Data
General
The sensors used in CACHE94 are listed in Table 3 and shown schematically in Figure 5 and Figure 6. Each sensor is associated with a system identification number (sid) that is composed of an ADAM name and channel number (e.g. cosmos 202). Channel numbers from 100 - 199 indicate an analog input channel and channel numbers from 200 - 299 indicate a serial input channel. Each sensor has one, and sometimes more, output variables associated with it.The variable names each have a data identification number (did) used to identify the archived data. (e.g. u.prop.5m and v.prop.5m). All data acquired during the deployment is archived. The application of the appropriate calibration and delay functions convert this raw data to engineering or scientific units. This processed data is available as 5 minute block averages of the means, variances and covariances obtained from the initial processing of the variables measured. These block averages are termed ``covars''. The format of the covars is relatively easy to decipher. Thus conc.co2.10m is the 5 minute mean of the concentration of CO2 measured using the fast response CO2 sensor at the height of 10 meters, while w:t.walk.10m is the 5 minute mean of the covariance of the vertical air motion measured by the Solent sonic anemometer and the temperature measured by the AIR fast response temperature sensor on the walk-up tower at the height of 10 meters. At the end of each day the raw data is transferred to an Exabyte magnetic tape. To reduce the risk of data loss the data for each day is stored on a separate tape. The raw data is then erased from the on-line hard disk. All covars and other derived products are available as NetCDF.
The available covars are presented in the available plots section of this report.
In April 1995 the five minute products will be recalculated using the currently up-to-date calibration parameters.
Wind
After the array was established, the orientations of the booms of the prop-vanes and the sonic anemometers were determined using a theodolite set to true north by means of solar sighting and an ephemeris. For CACHE94 this procedure proved to be somewhat difficult because of the restricted nature of the ridge-top site.
After optical alignment of the booms the prop-vanes are ``set'' by manually aligning the vane to the boom using the temporally fixed alignment jig. The prop-vanes are integrated sensors which, after the absolute alignment has been entered, yield wind components in a geographic coordinate system; where u is wind from the west and v is wind from the south. The propvanes are individually calibrated for wind speed response in the SSSF wind tunnel prior to the deployment and yield wind speed directly as a serial output.
The transducer alignment of the sonic anemometers do not accurately correspond to that of the boom. Hence for the sonic anemometers this bearing was used as a nominal value and need to be corrected to the true value by comparing the sonic anemometer and the prop-vane data. The ATI sonic anemometers output the three orthogonal wind components and the derived speed of sound temperature. The Solent sonic anemometer measures three non-orthogonal wind components from which the three orthogonal wind components are calculated. In both cases, the wind components are in an instrument-oriented coordinate system, where u is the wind along the instrument boom towards the tower and v is the wind from right to left as you face in the positive u direction. The sonic anemometer data has been despiked and, for the daily plots, a streamline rotation has been performed.
Psychrometry
The psychrometers are integrated sensors which yield relative humidity and temperature directly as output. From these measured values other moisture parameters can be calculated.The psychrometers are calibrated in the SSSF calibration lab prior to the field program.
Hydrosol data
The cloud droplet information was acquired using the Forward Scattering Spectrometer Probe and the two Particle Volume Measurement devices.
FSSP data
The FSSP data was logged directly to an OSU IBM PC, and fed digitally to ASTER via the Marigold ADAM. Data include the specification of FSSP range (which determined size intervals for droplet counting), the counts for each of fifteen size intervals, the time interval over which the counts accumulated (nominally 0.1 seconds), and three diagnostic variables for determination of data validity. All of these data were logged at 10Hz.
PVM data
The PVM's output data on the total liquid water content at 10 Hz. The PVMs respond to the volume of cloud-water in the sensor path. Each PVM response must be corrected by a time-varying clear-air signal, which was recorded periodically throughout the experiment.
Carbon dioxide
The carbon dioxide fluxes showed great promise to explain the CO2 exchange of the canopy on the slope around Cheeka Peak. However care must be taken in interpreting the 5 minute covar data since flowrate variations and transients due to zero switching introduce spurious flux artifact values. The data must be detrended and filtered before the final values are to be accepted.
Ozone
The ozone mixing ratio provided substantial information as to the origin of the air arriving at Cheeka Peak. Basically there were three important sources: oceanic air with low ~ 20 ppbv O3, continental air containing the photolyzed fume from the large scale forest fires which burnt throughout Washington state, and pollution from the Seattle urban area. The situation was complicated by the fact that continental and urban air could move offshore and then return admixed with true oceanic air.
Radiation
The radiation stand bearing the radiation sensors was positioned on the slope south of the Cheeka Peak site. The 3-meter-long horizontal beam of this radiation stand was 1 meter above the canopy, 3 meters above the ground and oriented nominally east-west. The level-adjustable assembly bearing the radiation sensors was mounted on the beam and leveled to within 15'. Five radiation sensors were deployed: a four component radiation array monitoring incoming and outgoing short and long wave radiation with Eppley radiometers, and a REBS Q6 net radiometer. The two long-wave radiometers and the net radiometer output analog voltages. The short-wave radiometers gave similar signals together with thermistor output defining the temperature of the instruments' case and dome. This enabled a correction to be made to compensate for the direct sunlight heating of the dome of the shortwave sensor.
It is important to appreciate that the short-wave radiometers and the net radiometer view a hemisphere with a cosine weighting factor. For the incoming radiation this presents no difficulty if the sensors are horizontally level. For the outgoing radiation it implies that the effective viewing area of the sensors at 3 meters height is only a few tens of square meters. This constraint puts a premium upon the representativeness of the surface under the radiation stand.
The Everest surface temperature sensor has a collimated 30° cone and with the mounting a meter above the surface facing down at 45° viewed a patch of the surface less than ten square meters in area. The sensor yields an output directly in degrees centigrade by using an assumed infrared emissivity of 0.95.
Soil parameters
Two locations near the radiation stand were instrumented with soil temperature and heat flux sensors. These were in areas of duff and of dead leaves. At each location four soil temperature probes were inserted at 1 cm, 3cm, 5 cm, and 7 cm depth, and a soil heat flux plate was buried at 8 cm. The output of the four soil temperature probes were connected to the data system via a self timing multiplexer. The multiplexer generated an indicator signal to enable the correct assignment of the different outputs. The temperature profile of the soil and the average temperature of the top 8 cm was determined. From the soil heat flux at 8 cm depth together with the heat stored in the top 8 cm of soil the surface heat flux was calculated. As there was an extensive canopy and as the below canopy was so complex this surface heat flux value is difficult to interpret.
Daily plots
A selection of covar plots were produced daily to provide a review of the operations. These are termed ``Daily plots''.
The daily weather plots presented in the appendix of this report are (a) temperature, humidity, pressure and rainfall, (b) wind speed and direction, (c) net radiation, sensible heat flux, latent heat flux, and surface heat flux, (d) z/L, u* and the Bowen ratio, (e) FSSP and PVM data, mean scaled standard deviations, and deposition velocities.
The plotted data are 5 minute covar plots, except for the turbulent flux data, sensible and latent heat fluxes, z/L, u* and the Bowen ratio, which are 20 minute averages.
The daily plots are presented in the appendix of this report.
Logbook
An electronic logbook was maintained during the field program. A total of 382 entries were made during the CACHE94 deployment in 17 different categories:
A2D/DATEL/FILTER ADAMS/NETWORK CO2/O3/H2O2
COMMENT FASTT FSSP/PVM/CPC
KRYPTON KRYPTON/IR LOG
OPS PROP PSYC
RAD/SOIL/RAIN SOFTWARE SONIC
STATUS WEATHER
The electronic logbook is available to complement this report.
Last Modified: 12:00pm MDT, July 4, 1996