This document is a standard product of the NCAR/ATD/SSSF ASTER facility, which gives an overview of the measurements taken and conditions during the DOWNWASH94 field experiment.

ASTER data are stored in two forms:

Also available is a computer-readable log of comments noted by ASTER personnel. The logbook can be read here (sorry, some comments are garbled and this version cannot be searched). For access to the ASTER data, please contact SSSF. For access to other data from the DOWNWASH program, please contact Dave Strimaitis at Sigma Research Corporation.

Several photographs are available in compressed TIFF format.

Results from this experiment are described in:

Schulman LL, Scire JS. The development of the Plume Rise Model Enhancement (PRIME): The EPRI plume rise and downwash modeling project. 9thJoint Conference on the Applications of Air Pollution Meteorology with the AWMA, Atlanta, GA. January 28-February 2, 1996.

Lloyd L. Schulman , David G. Strimaitis & Joseph S. Scire (2000) Development and Evaluation of the PRIME Plume Rise and Building Downwash Model, Journal of the Air & Waste Management Association, 50:3, 378-390, DOI: 10.1080/10473289.2000.10464017

Experiment Description

DOWNWASH94 (Full name: Combustion Turbine Plume Rise and Building Downwash Model Development) was an EPRI-funded program with principal investigator Lloyd Schulman (Sigma Research Corporation). This program was designed to provide data for the development of a new computer model to represent the plume rise and building-induced downwash from power plants, especially those with "less than Good Engineering Practice" stack heights. In particular, the research goal was to define conditions under which the exhaust plume from a combustion turbine would be drawn downward due to the turbulence induced by the structure housing the turbine.

Norman Bowne (ENSR Consulting and Engineering) was the manager for the field phase of this program which, in addition to ASTER, included an auxiliary weather station and tethersonde operated by ENSR and a scanning aerosal LIDAR operated by Norm Nielsen (SRI International). New Jersey Power and Light also operated a Doppler SODAR at a nearby plant. The entire EPRI program also included wind-tunnel and numerical modeling studies of plume rise and downwash.


This experiment was located in Sayreville, New Jersey (about 20 miles south of Newark) at the Sayreville Generating Station. This site was quite complex, with the CT stacks about 12m high, the three oil storage tanks to the west about 18m high, and the main power plant to the northwest about 45m high, with stacks going up to 67m. Bordering the plant were marsh lands and the Raritan River.


The DOWNWASH Tower Layout shows the location of the ASTER towers ("cav1", "cav2", and "far") used for this program relative to structures in the power plant. Most of the towers were located near the southern-most combustion turbine (CT4) to obtain mean and turbulent flow statistics in a two-dimensional cross section aligned with the expected mean flow direction. (The DOWNWASH Tower Survey, shows detailed locations of these towers.) The third tower ("upwind") was located in an unused parking lot near the plant entrance to provide reference flow measurements. The DOWNWASH Configuration documents which sensors were deployed on each of the towers. Each sensor is labeled by the quantities it measures, its name, its height on the tower, and the tower name. DOWNWASH Sensor Table lists the type, manufacturer, and other specifications of the sensors used for this project. "Cav1" had three levels of mean temperature and humidity sensors and two levels of turbulence sensors (three-dimensional sonic anemometer and temperature sensor). "Cav2" had three levels of mean wind speed and direction sensors (propeller-vanes). "Far" had one level with a mean wind sensor and another level with turbulence sensors, which can be combined to give a two-level wind profile. The reference "upwind" tower also had three levels, each with wind, temperature, and humidity sensors and one with turbulence sensors (including a fast-response hygrometer). Also included at the "far" tower were measurements of net, upward and downward visible and infrared radiation, radiometric surface temperature, and precipitation.

The non-standard (not included with ASTER) sensors used for this program were:


This program occurred during one of the worst winters on record for the northeastern U.S., characterized by freezing rain, cold temperatures which caused "brown-out" electrical power conditions, and flooding due to ice-jammed rivers. Flooding extended to the "upwind" and "far" towers in conjunction with a tidal surge on day 63, which damaged the data acquisition system for the "far" tower.

Albedo from the radiometers should give a good indication of when snow was on the ground (most of the time). Surface conditions (including snow depth) often were noted in the system logbook.

Setup of cav1 and cav2 was delayed until day 41 while CT4 was being repaired. During this time, data were being collected from these towers while they were laying close to the ground.

Known instrument problems:

Refer to the logbook for more details on sensor operation.


Day: Action

Daily Plots

The following plots summarize conditions during each day of the project. Each plot covers one Julian day (0000-2359 GMT) and is labeled with time in GMT at the bottom and local time (EST) at the top. The top panel displays temperature and specific humidity from the 2m upwind tower, pressure, and precipitation rates (if present). Below that is a plot of wind speed and direction from the 6m upwind tower, with dotted lines showing the best directions for downwash to occur. The next panel shows net radiation measured near the far tower, and sensible heat flux from 10m on the upwind tower. The bottom panel shows the Monin-Obukhov stability parameter, z/L, and the friction velocity, u*, calculated from the 10m upwind data. Since these fluxes and derived parameters are based on smoothed, 5-minute average statistics, they should not be used quantitatively and are only shown for guidance in selecting periods to analyze further. The latent heat flux and Bowen ratio, which normally are shown on this series of plots from ASTER, are not present due to the above-mentioned poor performance of the Krypton hygrometer.

Other plots

Steven Oncley<>
Last modified: Thu Sep 12 15:33:36 1996