Should we routinely be expressing uncertainty in our measurements?

In an excellent paper reviewing hydraulic head (water level) measurements, Vincent Post & Jos von Asmuth (2013[i]) argue that when measuring water levels

In-Situ's new water level meter 100 and 200 used for accurate groundwater level measurement

“it should become standard practice in hydrogeology to provide quantitative estimates of the measurement error.”

The basic tool of any Hydrogeologist is the electric water level meter, erroneously, but commonly, referred to as a “dipper” or “dip tape”, though, as I was once advised in an email from a very eminent Hydrogeologist, “dippers are a wading bird or a surveying instrument for measuring angles, not water levels”!

Whatever these devices are called, we all use and trust them to reliably and accurately record the depth to water level.

But is your trust misplaced? Have you ever thought about the margin of uncertainty associated with this most basic of measurement?

Most water level meters are supplied with tapes marked in intervals of at least centimetres, and increasingly, millimetres.  This means that it is possible to record a measurement to 3 decimal places (e.g. 6.345 m below top of borehole lining). But what is the error margin? Is it ± 0.5 mm (which it needs to be if readings are quoted to 3 decimal places) or is it something greater than this?

Measurement Standards

There are standards specified for measuring water levels in boreholes using tapes. Other standards exist for manufacturing of steel measuring tapes. Examples of both from
Europe[ii],[iii] and the USA[iv],[v] are given in Table 1.  It is important to differentiate between the two types of standard.  For example, the BS ISO standard is based on the repeatability of a measurement using the same tape. Manufacturing standards are based on a scientifically based reference length which recognise that all materials will stretch in proportion to their length when a weighted probe is attached (remember Hook’s Law?).

A measurement taken at 30 m depth using an EU Class II tape should be accurate to within 1 cm (0.63 cm), but at 100 m the accuracy should be within 2 cm. Consequently, quoting your water level measurements to 3 decimal places is probably not appropriate unless you use an EU Class I tape at less than 5m depth (or an ASME / EU Class I tape at less than 10 m depth).

Table 1: Summary of accuracy standards

standards used by manufacturers of water level meters

Standards used by Manufacturers of Water Level Meters

Most manufacturers do not specify a standard of accuracy. It is very difficult to manufacture tapes constructed of plastic to reliably and consistently record water levels within standards. Some better-quality designs use multiple steel conductor cable or Kevlar to reinforce the tape to achieve EU Class II standards (0.02%).  Steel tapes can be manufactured more easily to EU Class II standards and some, including In-Situ are aspiring to the new ASME (EU Class I) standard (0.01%).

Seeing is Believing

But how can you practically check the accuracy claimed by manufacturers?

It is very easy and inexpensive to purchase an EU Class I or Class II standard steel tape on-line. These tapes all have the class symbol marked at the start of the tape (see pic).

 groundwater level measurement

You can then use the steel tape as a standard to check the length of your water level meters.

level tape for groundwater depth measurementHere’s an example from a water level tape I came across that was being used to validate water level measurements from a data logger.

The photo shows the plastic dip tape at 7 m length lined up against an EU Class II steel tape. The zero points on both tapes are matched. The 7 m mark on the plastic tape lines up at 6.980 m on the steel tape. This is a margin of error of 2.0 cm over 7 m.

At 30 m depth, the dip tape would be overstating the water level measurement by 8.6 cm and at 100 m by 28.6 cm! For this application, this is an unacceptably high measurement error of 0.29%.

In Summary

If you are using water level tapes in situations where accuracy is not so important, then perhaps any tape will do. But surely we should all be aspiring to be more professional in the way we report measurements and clearly state the margin of error we are working to?  This is particularly important when using water level meters to validate measurements from other instruments (e.g. data loggers) or where there is a need to measure to mm accuracy (e.g. wetlands).

So, be wary with dip tapes. Validate their accuracy using a Class I or Class II steel tape. And do consider taking on board the recommendation by Post and von Asmuth to state measurement accuracy when reporting water level data.

©Peter Dumble 2016   Peter.Dumble@pdhydrogeology.com 

 

In-Situ Europe - The NEW water level meter 200 uses high-tensile steel tape for maximum accuracy.

 

[i] Post V.E.A & von Asmuth J.R, 2013. Review: Hydraulic head measurements – new technologies, classic pitfalls. Hydrogeology Journal 21, 737-750.

[ii] BS ISO 21413:2005. Manual Methods for the Measurement of a Groundwater Level in a Well”.

[iii] EU Directive 2014/32/EU on the harmonisation of the laws of the Member States relating to the making available on the market of measuring instruments.

[iv] ASTM D4750-87(2001), Standard Test Method for Determining Subsurface Liquid Levels in a Borehole or Monitoring Well (Observation Well) (Withdrawn 2010), ASTM International, West Conshohocken, PA, 1987, www.astm.org

[v] ASME B89.1.7 – 2009 (R2014). Performance Standard for Steel Measuring Tapes.