Simple Moving Average Applied to "ISO Method" CPAM Concentration Estimates: An Unexpected Result.

Area of Science:

• Environmental monitoring
• Radiological science
• Data analysis

Background:

• Continuous Particulate Air Monitors (CPAMs) are crucial for real-time airborne radioactivity detection.
• Raw CPAM data often exhibits high variance, necessitating effective filtering techniques.
• Existing filtering methods may introduce unacceptable lag or insufficient variance reduction.

Purpose of the Study:

• To investigate the effectiveness of a simple three-point moving average filter for CPAM data.
• To analyze the impact of this filtering on variance reduction and data lag.
• To understand the underlying reasons for observed variance reduction.

Main Methods:

• Applied a three-point moving average filter to CPAM airborne radioactivity concentration estimates.
• Analyzed the autocorrelation structure of the processed data.
• Derived the variance reduction factor analytically.

Main Results:

• Observed a surprisingly large amount of variance reduction using the three-point moving average filter.
• The filtering method demonstrated relatively little lag compared to the achieved variance reduction.
• The autocorrelation structure, arising from differencing integrated counts, was identified as the key factor.

Conclusions:

• A simple three-point moving average is a highly effective filter for CPAM data.
• This filtering technique balances significant variance reduction with minimal data lag.
• The analytical derivation confirms the effectiveness attributed to the data's inherent autocorrelation.