Search Results

You are looking at 1 - 4 of 4 items for :

  • Author or Editor: Graham Feingold x
  • Monthly Weather Review x
  • Refine by Access: All Content x
Clear All Modify Search
Hailong Wang, William C. Skamarock, and Graham Feingold

Abstract

In the Advanced Research Weather Research and Forecasting Model (ARW), versions 3.0 and earlier, advection of scalars was performed using the Runge–Kutta time-integration scheme with an option of using a positive-definite (PD) flux limiter. Large-eddy simulations of aerosol–cloud interactions using the ARW model are performed to evaluate the advection schemes. The basic Runge–Kutta scheme alone produces spurious oscillations and negative values in scalar mixing ratios because of numerical dispersion errors. The PD flux limiter assures positive definiteness but retains the oscillations with an amplification of local maxima by up to 20% in the tests. These numerical dispersion errors contaminate active scalars directly through the advection process and indirectly through physical and dynamical feedbacks, leading to a misrepresentation of cloud physical and dynamical processes. A monotonic flux limiter is introduced to correct the generally accurate but dispersive solutions given by high-order Runge–Kutta scheme. The monotonic limiter effectively minimizes the dispersion errors with little significant enhancement of numerical diffusion errors. The improvement in scalar advection using the monotonic limiter is discussed in the context of how the different advection schemes impact the quantification of aerosol–cloud interactions. The PD limiter results in 20% (10%) fewer cloud droplets and 22% (5%) smaller cloud albedo than the monotonic limiter under clean (polluted) conditions. Underprediction of cloud droplet number concentration by the PD limiter tends to trigger the early formation of precipitation in the clean case, leading to a potentially large impact on cloud albedo change.

Full access
Bjorn Stevens, Robert L. Walko, William R. Cotton, and Graham Feingold

Abstract

The production of anomalous supersaturations at cloud edges other than cloud base has presented a vexing challenge for modelers attempting to represent the evolution of a droplet spectrum across an Eulerian grid. Although the problem manifests itself most dramatically for models that explicitly predict on the supersaturation field, it is also present in models with bulk condensation schemes in which condensation happens implicitly. Although the problem has been discussed in the context of truncation errors associated with finite difference approximations to advection, this note demonstrates more generally that the cloud-edge supersaturation problem is a fundamental problem associated with the ubiquitous assumption that the forcings on the droplet spectra are well represented by the mean thermodynamic fields. In certain respects, this assumption is equivalent to failing to represent fractional cloudiness within a grid. Although well-known consequences of this problem are the underprediction of temperature and the erroneous representation of the mean buoyancy flux within a grid box, we also demonstrate that the spurious production of droplets can arise in response to the spurious production of supersaturations in models with detailed microphysical representations.

Full access
Wayne M. Angevine, Joseph Olson, Jake J. Gristey, Ian Glenn, Graham Feingold, and David D. Turner

Abstract

Proper behavior of physics parameterizations in numerical models at grid sizes of order 1 km is a topic of current research. Modifications to parameterization schemes to accommodate varying grid sizes are termed “scale aware.” The general problem of grids on which a physical process is partially resolved is called the “gray zone” or “terra incognita.” Here we examine features of the Mellor–Yamada–Nakanishi–Niino (MYNN) boundary layer scheme with eddy diffusivity and mass flux (EDMF) that were intended to provide scale awareness, as implemented in WRF, version 4.1. Scale awareness is provided by reducing the intensity of nonlocal components of the vertical mixing in the scheme as the grid size decreases. However, we find that the scale-aware features cause poorer performance in our tests on a 600-m grid. The resolved circulations on the 600-m grid have different temporal and spatial scales than are found in large-eddy simulations of the same cases, for reasons that are well understood theoretically and are described in the literature. The circulations [model convectively induced secondary circulations (M-CISCs)] depend on the grid size and on details of the model numerics. We conclude that scale awareness should be based on effective resolution, and not on grid size, and that the gray-zone problem for boundary layer turbulence and shallow cumulus cannot be solved simply by reducing the intensity of the parameterization. Parameterizations with different characteristics may lead to different conclusions.

Free access
Mikael K. Witte, Patrick Y. Chuang, Orlando Ayala, Lian-Ping Wang, and Graham Feingold

Abstract

Two case studies of marine stratocumulus (one nocturnal and drizzling, the other daytime and nonprecipitating) are simulated by the UCLA large-eddy simulation model with bin microphysics for comparison with aircraft in situ observations. A high-bin-resolution variant of the microphysics is implemented for closer comparison with cloud drop size distribution (DSD) observations and a turbulent collision–coalescence kernel to evaluate the role of turbulence on drizzle formation. Simulations agree well with observational constraints, reproducing observed thermodynamic profiles (i.e., liquid water potential temperature and total moisture mixing ratio) as well as liquid water path. Cloud drop number concentration and liquid water content profiles also agree well insofar as the thermodynamic profiles match observations, but there are significant differences in DSD shape among simulations that cause discrepancies in higher-order moments such as sedimentation flux, especially as a function of bin resolution. Counterintuitively, high-bin-resolution simulations produce broader DSDs than standard resolution for both cases. Examination of several metrics of DSD width and percentile drop sizes shows that various discrepancies of model output with respect to the observations can be attributed to specific microphysical processes: condensation spuriously creates DSDs that are too wide as measured by standard deviation, which leads to collisional production of too many large drops. The turbulent kernel has the greatest impact on the low-bin-resolution simulation of the drizzling case, which exhibits greater surface precipitation accumulation and broader DSDs than the control (quiescent kernel) simulations. Turbulence effects on precipitation formation cannot be definitively evaluated using bin microphysics until the artificial condensation broadening issue has been addressed.

Full access