The isotopic and hydrologic response of small, closed-basin lakes to climate forcing from predictive models: Simulations of stochastic and mean-state precipitation variations

A hydrologic and isotope mass-balance model is applied to two small, closed-basin lakes, Castor and Scanlon, in north-central Washington to describe the influence of hydroclimatic forcing on lake hydrologic and isotopic evolution. Simulations of lake responses to the combined effects of stochastic variability (i.e., random interannual fluctuations) and long-term (i.e., multidecade to century), mean-state changes in precipitation were conducted using 300 yr of randomly generated precipitation data as model inputs. Simulation results demonstrate that average (long-term) closed-basin, lake-water, oxygen-isotope values are dependent largely upon lake-water outseepage (i.e., subsurface outflow) rates, with lower (higher) outseepage resulting in decreased (increased) isotopic sensitivity to long term precipitation changes, and that the lake basin surface-area-to-volume (SA : V) ratio changes with depth influence the direction of the isotopic response. Simulation results also suggest that, as average lake volume decreases as a consequence of decreasing mean-state precipitation amounts, interannual lake-water isotopic variability in response to stochastic forcing will increase. Conversely, as the average lake volume increases in response to increasing mean-state precipitation amounts, interannual lake-water isotopic variations associated with stochastic forcing will decrease. Additional model experiments demonstrate that increased (decreased) variance in precipitation leads to increased (decreased) water volume within lakes over the long term, which (if stochastic variance changes are large enough) could result in decreased (increased) lake-water oxygen-isotopic sensitivity to stochastic precipitation. The oxygen isotope composition of closed-basin lake water is controlled by many factors including net groundwater flux, catchment size and soil characteristics, lake basin morphology, and hydroclimatic forcing through stochastic variability (i.e., random interannual fluctuations), and mean-state changes (i.e., changes in multidecade to century averages) in precipitation, relative humidity, and temperature (Gat 1970; Leng 2004; Steinman et al. 2010). In regions with highly seasonal climates such as the Pacific Northwest, precipitation and evaporation rates substantially vary intra-annually, whereas lake-basin morphology, lake seepage rates, and catchment hydrologic parameters such as soil thickness and depth, vary only on decadal and longer time scales, largely in response to interannual climate forcing (Johnson and Watson-Stegner 1987; Phillips 1993; Rosenmeier et al. in press). Changes in average climatic states, such as long-term increases in the variance of precipitation about an established annual mean, or the shifting of precipitation from one season to another, add additional complexity to the geochemical evolution and, more specifically, the isotopic composition, of lake water (Vassiljev 1998; Leng et al. 2001; Pham et al. 2009). Only by simulating lake sensitivity to these factors using quantitative models can the isotopic response of closed-basin lakes to climate change be predicted, a fact that has considerable relevance to paleoclimate interpretations of sediment core oxygen isotope (d18O) records (Ricketts and Johnson 1996; Cross et al. 2001; Jones et al. 2005). Groundwater flow regimes and associated lake seepage rates are of particular importance in the control of steadystate lake-water d18O values (Donovan et al. 2002; Smith et al. 2002; Shapley et al. 2008). In closed-basin systems that are isolated from regional groundwater (i.e., perched lakes), the origin of subsurface inflow is often precipitation falling on or very near the catchment (Almendinger 1993). In seasonal climates, under these conditions, base flow has an isotopic composition that is approximately equal to the weighted average isotopic composition of mean annual precipitation and, therefore, can effectively be considered runoff with an extended transit time (Henderson and Shuman 2009). In contrast, subsurface inflows to lake systems that are not isolated from regional groundwater aquifers can potentially undergo a much more complex isotopic evolution during transport to a lake (Smith et al. 1997). Subsurface outflow from closed-basin lakes occurs as seepage through the lake bed at rates that are typically very low due to the low hydraulic conductivity of silt and clayrich lake sediment. Outseepage, a potentially minimal hydrologic flux, is in all cases a fundamental control on the steady-state isotopic composition of closed-basin lake water because, through interaction with lake-basin morphology, it determines the proportion of water that leaves a lake through fractionating (evaporation) and nonfractionating (outseepage) pathways (Steinman et al. 2010). In this study, a numeric, hydrologic and isotope massbalance model was applied to Castor and Scanlon Lakes, north-central Washington, and was used to simulate lakewater hydrologic and isotopic responses to changes in * Corresponding author: [email protected] Limnol. Oceanogr., 55(6), 2010, 2246–2261 E 2010, by the American Society of Limnology and Oceanography, Inc. doi:10.4319/lo.2010.55.6.2246