Translocation of soil organic matter following reservoir impoundment in boreal systems: Implications for in situ productivity

To evaluate the effect of reservoir flooding on carbon cycling over time, we studied sedimentary environments in three natural lakes that existed prior to impoundment and that have been incorporated in the larger lentic system of a 70-yr-old boreal reservoir in Quebec. Elemental and biomarker analyses were determined in all core intervals and three soil profiles to characterize source inputs of organic matter to sediments. Following impoundment, a twoto threefold increase in lignin concentrations (5 to 10–25 mg [10 g dry wt]21) associated with similar decreases in selected terrigenous biomarker ratios (cinnamyl phenols : vanillyl phenols, C : V, 1.0–2.0 to 0.2–0.5; 3,5-dihydroxybenzoic acid : vanillyl phenols, 3,5-Bd : V, 0.2–0.5 to 0.1–0.2; and p-hydroxyl phenols : sum of vanillyl and syringyl phenols, P : [V1S], 0.3–0.7 to 0.1–0.4) illustrate the effect of soil erosion and subsequent translocation of surface soil organic matter (SOM) to sedimentary deposits. Using a mixing model based on mass-normalized biomarker yields in identified surface and mineral soil end-members, we show that although the proportion of mineral-derived SOM predominates in the receiving sedimentary systems during preflooding conditions (97–99% of allochthonous inputs), translocated surface SOM increased in postflooding sediments to comprise up to 5–30% of the total allochthonous organic matter inputs. Using a similar quantitative mixing model based on elemental C and N contents in the autochthonous and the mixed allochthonous end-members, we estimated that concomitant to the redistribution of eroded soil organic matter, reservoir flooding induces a 1.5to 2-fold increase in the fraction of autochthonous organic matter transferred to sedimentary environments of the reservoir. Results presented here suggest that reservoir flooding induces a change of state in lake systems that may perdure over time scales of decades to centuries. Large-volume reservoirs, built primarily for hydroelectric power generation, are among the largest man-made water infrastructures and hold at any given time 10–20% of the global mean river runoff (Rosenberg et al. 1997; Vörösmarty and Sahagian 2000). Beyond their recognized disruptions of water flow and impairment of water quality (Rosenberg et al. 1997; Hambright et al. 2004), large-scale reservoirs also result in significant shifts in trophic structure during the transition from a primarily lotic to a lentic system (Kalff 2002). Indeed, the creation of new reservoirs frequently leads to short-term spikes in nutrient levels and anomalously high in situ productivity including sharp increases in bacterial bio1 Present address: University of Colorado at Boulder, Department of Chemistry and Biochemistry, 215 UCB, Boulder, Colorado