Data quality and instrument calibration are very important aspects of the Department of Energy's Atmospheric Radiation Measurement (ARM) program. Our initial goal is to improve the quality of the data collected at the ARM field sites. Our long-term goal is to improve the level of instrument accuracy. The two projects in which I have devoted my time to this summer are concerned with achieving each of these goals.
To help improve data quality, it was necessary to find a computer program that would look at raw data being received daily via satellite from ARM Tropical Western Pacific field sites. We chose a Matlab program for its simple yet sufficient method of looking for time gaps in instrument data collection. By looking for time gaps in the data, one can compare with weather records and daily site logs to determine the cause of the gap. This understanding could lead to the reduction of gaps in the data, thus producing more consistent and usable data.
I furthered the achievement of the second goal by working with a radiation transfer code to improve the accuracy of the instrument calibration and reliability of the Global Climate Models (GCM’s) being used by the Tropical Western Pacific (TWP) branch of the ARM program. Dave I, II, and IV are radiation transfer codes Bill Porch used in his early days at Lawrence Livermore. These Fortran programs, if sufficiently modified to match a tropical atmosphere would provide us with a calculated total optical thickness. Optical thickness is a measure of how much or how little sunlight is able to cross the atmosphere.
Optical thickness is dependent upon the wavelength, the size of the particles present and the properties of the particles. For example, water vapor is responsible for major radiation absorption at wavelengths of 0.55 to 2.5 m m, whereas oxygen absorbs radiation at wavelengths of 0.12 to 0.20 m m. The Dave Radiation Transfer Code inputs temperature, pressure, ozone, water vapor, and carbon dioxide count specific to tropical latitudes. Then, by varying wavelength, aerosol size and zenith angles, optical thickness for each layer can be computed and summed for a total optical thickness of the atmosphere. This calculated thickness computed by the Dave code is compared with instrument calculations to check the accuracy of the instruments and models. In this case, the instrument that we are comparing data with is the Multifilter Rotating Shadowband Radiometer (MFRSR). It is important to note that when trying to correct for inaccuracies in the instrument or model that one does not use instrument data as a reference for the model, or vice versa, unless absolute accuracy is proven for one or the other. Instead, the best method is to use either the model or the instrument to determine what the other lacks. Therefore, by comparison and complete instrument logs, a conclusion can be reached on where to make the necessary adjustments.
Presentation: Data Quality Improvement Studies at ARM TWP
The Global Change Education Program is funded by the