Eutrophication is one of the most common threats to water quality of lakes world wide. The positive relationship between phosphorus loading and phytoplankton biomass is well established. However, little is known about the response of benthic algae (periphyton) to whole-lake fertilization. I used a combination of small-scale and whole-lake experiments to determine: 1) if the response of periphyton to fertilization depended on the substratum on which it grew; and 2) if the contribution of periphyton to whole-lake primary production changed along a nutrient gradient. Whole-lake experiments were conducted from 1991 - 1995 in five lakes at the University of Notre Dame Environmental Research Center. During 1993 - 1995, four of the lakes were fertilized with nitrogen and phosphorus.

Substratum strongly affected periphyton response to water column fertilization. Periphyton on sediments (epipelon) sequester dissolved inorganic carbon and other nutrients from the sediment pore water. Interstitial water had higher concentrations of nitrogen and phosphorus than did the water column. Therefore, when nutrients were added to the water column, there was no increase in epipelic algal biovolume or production. In contrast, biomass and production of periphyton on wood (epixylon) increased with fertilization. These substratum-specific responses occurred both in small-scale laboratory experiments and at the whole-lake scale.

From 1991 - 1995 biomass and production of epipelon, epixylon, and phytoplankton were monitored. In addition, surface area of sediments and wood was estimated in each lake. These data were used to model benthic and whole-lake (benthic + pelagic) primary production. Wood constituted up to 16% of benthic surface area. However, areal biomass and production of epipelon were up to 10 x higher than those of epixylon. Therefore, despite the positive response to fertilization, epixylon constituted a maximum of 3.6% of total benthic production . At ambient nutrient loading rates, epipelon accounted for 50 - 80% of whole-lake (epipelon + phytoplankton) primary production. However, whole-lake epipelic production declined due to light-attenuating phytoplankton blooms when the lakes were fertilized. When there was a large increase in phytoplankton production, whole-lake primary production increased with fertilization. However, when phytoplankton response to fertilization was moderate, declines in benthic algal production compensated for the increases in phytoplankton production, and there was no change in whole-lake primary production. Thus, the historical focus on phytoplankton greatly overestimates the response of whole-lake primary production to fertilization. These experiments demonstrate the importance of benthic-pelagic links to ecosystem processes in lakes.