COMPARISON OF LOAD ESTIMATION FROM GRAB SAMPLES AND CONTINUOUS FLOW PROPORTIONAL SAMPLING - SUPPLEMENTARY REPORT TO FR0005
Report No FR0092
R P LACEY
The objective of this work is to update the previous report (FR0005) on the study of methods of estimation of pollutant loads, with particular regard to progress that has been made with the statistical theory.
At the time of writing FR0005, satisfactory methods for quantifying the bias and precision of one of the commonly used procedures for calculating loads (the method based on the flow-weighted mean concentration) were not available. More satisfactory approaches have now been found.
This report gives better methods than were previously available for quantifying the bias and precision associated with a commonly used procedure for calculating loads.
Recalculation of the results of the study at Kingston upon Thames, to incorporate improvements in the method, has confirmed the conclusions of the earlier WRc report (FR0005) particularly the main finding that, compared with continuous monitoring, estimates of annual load based on weekly grab sampling (at Kingston upon Thames) would be accurate only to within ±25%, while estimates based on monthly grab samples could easily err by as much as ±50%.
This report does not repeat the full description of the Kingston work, but concentrates on technical details of the method of calculation. Table 9 and Figures 2-11 in the earlier report (FR0005) should be replaced by the corresponding table and figures in the present report.
Copies of the Report are available from FWR, price £15.00 less 20% to FWR Members