Small temporal and spatial variations in the distribution of dissolved oxygen in Crater Lake, Oregon, are used to estimate the mean age of the lake's deep water, the flux of labile organic carbon to the deep lake, and the influence of hydrothermal activity on the concentration of dissolved oxygen within the lake. An increase in the concentration of dissolved oxygen in the deep water during winter [1988][1989] indicates that 30-45% of deep water was replaced with well-oxygenated surface water. This deep-water renewal corresponds to a mean deep water residence time of 2-4 yr. The deep-water oxygen consumption rate is 2 l-26 pmol m-3 d-l, which occurs primarily through the oxidation of organic matter and, to a lesser extent, the oxidation of reduced inorganic species that are introduced to the lake via subsurface hydrothermal springs.Dissolved oxygen is a sensitive tracer of the physical and chemical processes occurring in an aquatic system. Oxygen in excess of atmospheric saturation in the upper water column is caused by photosynthesis, in situ warming of the water column by penetration of solar radiation, and air bubble injection (Jenkins and Goldman 1985;Emerson et al. 199 1). The decrease (consumption) in dissolved oxygen at depth typically results from oxidation of particulate organic material raining out of the euphotic zone. If the photosynthetic component of the oxygen excess in the upper water column is properly constrained, then this component should have nearly the same value, but opposite in sign, as the amount of oxygen consumed in deep water. This balance between surface excess photosynthesis and deep-water respiration assumes that only a small portion of the photosynthetically derived carbon is buried in sediments.Because of the consumption of oxygen at depth in an aquatic system, deep water is generally undersaturated with respect to atmospheric saturation. In temperate lakes, oxygen-rich wintertime surface water mixes with deep water, resulting in a net increase in dissolved oxygen at depth at the time of mixing. This change in oxygen pro-
AcknowledgmentsMark Buktcnica, James Milestone, John Salinas, and Scott Stonum provided day and night field support for this research; their efforts arc greatly appreciated. The field and analytical efforts of the OSU research team were invaluable. Eric Olsen and Marv Lilley provided the unpublished methane data. Special thanks are extended to Roberta Conard and Chris Moser.