next up previous
Next: Modeling Up: Comparison of microscale Previous: Introduction

Observations

The UW-VIL was deployed in Sheboygan, Wisconsin, for the Lake-Induced Convection Experiment (Lake-ICE) during December of 1997 and January of 1998. The site (4344N, 8742W, 176 m ASL) was located within 10 m of the western shore of Lake Michigan. The VIL's beam-steering-unit (the point at which lidar beam is transmitted from) was located approximately 5 m above the lake surface. Thus, horizontal scans (PPIs) at 0 elevation allowed us to map the horizontal distribution of aerosol and steam-fog in a plane approximately 5 m above and parallel to the surface of the lake. Figure 1 is a PPI scan of this type. Steam-fog and hygroscopic aerosol produced a high-scattering tracer near the lake surface.


Figure 1. PPI of range-corrected backscatter intensity showing the organization of the steam-fog on 13 January 1998 from a few hundred meters to 5.9 km offshore. At the shore the mean wind during this time was from 280-290 at 5-10 m s and the air temperature was -20 C. The open-cells range in horizontal size from about 100 m at 1 km offshore to about 500 m at 5.9 km offshore.

In addition to measuring aerosol scattering on horizontal slices through the surface layer, the VIL is capable of making vertical slices (RHIs) through the entire mixed layer and mapping the 3-D structure of aerosol scattering in the boundary layer. By rapidly moving the laser beam in a series of RHIs, each with a slightly increased azimuth angle, we can measure the 3-D structure. For example, a volume scan spanning 40 in azimuth and 15 in elevation angle requires about 2 minutes. A typical change in elevation angle between two laser pulses during an RHI is 0.23. By repeating such volume scans, we can also monitor the temporal evolution of the structures.

Perhaps the most interesting VIL observations during Lake-ICE were open-cell patterns in the steam-fog about 5 meters above the surface of the lake on 10 and 13 January 1998. Cold air advection was occurring on both of these days and visual observations confirmed clear skies over the lidar site and steam fog on the surface of the lake. On 10 January the minimum temperature reached -16.7 C at 14 UTC with the wind from 236 at 6.5 m s. On 13 January the air temperature dropped to -20 C and the wind was from 280-290 at 5-10 m s. The lake water temperature on these days ranged from 3 to 5 C. In this paper we focus on the 13 January case, but we intend to present other cases at the poster.


Figure 2. RHI of range-corrected backscatter intensity showing the vertical structure of steam-fog on 13 January 1998 from a couple hundred meters to 6 km offshore. Narrow columns of steam-fog and aerosol can be seen above the lake surface. A 500-m deep mixed layer formed over land appears to be advecting offshore which is also indicated in the upwind radiosonde sounding in figure 3.

The horizontal cell dimensions increase with increasing offshore distance and appear to be slightly elongated in the direction of the wind. Their somewhat hexagonal shape allows any one cell to share most of its walls with neighboring cells. Cell widths on the left side of figure 1 range from approximately 100 to 500 m while cell widths on the right range from 500 to 1000 m. The streaks across the image are caused by attenuation from the steam fog.

While the steam on the 10th did not appear to rise more than about 50-m above the lake, RHI scans from 13 January, such as figure 2, reveal narrow rising columns of steam which sometimes extend to the top of a 500-m deep mixed-layer. The columns are very bright near the surface and decrease in intensity with altitude. In figure 2, there is one such feature at about 4.4 km range that extends from the surface up to about 200 m. Some of these features may be steam devils and we hope that the VIL observations of them will enable us to quantify their size and number density.


Figure 3. NCAR ISS CLASS sounding from 13 January 1998 at 16:30 GMT was used to initialize our model. The VIL also indicates a mixed layer extending up to about 500 m (about 950 mb) at the coast which is being advected over the lake by the larger scale flow. The wind profile shown here has been rotated so that the surface wind vector is normal to the north-south shoreline in the model.



next up previous
Next: Modeling Up: Comparison of microscale Previous: Introduction



Ed Eloranta
Tue Nov 16 08:30:34 CST 1999