How important is microphysical variability to atmospheric cloud processes? Quantification of variability and impact on forecast model results.

Poster PDF

Description

To study a multitude of complex interactions between atmospheric ice particles that cannot be physically reproduced in the laboratory, a typical preliminary approach in modeling and experimentation is to reduce the number of variables that affect such interactions. We will present results from two related efforts to constrain and better understand particle interactions. First, an effort to constrain aggregation sticking efficiency estimates: In cold environments, the time required for particle fusion (the thermodynamic sticking of one particle to another) is long, greatly reducing the likelihood for ice crystals aggregating via sintering of crystal surfaces alone suggesting that the interlocking of branches should dominate the aggregation process. During the March 2000 DOE ARM IOP over the Southern Great Plains site, the University of North Dakota Citation research aircraft made high quality Cloud Particle Imager (CPI) observations at multiple levels in a near steady-state cloud composed mostly of bullet rosettes and aggregates of bullet rosettes at around -40C. In this study we hypothesize and identify the necessary processes for bullet rosette aggregation. Using a modified version of the Ice Particle Aggregation Simulator model (IPAS) that uses CPI images of bullet rosettes we estimate the likelihood of two bullet rosettes aggregating by interlocking arms. Initial results indicate that the aggregation of bullet rosettes is highly sensitive to estimated terminal velocities. Second, the impact of these results on model results: With the variability identified in the observations, similar variability is implemented into WRF using the P3 microphysics scheme, using an idealized 2D orographic case, with two approaches: 1) random fall speed fluctuations based on the observed fall speed PDF and 2) plus/minus one standard deviation of the mean. The simulations investigate the role of potentially random errors in our understanding of ice crystal fall speeds and systematic biases, respectively.