Errors in groundwater level measurement

As in any measurement activity, errors can occur in the data.

Inaccuracies in measurement

Imprecise data, including imprecision in spatial coordinates, ill-defined vertical datums, and imprecise dates of measurement introduces significant inaccuracies in measurements of water level.

Although modern measuring equipment and careful surveying can give water level to better than 0.01 metre precision, historical data will generally be at least one or two orders of magnitude less precise, and often worse.

As in some aquifers groundwater gradients are of the order of one metre/ kilometre this imprecision may be significant for detailed survey.

Multiple aquifers

Where multiple aquifers are present there may be significant differences in water level between aquifers, especially if there are perched water tables or confined aquifers.

All too often borehole completion details are absent, making it unclear which aquifers are measured.

Where boreholes are left open to multiple aquifers composite levels (unrepresentative of either aquifer) may be measured.

Karstic aquifers

In karstic aquifers and other aquifers where fissure flow dominate there may be no defined water table in a regional sense, but rather each fracture system may respond independently, leading to significant variations in level over short horizontal or vertical distances.


In clays a water table may be present, but not be a useful concept, if the low permeability of the clay prevents the draining of surface water.

Seasonal variations

Seasonal variations in level range from a few centimetres in some confined aquifers to several tens of metres in the interfluves of aquifers such as the chalk. Inter-annual variations show similar fluctuations in response to climatic variations.


Levels may be artificially depressed by pumping. Where regional depressions occur it is the current depressed level that is normally of interest, although reversion to the natural 'unpumped' level is likely if pumping ceases — frequently historical data has been gathered during periods when pumping was intense, and less data may be available after pumping stops.

Where unrecognised localised depression of the water table occur through pumping they may distort interpolations of regional levels that include measurements made in nearby boreholes.

More rarely levels may be higher in observation boreholes due to localised recharge leading to groundwater mounding. Measurements made during or immediately after the drilling of a borehole may be affected by the injection or extraction of water as part of the drilling process.


Contact Andrew McKenzie for further information