Technical Comments

Comment on “Reconciliation of the Devils Hole climate record with orbital forcing”

See allHide authors and affiliations

Science  21 Oct 2016:
Vol. 354, Issue 6310, pp. 296
DOI: 10.1126/science.aaf7718


Moseley et al. (Reports, 8 January 2016, p. 165) postulate an increase in dissolved thorium isotope 230Th with depth below the water table as the explanation for the differing ages of Termination II. Flow of geothermal water through the Devils Hole caverns precludes this explanation. Deposition of younger secondary calcite into the initial porosity of the calcite comprising their cores is a plausible alternate explanation.

Moseley et al.’s (1) postulated (not measured) increase in 230Th with depth below the water table—invoked to explain the different published ages for Termination II (13)—is doubtful in consideration of the modern hydrodynamic setting of the two Devils Hole caverns. These caves, developed within a Paleozoic carbonate aquifer, are within 1 km of a major fault (4) that dams groundwater flow within the aquifer creating the spring-fed Ash Meadows oasis bordering the caverns on the west. Groundwater flow through the aquifer occurs solely via open fractures and faults (4); the caverns comprise such open major fissures that trend northeast-southwest and are of extensional tectonic origin (5). That groundwater in the caverns is moving toward the oasis is well documented based on (i) the hydraulic gradient between the caves and the springs and (ii) identity of the chemistry, stable isotopes, and temperature of cavern water with that discharging from the major springs (4, 6, 7).

Temperature surveys (8) to a depth of 32 m below the water table provide additional evidence of the hydrodynamic nature of the water column within these caverns. These surveys, in the larger of the two caves, documented the following: (i) vertical convective mixing during winter months; (ii) thermal stratification during summer months with possible periodic convective mixing; and (iii) “horizontal groundwater flows through discrete zones of high permeability.” The temperature gradient during the summer months of minimal vertical convective mixing is about 0.14°C/32 m, or about 4.4°C/km. This gradient is a fraction of geothermal gradients elsewhere in the Great Basin, which vary from 20° to 50°C/km (9), but is similar to the very low gradients in other areas of upward flow of geothermal waters within fault zones (10). Additional evidence of active flow of water is provided by the chemical homogeneity of the water column between 5 and 38 m, as documented in table 1 of Plummer et al. (11), who noted, “Thermal convection and groundwater advection probably mix the water-table fluids to some depth along the fault plane.” That the water column is dynamic is most easily comprehended by noting that the temperature of the water surface (33.5°C) is ~15° warmer than 18.5°C, the mean annual air temperature (7).

Briefly, several lines of evidence indicate that the caverns are not standpipes isolated from active flow in the carbonate aquifer; they comprise major avenues of water flow in the aquifer. Moseley et al.’s postulated buildup of 230Th in the water column would be displaced by flow within the fissures and cannot be a valid explanation for the different ages of Termination II. During glacial terminations, the postulated buildup of 230Th with depth below the water table is even more difficult to envision. During such times of transition from relatively wet to relatively dry climates, the water table in the caverns was several meters above modern (12), with likely greater than modern groundwater flow and chemical homogenization of the water column in the caverns.

Using seven observations gleaned from the Devils Hole literature and from Moseley et al. (1), I present below a plausible alternate explanation for the discrepancy in ages of Termination II. (i) Devils Hole mammillary calcite was precipitated under equilibrium conditions and constant aquifer temperature during at least the past 180 thousand years (ky) (13). (ii) The internal partial pressure of CO2 in Devils Hole water (11) varies from 0.0123 to 0.0141 atm, or >30 times the modern atmospheric CO2 and >50 times that during Termination II when CO2 was ~230 parts per million by volume (ppmv) [figure 1 in (1)]. (iii) Outgassing of CO2 is occurring in the upper 5 m of the water column, as indicated by the following data in table 1 of Plummer et al. (11). The pH and calcite saturation index is higher, and the log Pco2 is lower at 5 m than at 25 and 37.5 m below the water table, leading Plummer et al. (11) to state, “More likely is the possible outgassing of CO2 to the atmosphere in contact with the water table at Devils Hole…” (iv) The porosity of mammillary calcite formed at >20 m depth is <<1%, whereas the porosity of folia, formed at the water table via outgassing, may reach 15 to 20% (12). (v) Moseley et al.’s core DH2-D was precipitated at an average rate 43% greater than DH2-E, their slightly deeper core, and 54% greater than DH-2, which was deposited 21 m below the water table [supplementary materials of (1)]. (vi) Calcite comprising DH2-D and DH2-E precipitated at a rate 10 times their average growth rate in these cores after ~9 thousand years ago (ka) [figure 1 in (1)] when the water table declined close to modern levels (12), indicating that calcite formed close to the water table grows rapidly. (vii) Folia (not mammillary calcite) are present in Moseley et al.’s highest core (DH2-D) between 121 and 116 ka, and in the next highest core (DH2-E) at ~117 ka [table S1 in (1)], showing that during Termination II, calcite in both of these cores formed in the shallow zone of outgassing observed in the modern DH.

The observations listed above permit the conclusion that Moseley et al.’s U-series ages are inaccurate in the vicinity of terminations. This conclusion relies on a key assumption, namely, mammillary calcite precipitated in the upper few meters of the water column had significant (>5%) primary porosity due to outgassing with attendant rapid deposition of calcite, in contrast to the dense calcite (porosity <<1%) comprising cores deposited at depths >20 m below the water table (2). If this assumption is correct, then the younger U-series ages recorded by Moseley et al.’s cores during Termination II reflect the deposition of secondary (and younger) calcite into the initial porosity of the calcite comprising their cores. Notably, the 8- to 10-ky discrepancy in ages between their shallow cores and those deposited at depths >20 m (2, 3) occurs during terminations because it is at this time that the water table is falling rapidly, outgassing is accelerated, and porous calcite is deposited in the upper few meters of the water column. The older age of Termination II in core DH2-E than in DH2-D is to be expected because DH2-E, having been deposited 1.3 m deeper in the water column than DH2-D, had less initial porosity. The conclusion outlined above can be verified or falsified by a thin-section examination of the cores. If correct, the segment of the cores covering Termination II will display numerous secondary growths of calcite within a matrix of crystal morphologies distinct from the dense elongate crystals typical of mammillary calcite deposited below 20 m. (14).

References and Notes

Acknowledgments: I thank T. B. Coplen and W. D. Sharp for helpful review comments.
View Abstract

Navigate This Article