We sampled surface waters of 31 lakes in the Northern Highland Lake district of Wisconsin and the Upper Peninsula of Michigan during July and August of 2000 (Table 1). The lakes were chosen to span wide and orthogonal ranges in DOC and TP concentrations and for their close proximity to the Trout Lake Station in Vilas county, Wisconsin. The order in which the lakes were sampled was randomized.
Limnological samples were collected for each lake at 0.5 m depth as follows. DOC samples were collected as the filtrate through Whatman GF/F filters, and were analyzed on a Shimadzu model 5050 high temperature TOC analyzer. Color was also measured from this filtrate as absorbance at 440 nm on a Spectronic Genesys 2 spectrophotometer using 10 cm quartz cuvettes. Chlorophyll a was collected by filtering 200 ml of lake water, and then freezing filters for at least 24 hours, followed by methanol extraction for 24 hours. Fluorescence was determined before and after acidification to correct for pheopigments. Total phosphorus was analyzed on a Lachat autoanalyzer after persulfate digestion of a whole water sample. DIC and was measured on a Shimadzu GC-8AIT (TCD detector) gas chromatograph. DIC was determined from the headspace of acidified samples, which was injected into the GC. pH was measured using an Orion digital pH meter with automatic temperature compensating electrode. Temperature and dissolved oxygen profiles were measured using a YSI temperature/dissolved oxygen meter. Spot measurements of surface water DO were made on quadruplicate samples, using Winkler titrations as described in Bade and others (1998).
We deployed a buoy that sampled dissolved CO2, DO, water temperature, photosynthetically active radiation (PAR), and wind speed for 2-4 days on each lake. All water measurements were made at a depth of 0.5 m. Wind speed was measured one meter above the lake, using an RM Young model 03001, and PAR was measured 10 cm above the lake surface using a Li-Cor model 190SA quantum sensor. Electronic control and data collection were managed by a Campbell Scientific CR10X data logger. DO and water temperature were measured with a YSI model 600-XLM sonde fitted with a Rapid Pulse oxygen probe (model 6562) and temperature sensor. The sonde was attached to the buoy at the opposite end from the CO2 equilibration chamber (described below).
We measured dissolved CO2 independently from DO. We equilibrated a closed loop of atmospheric gas in an equilibration chamber submerged to 0.5 m. The equilibrated gas volume was about 234 ml. We recirculated gas for the last 10 minutes of every 30 minute period, with a flow rate of about 9 ml s-1. A pump exchanged lake water every minute during equilibration. Equilibrated gas was diverted to the IRGA, equipped with a 14 cm sample cell for lakes with CO2 concentration under 2000 ppm or a 5 cm sample cell for lakes with CO2 concentration between 2000-20000 ppm. Following analysis of equilibrated gas, solenoids were activated to route atmospheric gas (taken 10 cm above the water) for CO2 analysis.
Time series data are included for 25 of the lakes.
Additional detail of the methods available in Hanson et al. (2003)
Hanson, P. C., D. L. Bade, S. R. Carpenter, and T. K. Kratz. 2003. Lake metabolism: Relationships with dissolved organic carbon and phosphorus. Limnol. Oceanogr. 48: 1112-1119.