A New Noble Gas Paleoclimate Record from Texas: Perils and Promise Explained

Typically, it is assumed that because of dispersion, signals from rapid climate change are lost in groundwater systems. This plus the uncertainties in 14C groundwater dating have meant that significant negative shifts in noble gas temperatures (NGTs) have usually been associated with large scale events such as the Last Glacial Maximum (LGM). New noble gas measurements on 49 samples from the Carrizo aquifer in South Texas have been used to construct a detailed paleoclimate record over the past ~10ka (Castro et al., 2007). The build up of radiogenic 4He was used to constrain a 3D hydrogeological model of unprecedented detail for the Carrizo and confirms earlier work (Castro and Goblet, 2003) suggesting that the large NGT drop in this area occurred ~6ka and not at the LGM. A study of the effect of dispersion using aquifer characteristics derived from the 3D model demonstrate that rapid shifts in climate can be recorded well into the Pleistocene. The NGT record using standard models matches mean air temperature (T) in the recharge area, but shows a ~ 5 °C bias to low T relative to present ground T, with a sudden further drop in T at about 1ka and a minimum ~ 8 °C below present day values at about 6ka. This is significantly younger than the LGM and well below the expected temperature at this time. Hall et al. (2005) suggested an NGT model that assumes that O2 depletion at the water table enhances noble gas partial pressures and this model successfully corrects for the 5 °C temperature bias seen in recharge area samples while simultaneously improving NGT model fits. The model also shows a marked change in the amount of excess air seen in samples, which drops suddenly at ~ 1ka. We interpret the temperature drop and the reduction of excess air at this time as indicating a change mostly in recharge conditions rather than atmospheric T, with the water table approaching the surface. It is not an indicator of the LGM as had previously been assumed (Stute et al., 1992). Castro and Goblet, 2003, Geophys. Res. Lett., 30 (24), 2251, doi:10.1029/2003GL018875. Castro et al., 2007, Earth Planet. Sci. Lett., 257, 170-187, doi:10.1016/j.epsl.2007.02.030. Hall et al., 2005, Geophys. Res. Lett. 32(18) L18404, doi: 10.1029/2005GL023582. Stute et al., 1992, Science, 256, 1000-1001.