Environmental variables and plankton communities in the pelagic of lakes: enclosure experiment and comparative lake survey

Most primary production of lakes and oceans occurs in the
well-mixed surface layer that is subject to strong seasonal and
geographical variation. With increasing mixed surface layer depth
average light supply and specific nutrient supply decrease and so
do light-dependent production rates and depth-dependent sinking
loss rates of phytoplankton. Changes in mixing depth are expected
to have important consequences for the dynamics of phytoplankton
biomass, algal nutrient stoichiometry, light availability and
nutrient retention in the mixed layer. Light absorption by enhanced
concentrations of abiotic substances (humic substances, clay
particles) furthermore negatively affects light availability and
production. I tested the predictions of a dynamical “closed system”
model concerning the effects of mixing depth and background
turbidity (Kbg) on phytoplankton biomass, light climate and
nutrients in a field enclosure experiment. The natural
phytoplankton community was exposed to high and low background
turbidity along a gradient of mixing depth. For sinking algae, the
model predicts that phytoplankton biomass should be most strongly
limited by sedimentation losses in shallow mixed layers, by mineral
nutrients at intermediate mixing depths and by a lack of light in
deep mixed layers. As predicted, phytoplankton volumetric and areal
biomasses showed a unimodal relationship to mixing depth and were
negatively affected by background attenuation. With increasing Kbg
the biomass peak shifted towards shallower mixing depth. The
concentrations of dissolved and total nutrients were positively
affected by increasing mixing depth but only marginally related to
Kbg most likely due to a variable carbon to phosphorus cell quota.
For thermally stratified lakes I derived the following predictions
from a dynamical “open system” model which includes variable algal
cell quota: within a realistic mixing depth range (3-12m) light
availability, phytoplankton density, and the carbon:phosphorus
ratio of algal biomass should all be negatively related to mixing
depth, while algal standing stock should be unimodally related, and
total and dissolved nutrients be horizontally or positively related
to mixing depth. All these prediction were in qualitatively good
agreement with data from 65 central European lakes sampled during
summer stratification. Notably, I observed the predicted negative
relationship between phytoplankton density and mixing depth in
spite of the rather limited range of mixing depths typical for
medium sized temperate lakes. Furthermore, I found a strong
negative relationship among zooplankton biomass and mixing depth.
In a comprehensive comparative lake study of 40 northern German
lakes, I sampled the surface mixed layers for a set of variables
and focused on the taxonomic composition of phytoplankton and the
relationships of taxonomic classes to environmental variables. I
used high performance liquid chromatography to analyse the
phytoplankton samples for 13 photosynthetic pigments and calculated
the contributions of seven algal classes with distinct pigment
signatures to total chlorophyll a using CHEMTAX, a matrix
factorisation program. In multiple regression analyses, I examined
the relationships of phytoplankton biomass and composition to total
nitrogen (TN), total phosphorus (TP), total silica (TSi), mixing
depth, water temperature, and zooplankton biomass. Total Chl-a was
positively related to TN and TP and unimodally related to mixing
depth. TN was the factor most strongly related to the biomasses of
single taxa. I found positive relationships of chrysophytes,
chlorophytes, cryptophytes, and euglenophytes to TN, and of diatoms
and chrysophytes to TSi. Diatoms were negatively related to TN.
Cryptophytes and chlorophytes were negatively and cyanobacteria
positively related to zooplankton. Finally, the relative biomasses
of chrysophytes and cryptophytes were negatively related to mixing
depth. Most results were consistent with theoretical expectations.
Some relationships may, however, have been masked by strong
cross-correlations among several environmental variables.