INTRA- AND INTER-CAVE DRIP δD AND δ18O ISOTOPIC LAGS AND DAMPING FROM MONITORING IN THREE CAVES ALONG AN ALTITUDINAL GRADIENT, VANCOUVER ISLAND, BRITISH COLUMBIA, CANADA

The interpretation of speleothem paleoenvironmental records requires understanding of spatial-temporal variations in vadose drip water chemistry and isotopic composition. This study reports on intra- and inter-cave differences in δD, δ¹⁸O and electrical conductivity, using 1.5 years of monthly water samples from three adjacent drips (<20 m apart) in each of three caves at increasing elevation (0, 550, and 740 m ASL) on the steep headland of Tahsis Inlet fjord on the Pacific coast of Vancouver Island. Despite varied patterns of drip hydrology, all eighteen drip isotope records displayed strong and regular seasonal responses that could be approximated with sine waves, which we compare with the mean monthly local air temperatures at sea level, and precipitation at the Victoria GNIP station (~300 km to the south).

The expected δ¹⁸O depletion of -0.15 to -0.5‰/100 m of ascent was not observed, and instead there was overlap in isotopic ranges (at 1 SD) between all three caves. Notwithstanding, the drips at 740 m ASL are tightly aligned to the GMWL and ¹⁸O-depleted; the drips at 550 m ASL and at sea level plot along the GMWL, or between it and the Victoria LMWL, with the exception of the slowest drip (sea level). The δD and δ¹⁸O drip amplitudes were damped on average 74% and 73% respectively compared to the Victoria data. A first conclusion from this study is that the substantial intra- and inter-cave variation of δD and δ¹⁸O in drips was sufficient to mute the altitudinal isotopic lapse rate in this local coastal rainforest environment.

The δD and δ¹⁸O drip isotopes lagged the Victoria record by 155±26 days and 165±50 days respectively. The slowest drip had the longest lag times. The shortest lag (87 days for δ¹⁸O, cave at 550 m) implies a shortest residence time in the aquifer. When the calendar dates of the drip isotope records were shifted to fit the time lags there was excellent agreement with local sea level air temperatures. A second conclusion, is that in these caves, most or all of the volumes being discharged have a residence time of less than one climatic year for all components of the source reservoir combined – forest, litter, soil, epikarst and the bedrock transmission zone to the drip point. This reflects the negligible matrix porosity of the host rock and the super-humid conditions. Drip recharge was evidently by simple piston flow.