Poster Presentation

Data Processing and Curve Fitting of Counts Data

  • Liam Taylor (University of Minnesota, Morris)
  • Gordon McIntosh (University of Minnesota, Morris)
  • Alaina Swanson (University of Minnesota, Morris)
  • Erick Paul Agrimson (St. Catherine University)
  • Kaye Smith (St. Catherine University)


Various data analysis methods were explored to more accurately and consistently determine the Regener-Pfotzer (RP) maxima for high altitude cosmic radiation. The radiation has been measured during 15 balloon flights using Geiger counters with five second accumulation times. Of the 15 flights, 10 of them included omni-directional counts data, and 8 of them included vertical coincidence counts data. Count data from altitudes greater than 10 km were analyzed to determine the maxima. The data analysis methods used were moving average filtering and summation of Geiger counts into one minute intervals. Moving average filtering did not give reliable results, so the summation method was chosen. Once the data were summed, several different curves were fit to determine where the RP maximum occurred. The curves tested include second and third order polynomials as well as cubic spline interpolation of the data averaged over 1 km intervals. Second order polynomial fitting did not fit the data well. Third order polynomials and cubic splines gave better results. Third order polynomial fitting was chosen due to its ease of use and the similarity of the results given by the cubic spline interpolation (within 1%). The omni-directional RP maxima occurred at an average altitude of 21.8 km ± 1.7 km, while the vertical coincidence RP maxima occurred at an average altitude of 18.5 km ± 1.1 km. In addition, the vertical coincidence RP maximum occurred at 65 hPa ± 9 hPa, while the omni-directional coincidence RP maximum occurred at 38 hPa ± 13 hPa.

How to Cite:

Taylor, L., McIntosh, G., Swanson, A., Agrimson, E. P. & Smith, K., (2017) “Data Processing and Curve Fitting of Counts Data”, Academic High Altitude Conference 2017(1). doi:



Published on
27 Oct 2017