Impacts of Pleistocene glacial loading on abnormal pore‐water pressure in the eastern Michigan Basin

Abstract

The hydromechanical effects of Pleistocene glacial loading on the Michigan Basin are assessed using numerical analysis based on coupled stress and pore-water pressure. The two-dimensional model domain included the Basin cross section and extended 10 km into the Precambrian. In the analysis, we considered the effects of the number of glacial loading cycles, the presence and connectedness of a deep Cambrian aquifer, the direction of glacial advance, the effect of a wet versus dry glacier/soil interface, topographic effects, density-driven flow effects, and lithosphere flexure on the development of anomalous pressures. The modeling results were compared with data collected from highly instrumented wells completed in the eastern margin of the Basin. The present-day results define regions of significant underpressure in the upper Ordovician and lower Silurian formations characterized by very low hydraulic conductivity and regions of overpressure where hydraulic conductivity is higher. To achieve the degree of underpressure observed in the instrumented wells using the model, a specific loading cycle applied over 100 000 years was repeated four times. As the number of loading cycles increased, the Paleozoic formations reached a state where the underpressures remain constant. Our results also illustrate the difference in poroelastic modeling between the application of mechanical loads on the land surface and the application of an equivalent hydraulic head, where the latter developed overpressures rather than the observed underpressures. The modeling also shows that the overpressures observed in the Cambrian formations are most likely to be the result of density-driven flow. Finally, the simulations show that the effects of lithosphere flexure in the hydromechanical models results in the development of lateral stresses that generate overpressures rather than underpressures in the southern half of the domain. As there are no suitable observation points, these results remain unconfirmed, and further study is warranted.

Assessing Alternative Scenarios for the Cause of Underpressures in the Ordovician Sediments along the Eastern Flank of the Michigan Basin

“…The Cambrian units have a very high horizontal hydraulic conductivity of 3.0 × 10 −6 m/s [5,61] which can dissipate transient hydraulic perturbations efficiently. Similar findings were reported by Khader and Novakowski [30] using a twodimensional model. The analyses of Sykes et al [37] included one simulation that allowed the Cambrian aquifer to drain and an underpressure in the middle Ordovician formed by the end of one glacial cycle; however, additional glacial cycles were not simulated.…”

“…Following the work of Sykes et al [37], the twodimensional model studies of Nasir et al [29] and Khader and Novakowski [30] illustrated that formation underpressures could be partially attributed to the removal of glacial ice loads. The authors suggested that discrepancies between the model predicted results and observed underpressures could be the result of factors not evaluated, such as the presence of a gas phase, osmotic pressure, and/or erosion.…”

“…They include, among other osmosis, the presence of a nonwetting gas phase, hydromechanical effects due to exhumation, cyclic glacial loading, and crustal flexure [22][23][24][25][26][27][28][29][30]. Abnormal underpressures in geologic formations have been reported in many regions of the globe [2], including China [31,32], Canada [33][34][35], and Switzerland [36].…”

Ice sheet load cycling and fluid underpressures in the Eastern Michigan Basin, Ontario, Canada

“…[] obtained underpressures in simulations when upper boundary head, representing a surficial aquifer, did not increase during episodes of ice cover. Khader and Novakowski [] also obtained underpressures when head in the Cambrian aquifer rose only minimally during glaciation. To emulate these results we considered variants of the base case in which first G 2 (middle boundary) and then G 3 (bottom boundary) was set at 0.1, so that heads increased by only about one tenth of ice height during glaciation.…”

“…In contrast, both Nasir et al . [] and Khader and Novakowski [] obtained underpressures more or less like those observed by simulating hydrologic changes and mechanical loading in two‐dimensional domains. The different outcomes may reflect different representations of conditions during glacial cycles or subsurface properties, suggesting that the response of the groundwater system can depend strongly on them.…”