Spreadsheet for automatic processing of water quality data : 2010 update, calculation of percentiles and tests for seasonality

SKU:
SR_2010-042-pdf
$0.00
(Inc. GST)
$0.00
(Ex. GST)
Write a Review

Daughney, C.J. 2010 Spreadsheet for automatic processing of water quality data : 2010 update, calculation of percentiles and tests for seasonality Lower Hutt, N.Z.: GNS Science. GNS Science report 2010/42 16 p.

Abstract: In 2005, a spreadsheet was developed for automatic processing of water quality data obtained through the National Groundwater Monitoring Programme. This spreadsheet was designed to automatically process water quality results for many different analytes in each of many different samples collected from many different sites. The spreadsheet was developed because software with comparable functionality was not commercially available. Excel was selected to ensure portability. Two years later, the spreadsheet was updated with additional functionality and renamed “2007 NGMP Calculator”. This report provides a description the most recent version of the spreadsheet, which is named “2007 NGMP Calculator Vers NS-4”. The latest version of the spreadsheet is almost identical to Version NS-3, hence the decision to retain “2007” in the file name. However, there are two new capabilities in Version NS-4: The minimum, maximum and the 5th, 25th, 75th and 95th percentiles are now included in the reported distributional parameters, along with the median, average, median absolute deviation and standard deviation. As in the previous version of the spreadsheet, these distributional parameters are calculated on a per-site and per-analyte basis using the log-probability regression method. - Seasonality is assessed with the Kruskal-Wallis test. As in the previous version of the spreadsheet, the user can opt to divide the year into up to 12 different seasons, and the user can control the Julian day on which the first season starts. The same season definitions are used for the Mann-Kendal trend test and for the Kruskal-Wallis test for seasonality. Both tests are also performed with using the same user-specified confidence level. (auth/DG)