Search Results

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

  • Author or Editor: K. Shankar Rao x
  • Journal of Applied Meteorology and Climatology x
  • Refine by Access: All Content x
Clear All Modify Search
K. Shankar Rao
,
Richard M. Eckman
, and
Rayford P. Hosker Jr.

Abstract

During the 1984 ASCOT field study in Brush Creek Valley, two perfluorocarbon tracers were released into the nocturnal drainage flow at two different heights. The resulting surface concentrations were sampled at 90 sites, and vertical concentration profiles at 11 sites. These detailed tracer measurements provide a valuable dataset for developing and testing models of pollutant transport and dispersion in valleys.

In this paper, we present the results of Gaussian puff model simulations of the tracer releases in Brush Creek Valley. The model was modified to account for the restricted lateral dispersion in the valley, and for the gross elevation differences between the release site and the receptors. The variable wind fields needed to transport the puffs were obtained by interpolation between wind profiles measured using tethered balloons at five along-valley sites. Direct turbulence measurements were used to estimate diffusion. Subsidence in the valley flow was included for elevated releases.

Two test simulations—covering different nights, tracers, and release heights—were performed. The predicted hourly concentrations were compared with observations at 51 ground-level locations. At most sites, the predicted and observed concentrations agree within a factor of 2 to 6. For the elevated release simulation, the observed mean concentration is 40 pL/L, the predicted mean is 21 pL/L, the correlation coefficient between the observed and predicted concentrations is 0.24, and the index of agreement is 0.46. For the surface release simulation, the observed mean is 85 pL/L, and the predicted mean is 73 pL/L. The correlation coefficient is 0.23, and the index of agreement is 0.42. The results suggest that this modified puff model can be used as a practical tool for simulating pollutant transport and dispersion in deep valleys.

Full access
Carmen J. Nappo
,
K. Shankar Rao
, and
Jerold A. Herwehe

Abstract

The characteristics of pollutant transport and diffusion of a passive contaminant in a two-dimensional katabatic flow over a simple slope are examined using a primitive equation hydrodynamic model. It is shown that pollutants released above the drainage layer can be entrained into the layer and diffused to the ground surface. For elevated release within the drainage layer, subsidence in the flow leads to relatively high surface concentrations of pollutants close to the stack. Pollutants released at ground level can spread through the entire depth of the drainage layer. This vertical diffusion is more effective for a shallow slope, resulting in higher concentrations at all heights, than for a steeper slope. These dispersion characteristics are quite different from those for stable flows over flat terrain. The differences result from increases of boundary-layer depth, wind speed, and turbulence as the katabatic flow develops downslope.

The katabatic flow and dispersion model is tested by simulating the perfluorocarbon and heavy methane tracer releases for Night 4 of the 1980 ASCOT field study in Anderson Creek Valley, California. These tests show that the observed concentrations and the depth of the drainage layer in the lower region of the slope are underpredicted because the model could not simulate the convergence of drainage air (pooling) in the valley basin. The nightly average values of the observed concentrations, however, are predicted well. It is concluded that the model is applicable to nearly two-dimensional open slopes.

Full access
Christoph A. Vogel
,
Dennis D. Baldocchi
,
Ashok K. Luhar
, and
K. Shankar Rao

Abstract

Several methods for estimating surface energy balance components over a vegetated surface are compared. These include Penman-Monteith, Deardorff, and multilayer canopy (CANWHT) models for evaporation. Measurements taken during the 1991 DOE-sponsored Boardman Area Regional Flux Experiment over a Well-irrigated, closed wheat canopy are used in the comparison. The relative performance of each model is then evaluated. It is found that the Penman-Monteith approach using a simple parameterization for stomatal conductance performs best for evaporation flux. The Deardorff model is found to have the best relative performance for sensible heat, while the CANWHT model gives the best results for net radiation and soil heat flux. The Priestley-Taylor model for evaporation and a resistance-analog equation for sensible heat flux are also tested.

Full access
Ronald J. Dobosy
,
K. Shankar Rao
,
John W. Przybylowicz
,
Richard M. Eckman
, and
Rayford P. Hosker Jr.

Abstract

Fluxes and flux-divergences of mass and momentum in Brush Creek Valley, computed from measurements taken by Tethersondes and Doppler sodars in the 1984 ASCOT experiment, are presented. Estimates of mass influx from open sidewalls in Brush Creek, derived from concurrent tower measurements, are also given. Mass and momentum fluxes calculated from single-profile data were within a factor of 1.5 of those obtained by integrating Doppler lidar data. Flux-divergences for budget calculations should be derived from a Doppler lidar or equivalent remote sensor data, because single-profile measurements were found to have sampling errors which are too large for reliable flux divergence estimates. The mass influx from the sidewalls was insufficient to account for the mass flux-divergence in the main valley. This imbalance in the drainage flow mass budget is speculated to be due to the inflow from the small box-canyon tributaries, rather than from subsidence of air above the main valley.

Full access