Qing Wang (now at the Naval Postgraduate School) and Don Lenschow study the role of cumulus clouds that grow upward to penetrate decks of marine stratocumulus from below. This process is most common in the transition zones where the large stratocumulus decks adjoin regions of trade-wind cumulus. Using observations from the Atlantic Stratocumulus Transition Experiment (ASTEX, conducted in June 1992 with the NSF/NCAR Electra aircraft), Qing and Don analyze the effects of the penetrating cumulus on the evolution of stratocumulus layers and on the dynamics of the atmospheric boundary layer.
Their study found that the stratocumulus layer, usually several hundred meters deep, is often decoupled from the surface so that the moisture from surface evaporation is not immediately available to replenish the layer. Yet the stratocumulus persist for days despite the entrainment of overlying warm, dry air into the cloud layer. Wind data from the Electra helped solve the puzzle. Moist air near the surface was converging beneath isolated cumulus, rising through the cloud, and spreading out horizontally in the stratocumulus layer to replace the moisture lost through entrainment. Other Electra measurements confirmed the role of the cumulus, as they showed large horizontal variations in such quantities as temperature, ozone, and liquid water.
In the same issue of JAS, Don joins lead author Chin-Hoh Moeng and David Randall (Colorado State University) in a modeling study of how stratocumulus decks dissipate. When these decks form in an environment of weak wind shear and cool ocean, the turbulent motions within them are driven by two kinds of cooling at cloud top. One is radiative, caused by heat radiating from the cloud tops into space. The other is evaporative, caused when drier air from above mixes into the cloud and is cooled by water evaporating into it.
Chin-Hoh and coauthors use a series of numerical simulations to compare the roles of these two processes in cloud breakup due to entrainment of air from above. The entrainment occurs when turbulent eddies penetrate upward at the cloud top and engulf air from above into the cloud. Since an inversion exists just above a stratocumulus deck, this process typically brings warmer and drier air into the cloud.
The authors "turned off" each type of cooling to examine the other. They found radiative cooling to be self-limiting: as the cooling produces entrainment, the entrained warm air reduces the effectiveness of the cooling in generating turbulence, which limits the subsequent entrainment. Thus, the cloud remains solid. In contrast, evaporative cooling stimulates a runaway process: the entrainment it produces brings dry air into the cloud, enhancing the rate of evaporation and stimulating even more cooling and entrainment. This leads to a quick breakup of the cloud deck.
The authors also note that other processes such as solar heating and drizzle could influence the stratocumulus evolution. Work is now under way using ASTEX and other data sets for real-world evaluation of the model results.
A major collaboration among SCD, MMM, NASA, the Ohio Supercomputing Center (OSC), and several other parties promises to stretch the boundaries of satellite-based data exchange for the huge outputs produced by atmospheric models. The project, called CO-OP 3D, is supported by the Advanced Research Projects Agency (ARPA). Its goal is to demonstrate the feasibility of data exchange via satellite for real-time model simulations.
High-speed links between NCAR, the OSC, and NOAA's Great Lakes Environmental Research Laboratory were routed through NASA's Advanced Communications Technology Satellite (ACTS). This space-based network was used for a real- time demonstration on 13 September at the ACTS Results Conference to carry out a nested-grid simulation at two sites. NCAR modeled the atmosphere above Lake Erie at 54, 18, and 6 kilometers (triple nests) using the Penn State/NCAR mesoscale model, version 5 (MM5).
Meanwhile, the OSC ran a 2-km MM5 model and 2-km lake model. Information was exchanged between the two runs in real time via the ACTS satellite using the Parallel Virtual Machines (PVM) message-passing system. Data was remotely rendered via the satellite using Explorer (a visualization package), so that scientists at each site could view model output as it progressed. Videoconferencing--also routed through the satellite--between scientists at NCAR and OSC allowed scientists to discuss model output face to face.
Involved in the collaboration were Chris Fair, Basil Irwin, and Marla Meehl (SCD Networking Group); Bill Boyd and Jordan Powers (MMM); and many others. Principal investigators were Bill Buzbee (SCD director) and Bill Kuo (MMM). The project wraps up on 31 October; a final demonstration will take place in the SCD visualization lab late in the month.