8.1 Well and Piezometer Installation and Use

Ideally, wells being used to estimate the capacity of an aquifer to deliver water should fully penetrate the aquifer. As many peat deposits are on the order of a few meters thick, this is generally not a problem. Wells in peatlands are normally used to determine the static water table (not aquifer yield), and so a fully penetrating well is not necessary. However, a well should extend past the acrotelm because, by definition, the water table can be below the bottom of the acrotelm. Practically, the well should be deeper to provide stability, ideally anchored in the mineral substrate. Many practitioners will use a 1.5 m length of PVC (polyvinyl chloride) pipe to allow at least 1 m penetration (and 0.5 m stick-up), which is sufficient in bogs and fens. In swamps, the water table can be deeper and so the length should be chosen accordingly.

Piezometers by their nature are designed to monitor hydraulic head at a given depth interval. The length of the slotted intake is up to the user; however, an intake length of 20 cm is often practical for depths within the catotelm (usually < 4 m thick). Even shorter slot lengths (10 cm or less) may be warranted, particularly in the upper layer where hydraulic conductivity can change profoundly with depth. We note that the chosen screen length is a function of the research goals, where smaller screen lengths increase the spatial accuracy of the measurement at the expense of increasing the number of measurement points required to mathematically obtain spatial averages that are otherwise generated by longer screen lengths. While very short screen lengths—or in some cases tubes open only at the bottom (for example to measure only hydraulic head)—can increase spatial discretization, time lag (time required for piezometers to equilibrate) may be excessive and instantaneous measurements may not reflect the current state of the system, particularly where hydraulic conductivity is low. With piezometers, the slotted interval should be centered with respect to the desired depth it is intended to represent (e.g., a 20 cm slotted intake placed 40 to 60 cm bgs represents average head or hydraulic conductivity at the 50 cm depth). Then, as long as the interval is below the water table, the hydraulic head and the head recovery during tests reflects the same part of the peat deposit, with the caveats discussed below.

Measuring the water level within a well or piezometer is often more problematic in peat than it is in mineral soils. This is because the water level will rise as the observer approaches the pipe (the peat is highly compressible) and electronic devices often do not work in bogs due to the very low ionic concentrations in the water. The high compressibility poses a related problem: the pipe elevation (and stick-up) can change between measurements. As the water table declines over the season, the peat surface can decline. For short pipes, their elevation may decrease accordingly. For longer installations, the pipes may be stable, but when the surface elevation decreases the stick-up increases. The seasonal surface fluctuations are generally reversible. It is good practice to place a ring or marker near the base of the pipe and manually measure stick-up each time a head measurement is made. Also, due to the uneven nature of the peatland surface, the stick-up always needs to be measured on the same side of the pipe. If the pipe length is sufficient (i.e., the base is stable), then the change in surface elevation can be deduced as discussed by Price and Schlotzhauer (1999).

Shallow pipes are also susceptible to frost heave. Anchoring the tube to a metal rod pounded into the mineral substrate can reduce their movement. Nevertheless, it is good practice to survey the pipe-tops each season if multiyear measurements are being made or if vertical or horizontal hydraulic gradients are being assessed. Use of a logging pressure transducer suspended from the well can reduce errors caused by the weight of an approaching observer. However, it is still subject to well/piezometer stability.

Surface level can also be measured with an ultrasonic sensor but is only necessary when surface elevation changes are large and part of the monitoring objective (Fritz et al., 2008). For water level measurement in a bog, where no electrical current passes through the weak ionic solution, a blowstick can be used. A blowstick is a length of flexible tube inserted into a slightly larger diameter graduated rigid tube; or a flexible graduated tube that allows the user to hear bubbling caused by blowing through the tube as it is inserted. The accuracy of these devices is about ± 5 mm, although better accuracy can be achieved by consistent use by the same operator.

Inaccuracy can be problematic in peatland studies if the intention is to measure vertical hydraulic gradients, which are commonly of a similar magnitude as the error of the measuring device, notwithstanding the errors caused by subsidence/compression and surveying. Calculating the vertical gradient between widely spaced piezometers (≥ 1 m) reduces the effect of the measurement error, proportionally. Vertical gradients can also be determined by measuring the hydraulic head in a piezometer relative to the local water table, measured in a well or shallow pit in the acrotelm. To reduce the errors associated with the compressible nature of peat, local boardwalks (e.g., 15 to 30 cm wide horizontal boards affixed to vertical wood pilings driven down to the mineral soil beneath the peat) are recommended so practitioners are not standing on the peatland surface.