Electronic Data Collection and Analysis System
Yin Foo, Des; Foley, Margaret
E-Publications; PublicationNT; E-Books; Report ; 39/1992
1992-08-01
Made available via the Publications (Legal Deposit) Act 2004 (NT).
Date:1992-08
English
Power and Water Authority
Darwin
Report ; 39/1992
application/pdf.
Check within Publication or with content Publisher.
https://hdl.handle.net/10070/229024
https://hdl.handle.net/10070/673447
Technical Report WRD92039 Viewed at 15:07:17 on 29/07/2010 Page 176 of 192. D5.1 NOTES ON CALIBRATING LOGGED DRAWDOWNS BY MANUAL READINGS "DATMAN" can perform a Linear recalibration of a data set by another ("T:rue") set. This converts an original data set {O}, to ReCalibrated data {R}, using a "true" set {T}. R = s*O + 0 where s is the slope and 0 the offset of the Best straight Line through a T.v. 0 graph. This type of Data manipulation is valid only where there is : _ a Ivide range of (T) values compared that of the complete {OJ set; (A) an (almost) unique value of T for given (OJ values.(B) This lire-calibration" is useless if e.g. Manual readings are only available for 10 to 10.2m DD, when the logged DD rangE;s from 0 to 12m ( because any introduced error in the slope; will be ;nagnified by 12/0.2 (=60) ill this example), or if at different times different Manual readings are recorded at the same 10<Jger reading. "DATJ'IlAN" shows a {T} .v. {OJ graph and shows the slope,offset iind correlation co-efficent calculated from a Best straight .Line through the points. In general the unsuitability of the data to recalibration as regards uniqueness will be visible as scatter on the graph and a low value of "r squared" (which is often above 0.99). Note that; if {T} and {O} do not share the same times, currently (Dec '91) automatically interpolates values of {Or for {T} times f which can lead to errors either pre- or post- dates {OJ. If meaningless data must be deleted from {T}. particularly if {T} this happens, the