Primary production estimation from continuous oxygen measurements in relation to external nutrient input

In experimental ditches gross and net production were estimated from continuous dissolved oxygen measurements for individual days over a two year period. There are two series of three ditches: the first is dominated by benthic algae, and the second by submerged macrophytes. Each series receives three different levels of external nutrient (N,P) loading, referred to as reference, medium-loaded and highest-loaded level. Generally, there were no vertical oxygen gradients, and a one-box model satisfied. For periods with O2-stratification a two-box model has been developed. This model assumes that primary production only occurs in a well mixed top layer. In both series, the cumulative gross primary production over two years was positively related to the level of external nutrient input. In the ditches dominated by benthic algae, a continuous increase in cumulative net systems production, that was positively related to the level of external nutrient input, was observed. In the macrophyte ditches, the cumulative net systems production was positively related to the level of nutrient input during the first half year, but after that showed an irregular pattern. The maximum, areal daily gross productions during the summer, corrected for temperature, increased from 1990 to 1991 in all ditches dominated by benthic algae, but decreased in all macrophyte ditches. The accumulation of biomass in the ditches with benthic algae was calculated from the cumulative net production, and showed a good agreement with measured values in the reference ditch, but yielded higher values than measured in the medium-loaded and highest-loaded ditches. Reaeration was calculated using a system specific, empirical wind-reaeration relationship. The wind speed was on average higher during the afternoon than during the rest of the day. High wind speeds thus coincide with high oxygen saturations, and the use of a variable reaeration coefficient during a day results in higher gross and net production estimates than the use of a constant value throughout a day.