar or bulk models and are expressed using an approximate function. Exponential and gamma distribution are well-known as approximation functions, but there are DSDs of shapes that cannot be represented by these functions. One of them is a bimodal DSD with two peaks. Previous modeling studies have indicated that the bimodal DSD is formed when the collision-breakup process reaches equilibrium. On the other hand, recent observation-based studies have discussed the influence of convective activity within the precipitation system on forming the bimodal DSD. However, observations have not been able to quantitatively study the microphysical changes of individual particles and have yet to reveal the formation mechanisms within the precipitation system. In this study, we investigated quantitatively the process of the formation of the bimodal DSD by two-dimensional simulation of multicellular convection with the bin method. The simulation results showed that the bimodal DSD was formed during the updraft and downdraft in the mature stage of the multicell. Additionally, the bimodal DSD was formed at lower altitudes where there was inflow into the precipitation system. Particles that constituted the maximum of the bimodal DSD were found to have been advected by the inflow. Particles that constituted the local maximum dropped against the updraft. In contrast to these, particles that constituted the local minimum were less affected by the inflow and had difficulty dropping against the updraft. These results suggested that the bimodal DSD was formed by horizontal and vertical size sorting because of inflow and updrafts in the mature multicellular convection. In the future, it is necessary to simulate the reproduction of observed cases and compare them with observations.

Accepted on Sep 8, 2022.

Accepted on Sep 7, 2022.