At a certain water level in the marsh, Well A1 conforms to this "Wet Meadow" Phase. When a general rise in water level occurs following a rainfall event, the marsh surface itself rises only a small amount at A1. Indeed, changes in marsh surface elevation may reflect wetting with expansion and drying with contraction of the peaty layer. It is not known how the boundary between the peaty-muck and deep muck behaves to these small changes in water level. However, once inputs to the marsh cause the water to exceed about 3 feet elevation, the vegetation mat in the vicinity of Well A1 separates from the deeper muck and floats. We have confirmed that water intercedes between the layers. Under such high water conditions, Station A1 behaves as a floating mat (for example, after mid-January in 2003 in the graph above).

Note that following the rainfall on October 15, the soil saturation zone rose to a point where water appeared briefly on the surface (this was observed). Afterwards, water level dropped, although the marsh surface itself declines (maybe contracts) less quickly. Water level changes are magnified in comparison with water level changes in the nearby ponds(fine dashed line), because the well measurements are made in an aquifer: inputs and withdrawals occur in voids in the surrounding material so the same amount of input will raise the aquifer level more than the nearby pond level. During this phase it seems evident that the PVC pipe is embedded in bottom muck.

Another curious phenomenon occurred here in early May 2003. Although water level in the marsh was dropping steadily, movement of the mat appeared to level off, then actually rise. This behavior might have been expected as the mat bottomed out on the muck layer. However, from the graph, it appears the mat should have sunk another 6 inches before contacting the muck layer. It is possible, that this did occur; the intervening space between the muck and the bottom of the mat being a thinner muck easily penetrated by the drop line and not just water. Core samples made on April 12 suggest that the water layer was only a few inches thick at that time.

Elevation measurements surveyed in with a leveling rod are initially variable, then settle down to a relatively constant value. This effect may represent initial tamping down of the upper surface of the clay with the rod (that is, an effect of the measuring process). Overlaying the clay layer is a layer of "muck" through which the rod penetrates but the drop line does not. This layer also appears "unstable" initially (where compared with rod measurements), but then increases in thickness from early February through April, starting about when the mat separated from the bottom.

The results of cores (A1-1 and A1-2) made with a 4-foot long 0.5-inch diameter glass tube are shown on the right. The core locations are within 3 m of the well. All field measurements were related to water level at the time. The top of the mat is set at the same elevation as measured at the well, for the reason that the mat is depressed an unknown amount while taking measurements. The thickness of the floating mat (upper two, dark layers) was measured separately after a hole was made through it for insertion of the glass tube. This step was necessary in order to obtain an accurate core (minimal compression) of the liquid layers beneath. The water layer is shown in white (around 2.0 elevation in A1-1), with slurry increasing in density down to a dense, clay layer (bottom layer in profile) that was only penetrated far enough to hold the core in the tube for measurements. A "V" marks the depth of penetration of the weighted drop line. Note that this depth above the clay is approximately the same as the long-term average soft muck thickness of 0.41 ft.

The dry season change in water level represented by measurements made between August 15 and October 11, 2002 (dry season WL decline) was -0.009 ft/day (-2.7 mm/day). This change closely matches that of the marsh water level (-2.8 mm/day) for the same time period, but is nearly double the -1.4 mm/day change observed at Well B1 where the surface mat is floating.