Average annual carbon dioxide concentrations in eight neighboring lakes in northern Wisconsin, USA
Carbon dioxide concentrations in the surface waters of lakes are often not in equilibrium with the atmosphere. Analyzing data from 4665 samples taken from surface waters of 1835 lakes worldwide, COLE et al. (1994) found that less than ten per cent of the samples were within 20 \% of atmospheric equilibrium and that most were oversaturated. Since movement of C02 across the air-water interface will always tend towards C02 equilibration of lake water withthe air above it, the fact that lake water is rarely in equilibrium suggests strongly that processes other than movement of C02 across the air-water interface dominate in controlling C02 concentrations in water. A number of processes, acting at different time scales, affect C02 concentrations in surface waters of lakes. At a time scale of days to weeks, C02 concentration is determined largely by the balance of photosynthesis and respiration (e.g. HAMILTON et al.1994 ). Seasonal patterns· in surface water values can be caused by C02 uptake by algae during summer and entrainment of high C02 bottom waters during periods of mixing. In dimictic lakes, this leads to high C02 in surface waters in spring and fall relative to summer. At longer time scales of years to decades, the average surface water C02 concentration is affected by net movement of C02 across the air-water interface, loading of carbon from terrestrial or wetland sources, and by removal of carbon by permanent burial in sediments or outflow from the lake. In this paper, we focus on this longer time scale of years and on the role external loading of carbon plays in the annual C02 dynamics of eight neighboring lakes in the Northern Highland Lake District in northern Wisconsin, USA. We address three questions. First, how do calculations of the partial pressure of carbon dioxide (pC02) based on measurements of pH and dissolved inorganic carbon (DIC) compare with measurements made using an equilibration technique? Second, how different are the open-water-season average C02 concentrations in these eight lakes? Third, what factors account for any observed differences? Groundwater flow is an important hydrologic feature of lakes in the region (KENOYER \& ANDERSON 1989, KENOYER \& BowsER 1992, KRABBENHOFT et al. 1990 a, b), because approximately 40 m of noncalcareous sandy tills and outwash overlie granitic bedrock in this region (OKUEZE 1983). The eight study lakes are arranged along a groundwater flow system with lakes high in the flow system receiving most of their water direcdy from precipitation, and lakes lower in the flow system receiving as much as 35 \% of their input as groundwater (WEBSTER et al. 1966). The lakes are within 10 km of each other, so they are subject to the same weather and share a similar geologic setting. Seven of the lakes are primary study lakes of the North Temperate Lakes LongTerm Ecological Research program (MAGNUSON \& BowsER 1990). The eighth lake, Little Rock Lake; was the site of a whole-lake acidification experiment in which this hourglass-shaped lake was divided in two by a nylon curtain and one-half was acidified (BREZONIK et al. 1993). We use data from the unacidified basin of the lake. Some limnological characteristics of the lakes are given in Table 1.