Page 1:

• Stable PBL
• Interaction between heterogeneous surface and atmosphere
• Interaction between ocean and atmosphere
• Subfilter-scale turbulence in LES

Go to Page 2:

• International H2O Project (IHOP)
• Convective PBL and turbulence dispersion
• Wildfire research

Related websites:
http://www.joss.ucar.edu/cases/ (CASES 97)
http://www.mmm.ucar.edu/science/abl/cases/cases.html (CASES 99)

CASES-97

During the last few years, Margaret LeMone, with Kyoko Ikeda, and Robert Grossman (then at University of Colorado, now at Colorado Research Associates) analyzed the horizontal variability in the potential temperature at 2 m T2m. They found that the temperature varied linearly with station elevation, and the elevation-dependence was dry-adiabatic for sufficiently high Froude numbers. This year, the study was extended to include the evolution of the pattern through the night, and the departures from the linear trend.

Variation through the night: The evolution depended on ambient conditions. For clear-sky, light wind conditions, the horizontal variability in T2m increased through the night, as a watershed-scale circulation developed. As the total horizontal variability (standard deviation) increased, the standard deviation with respect to the best fit decreased. Other researchers found maximum total variability at earlier times. The results of this study suggest that the time elapsed between sunset and maximum horizontal variability is related to the scale of the terrain, the size of the array, and the spacing between stations. For strong winds and weakly stable thermal stratification, maximum horizontal variability occurred a few hours after sunset, and is linked to horizontal advection and vertical mixing.

Departure from the linear trend: Sites with thick surface cover (winter wheat) had cooler minimum temperatures than sites with bare ground, compared to the linear trend. Sites on ridges tended to be "too warm" and sites near bluffs tended to be "too cold."

CASES-99

During the past year, Jielun Sun and Sean Burns focused on processing the pyrgeometer data collected from CASES-99 and analyzed nocturnal heat budget and turbulence intermittency. Burns conducted careful radiation data processing and discovered mismatches between data loggers and the pyrgeometers during the field campaign, and consequently devoted considerable effort to correct the errors.

Sun focused on origins of intermittent turbulence by analyzing the night of October 18, 1999 with all the available observations, by closely collaborating with colleagues involved in CASES-99. She found that the passage of the atmospheric disturbances of a density current, solitary waves, and internal gravity waves were responsible for the intermittent turbulence. Turbulence adjacent to the ground can be generated by directional wind shear, pressure change, and drainage flows. Turbulence above the ground can be generated by thermal instability from overturning of cold air above warm air by atmospheric waves or large eddies, and can also be generated by shear instability from wind gusts associated with the atmospheric disturbances. Quiescence from turbulence suppression can come from descending motions associated with passages of atmospheric disturbances due to the mass balance, energy dissipation in nocturnal cooling air, and directional shear which strengthens local stability. Occurrence of these instability and stability factors at various time and heights lead to apparent intermittent turbulence.

Sun, in collaboration with Steve Oncley, Tony Delany, and Tom Horst (all NCAR/ATD), designed instrumentation for the radiative flux divergence measurements during CASES-99, which resulted in both observed radiative flux and sensible heat divergence over a relatively deep layer (between 2 m and 48 m) for the first time in history. Complications associated with heterogeneous surfaces were considered in the radiation flux divergence measurement, to avoid the ambiguity caused by two different fields of views from the radiative flux measurement at two levels. Through analysis of this data set, Sun studied the roles of the sensible heat and radiative flux divergence in the heat balance and their contribution to the applicability of Monin-Obukhov similarity theory in stable boundary layers. It was discovered that, in general, the largest radiative flux divergence occurs in the early evening under weak wind conditions following several days of warm trend. Averaging over the entire field campaign, the radiative flux divergence tends to be smaller than the local cooling and the sensitive heat divergence between two levels. However, the day-to-day variation of the sensible heat flux divergence tends to be larger than the radiative flux divergence. The radiative cooling observed between 2 m and 48 m is much smaller than the radiative cooling observed below 10 m in the literature. This implies that the radiative flux divergence may decrease with height as suggested by numerical radiation transfer models. Wind speed can change not only the sensible heat flux, but also the surface longwave radiation due to variations of surface area-exposure of warmer grass stems and soil surfaces compared to cool grass top, leading to fluctuations of the radiative flux divergence throughout the night.

Sun and Burns also studied dominant eddy scales of the wind, vertical velocity, water vapor specific humidity and temperature in the nocturnal boundary layer by using the Haar wavelet transform. They found that about 50% of the nights, the horizontal wind eddy scale is about 10 min while the vertical velocity is about 3 min, which indicates "pancake" type of motions in the nocturnal boundary layer. Other work included collaboration with Greg Poulos (Colorado Research Associates) on comparisons between model flux parameterizations and observed flux data from CASES-99.

Influence of soil moisture heterogeneity on the PBL

Edward Patton (Pennsylvania State University), Sullivan, and Moeng used their clear-PBL large-eddy simulation code that was recently coupled with the NOAH (National Center for Environmental Prediction / Oregon State University/Air Force/Office of Hydrology) land-surface model to study the planetary boundary layer (PBL) response to large-scale soil moisture heterogeneity (ranging from 2-30 km). In the presence of heterogeneity, the atmosphere transports moisture differently depending on the initial moisture state in the overlying atmosphere, wet versus dry. In both situations, land-surface heterogeneity induces organized motions that scale with the heterogeneity (see Movie 8). However, depending on the moisture state of the overlying atmosphere, the phase-correlated component can either be the sole contributor to the vertical water vapor mixing ratio flux, or make zero contribution (Figure 35 b.1). One of the important findings of this study is that if researchers plan to use the eddy-correlation technique to measure vertical water vapor mixing ratio fluxes at a point within a region of large-scale moist or dry soil conditions, they could dramatically misestimate the vertical fluxes if they ignore the contributions to the flux from organized motions, induced via heterogeneous surface forcing.

Figure 35. Vertical profiles of the normalized total mixing ratio flux (green) and the contribution to the total from the large-scale organized motions (blue) and the background turbulence (red). The top row presents results from cases where the atmosphere is initially dry (wetting PBLs), and the bottom row from cases where the atmosphere is initially wet (drying PBLs). The left-most panels are the control cases where the land-surface is horizontally homogeneous. The middle and right panels present results from cases with horizontally heterogeneous surface forcing with ? = 5 km and ? = 2 km respectively.

Atmospheric response to spatial variations of soil moisture

Related website: http://www.mmm.ucar.edu/science/abl/sgp/sgp.html

Sun investigated responses of the atmospheric moisture flux to temporal and spatial variations of soil moisture and spatial variation of vegetation, by analyzing the aircraft data collected from the Southern Great Plains (SGP) experiment. She focused on the atmospheric moisture flux collected by the Canadian Twin Otter aircraft along the El Reno and Kingfisher tracks, where both atmospheric turbulence and Electronically Scanned and Thinned Array Radiometer (ESTAR) soil moisture were well sampled. She found that the latent heat flux is influenced by the relatively fast temporal variation of soil moisture and relatively slow spatial variation of evapotranspiration from the vegetation root zone. The soil moisture content determines the importance of the spatial variation of the vegetation in the spatial variation of the atmospheric moisture flux. Extremely wet and extremely dry conditions would diminish the contribution of the spatial variation of the vegetation in the atmospheric moisture flux. The spatial variations of the vegetation play a dominant role in the spatial variation of the latent heat flux when the soil moisture is below about 20%, but not dry enough to cause severe vegetation damage.

Horizontal transport of CO2 over complex terrain

Sun led an investigation of horizontal transport of CO2 at Niwot Ridge, in collaboration with Steve Oncley, Tony Delany, Britt Stephens and Teresa Campos (all NCAR/ATD), and Alex Guenther and Andrew Turnipseed (both NCAR/ACD), Russ Monson (University of Colorado), Dean Anderson (U.S. Geological Survey), and Brian Lamb (Washington State University). A pilot experiment investigating the horizontal transport of CO2 within a forest canopy layer was conducted in August/September of 2002. The CO2 concentrations were measured at three to four levels within the canopy layer, at eight locations, within a 300 m2 area. To ensure the relative accuracy between locations, the CO2 concentration was measured by two central CO2 analyzers and 18 inlets, from various levels, at various locations. In addition, a central CO2 inter-comparison system was designed to provide quality-control for the CO2 measurement from different CO2 sensors. Further information related to this work, which was supported by NCAR Opportunity Funds and the NCAR Biogeosciences Initiative, is available at http://www.mmm.ucar.edu/science/abl/forest/.

Investigation of impacts of cold land transition in the global climate change

Related website: http://www.mmm.ucar.edu/science/abl/floss/

Burns assisted Larry Mahrt (Oregon State University) in setting up MMM thermocouples on a 20-m tower near Walden, Colorado, in early December 2001, during Fluxes over Snow Surfaces (FLOSS). He also was involved in monitoring the thermocouple temperatures during the experiment to ensure the quality of the data. The FLOSS field experiment was conducted in collaboration with Steve Oncley, Tony Delany, and Steve Semmer (all NCAR/ATD), and Larry Mahrt, Dean Vickers, and Richard Cuenca (all Oregon State University).

Effect of wave breaking on the ocean boundary layer

Sullivan, James McWilliams (University of California, Los Angeles) and Moeng, in collaboration with Ken Melville (Scripps Institute of Oceanography), are developing a stochastic wave breaking model for use in simulations of the ocean boundary layer (OBL). Turbulence-resolving simulations of the OBL typically impose a mean (constant) surface wind stress at the water surface, despite the known spatial and temporal variations in stress. This assumption neglects intermittent processes and, in particular, wave breaking, which is believed to be an important mechanism for transferring momentum and energy from waves to the ocean currents. The wave breaking model for the OBL neglects the complexity of a full air-water microphysical interface and, instead, focuses on important bulk processes, intermittent momentum and energy transmission from breaking waves. Direct numerical simulations are initially being used to test various aspects of this model. Preliminary results are shown in Figure 36 c.1.

Figure 36. Visualization of the horizontal current field u, panels left and right of a), and vertical momentum flux uw, panels left and right of b), at z/h = -0.014 below the water surface. Snapshots in the upper panels are for an OBL driven by a constant surface stress while the flow fields in the lower panels contain strong breaker forcing. Wave breaking is observed to greatly disrupt the elongated near-wall streaky structures associated with wall-bounded shear flows. Intense impulses from the breaking waves impart strong vertical momentum flux, both positively and negatively signed, to the water column. Even though the observed white cap covereage of the breakers at the surface is very intermittent (less than 2\%) the presence of breaking eventually stirs the entire surface layer of the OBL.

Effect of ocean waves on the atmospheric surface layer

Anna Rutgersson (Uppsala University and the Swedish Meteorological and Hydrological Institute) and Sullivan are investigating the effect of waves on the turbulent structure in the atmosphere using Direct Numerical Simulations (DNS) databases. There have been experimental indications that the effect of very fast waves (swell) over the ocean is different than previously believed. This could have large impact on global climate models, since the world oceans cover such a large part of the globe. The DNS data show significant wave-induced peaks in power spectra that extend up to higher levels for very fast moving waves. When separating the vertical momentum flux into quadrants, the cases with waves, and especially very fast waves, have larger amplitude transport events, with both positive and negative transport related to the waves. This can be seen, both in the DNS-data, and from measurements over the sea. Figure 37 c.2 shows the relative contribution from the four quadrants from DNS (blue) and measurements taken over the Baltic Sea (red). In spite of the differences between DNS-simulations and measured situations, the structure of the quadrants is similar. All quadrants have larger contributions during swell in both the measurements and the DNS.

Figure 37. Quadrant analysis from DNS (blue) and measurements from the Baltic Sea (red) during growing sea (upper) and swell (lower). Note that there are different scales on the axis for growing sea and swell.

Relationship between the atmospheric momentum transport and oceanic waves

Related websites:
http://www.mmm.ucar.edu/science/abl/showex/showex.html
http://www.mmm.ucar.edu/science/abl/cblast/

Sun and Burns, in collaboration with Douglas Vandemark (NASA Goddard Space Flight Center), Mark Donelan (University of Miami), Tim Crawford, Jerry Crescenti, and Jeff French (all NOAA Air Resources Laboratory), and Larry Mahrt (Oregon State University), continued investigation of retrieving two-dimensional wave spectra, using three laser altimeters, on board the LongEZ aircraft. Sun and Burns found that the wavelet analysis method is able to retrieve wavenumbers, based on wave slopes, simultaneously measured by the three laser altimeters. Therefore, the wavenumber is independent of where the wave slope is collected (i.e., whether the wave slope is collected along a straight line under the aircraft track, whether the aircraft flies at an angle from the wave propagation direction, or the aircraft flies at variable speed). However, when the wave slope is too small to be measured by the three laser altimeters, or the random noise of the measured wave slope is large, the wave analysis method may fail to retrieve the wave propagation direction. This result implies that the size of the triangle where the three laser altimeters were located can affect retrievable wavelengths.

Related websites:
http://www.atd.ucar.edu/sssf/projects/hats
http://www.whoi.edu/science/AOPE/dept/CBLASTmain.html

Sullivan, Lenschow, and Moeng, in collaboration with Thomas Horst (NCAR/ATD), Jeff Weil (Cooperative Institute for Research in the Environmental Sciences/Colorado University), and Jan Kleissl, Charles Meneveau, and Marc Parlange (all Johns Hopkins), continued their analysis of the sonic array data taken during the Horizontal Array Turbulence Study (HATS) field campaign. The objective of this effort is to improve subfilter scale (SFS) parameterizations in LES models. From the HATS dataset, they found that subfilter-scale (SFS) turbulent motions in the atmospheric surface layer are sensitive to the relative positions of the spectral peak of the vertical velocity ?w, and the filter cutoff scale ?f. A priori testing of SFS models with the HATS dataset shows that the turbulent kinetic energy and Smagorinsky model coefficients Ck and Cs, derived from the measurements, are smaller than the theoretical estimates based on the assumption of a sharp spectral cutoff. The correlations between measured and observed SFS fluxes and energy transfer are relatively low for an eddy viscosity model, and mixed SFS models that explicitly include a Leonard term exhibit higher correlations (see Figure 38 d).

Figure 38. Correlations between the HATS observations and an eddy viscosity model, and a mixed model (eddy viscosity plus a Leonard term) for the different components of the SFS flux tensor tau. The colored symbols denote different sonic array configurations. The correlation for the vertical momentum flux tau is shown as a solid line. Note the improved correlation obtained by explicitly including the Leonard term.