Notes on Local Meteorological Data (DRAFT)
Meteorological measurements have been made at several stations around Los Alamos as part of the Laboratory's environmental monitoring and emergency preparedness operations. Some of these stations are now inactive, but the data from both the active and inactive stations are archived and available for analysis. You can obtain either raw data, which are values reported by the dataloggers on a 15-minute or 24-hour interval, or you can obtain input files for running some of the frequently used atmospheric dispersion and dose assessment models.
Raw Data
Links to an overview of the time and geographical coverage of the data and brief station descriptions are given below. Links to detailed station descriptions, which include important data quality information, are provided from several points in this documentation.
Before using the data, we urge you to consult the data quality discussions in the detailed station descriptions. Although considerable effort has been made to ensure that these data are of high quality, problems remain. In applications where meteorological data may play a critical role, we recommend that you discuss your application with a meteorologist.
Where data are known to be bad (or they are missing), an asterisk (*) is inserted into the record. In general, data for the last one to two weeks have not been thoroughly reviewed, so caution is advised. The part of the data record that has been reviewed is noted at the top of the Raw Data Request Form for each station.
You can begin your request for raw data from here.
Model Input Files
Using the raw data, you can generate meteorological input files for some of the frequently used atmospheric dispersion and dose assessment models. All of these data sets are based on wind measurements made at 11 m above the ground. The method of determining atmospheric stability is common to all the data sets and is described (below) in the section discussing the CAP88 and GENII input files.
Model Input for ISC3 and MACCS
The ISC3 model (short-term version) and the MACCS model use hourly averages of wind speed and direction over a year. The user specifies the station and the year of interest. In these files, wind directions have been converted to the required "flow vector," which gives the bearing of the direction TOWARD which the wind blows. Data fields in the ISC3 files for which the data are bad or missing are filled with 9s. When you generate a MACCS input file, an asterisk (*) is inserted into the data fields for which data are bad or missing. If you generate an ISC3 input file for a leap year, you may have to remove 24 records to get the code to run. We automatically remove data for 2/29 from MACCS input files.
The ISC3 and MACCS models require hourly values of the mixing depth. Since there are no local climatological values for this quantity, we have used values that G.C. Holzworth estimated for Albuquerque (see Table B-1 in "Mixing Heights, Wind Speeds, and Potential for Urban Air Pollution throughout the Contiguous United States," EPA Publication No. AP-101, January 1972). In the input files you generate, the data fields for the "rural" and "urban" mixing depths are filled with seasonal average values of the morning mixing depth (for Albuquerque), and those values are repeated for every hour during the season.
You can begin your request for ISC3 or MACCS data from here.
Model Input for CAP88 and GENII
CAP88 and GENII require joint-frequency distributions of wind direction and wind speed organized by stability category. The user specifies the station and the time period of interest. These statistical data summaries are often referred to as STAR data, for STability ARray data. In files for CAP88 and GENII we use the standard meteorological convention for wind direction, which is to give the bearing (0° to 360° ) of the direction FROM which the wind blows.
There are three differences between the way we generate these data summaries and the way they are generated from National Weather Service data. First, we use 15-min data instead of hourly data. Second, we do not distribute occurrences of "calms" (speed less than 0.45 m/s) within each stability class in proportion to the sum of the frequencies of winds in the first two wind speed classes. The wind direction class we assign to a calm is based on the wind direction reported by the datalogger during the calm; we feel this is just as valid as other methods used to estimate wind direction during a calm. Third, the stability category is based on a measurement of turbulence instead of the original Pasquill method.
Stability categories are based on sigma phi, which is the standard deviation of fluctuations of the wind vector in the vertical plane. Sigma phi is determined from the standard deviation of fluctuations in the vertical velocity (sdw) divided by the mean wind speed (spd) and converted from radians to degrees. Then the stability category is determined by comparing sigma phi against ranges of values recommended by the EPA (see "On-site Meteorological Program Guidance for Regulatory Modeling Applications," EPA Publication No. EPA-450/4-87-013, 1987; or, see NTIS PB 87-227542).
As recommended by the EPA, we have adjusted the sigma phi ranges to reflect a local roughness of 0.4 m. In addition, all occurrences of unstable categories during the nighttime, stable categories during the daytime, daytime wind speeds greater than 6 m/s, and nighttime wind speeds greater than 5 m/s are classified as neutral. We define "daytime" as any period of time for which the average value of the incoming short-wave irradiance (swdn) exceeds 5 watts/m^2.
Spatial and temporal averaging shows that the unstable categories (A, B, and C) occur 35% of the time, the neutral category (D) occurs 39%, and stable categories (E and F) occur 27% of the time at Los Alamos. The frequency of occurrence of any given stability category can vary by 10 percentage points from year to year at a single station, or by as much as 20 percentage points from station to station.
Users of CAP88 have to enter a "lid height." We suggest using a value of 1600 m, which is the average of the annual mean morning and afternoon mixing depths from Holzworth.
You can begin your request for CAP88 or GENII data from here.
Copyright © U C1997 - Disclaimer