PhD Graduation

There is concern that a warmer climate may boost carbon emissions from lakes and promote the chance that they lose their vegetation and become dominated by phytoplankton or cyanobacteria. However, these hypotheses have been difficult to evaluate due to the scarcity of relevant field data.  To explore potential climate effects we sampled 83 lakes along a latitudinal gradient of more than 6000 km ranging from Rio Grande do Norte in Brazil to the South of Argentina (5-55 ^oS). The lakes were selected so as to be as similar as possible in morphology and altitude while varying as much as possible in trophic state within regions. All lakes were sampled once during summer (subtropical, temperate and tundra lakes) or during the dry season (tropical lakes) between November 2004 and March 2006 by the same team.

In the first chapters I address the question how climate might affect the chances for shallow lakes to be dominated by submerged plants. It has been shown that temperate lakes tend to have two contrasting states over a range of conditions: a clear state dominated by aquatic vegetation or a turbid state. The turbid state is typically dominated by phytoplankton and often characterized by poorer water quality than the clear state. The backbone of the theory explaining this pattern is a supposed positive feedback of submerged vegetation on water clarity: vegetation enhances water clarity and clearer water, in turn, promotes vegetation growth. The theory furthermore asserts that submerged vegetation coverage diminishes when nutrient concentrations increase until a critical point at which the entire vegetation disappears due to light limitation. Both aspects of the alternative state theory have been well studied in temperate shallow lakes, but the validity of the theory for warmer lakes has been questioned. In chapter 2 a graphical model is used to show how climate effects on different mechanisms assumed in the theory may affect the general predictions.  An analysis of our data presented in chapter 4 reveals that submerged vegetation has similar overall effects on water clarity across our climatic gradient. Nonetheless, the results hint at differences in the underlying mechanisms between climate zones. For example, the data suggest that the positive effect of vegetation on top-down control of phytoplankton by zooplankton is lost at high densities of fish that are often found in warmer regions. The main factor explaining differences in the water clearing effect of vegetation among lakes in our data set was the concentration of humic substances. In lakes with a high concentration of humic substances vegetation did not enhance the water clarity.

Combining our data with results from North America and Europe (chapter 3) we found that the critical nutrient level to maintain substantial submerged vegetation coverage is much less predictable in warm lakes than in colder lakes. This might be related to a poor correlation between phytoplankton biomass and nutrients in warmer areas (chapter 2). Despite the large unexplained variability in plant coverage in warm lakes our data suggest that the phosphorus concentration allowing substantial submerged macrophyte coverage increases steeply with the average number of frost days in a region. This implies that a reduction of phosphorus levels may be needed in regions where climate warming reduces the number of frost days if we aim at maintaining an equal submerged vegetation coverage.

In chapter 5 I show that nitrogen limitation may play an important role in shallow lakes as well. Our results indicate that submerged vegetation may tolerate higher phosphorus concentrations when nitrogen concentrations are low. Contrary to our expectations, evidence for nitrogen limitation was not systematically related to climate in our lakes. Also, unlike studies in other regions I did not find a relationship between the occurrence of cyanobacteria and the ratio between total nitrogen and phosphorus.

To explore the potential effect of climate on cyanobacterial dominance we combined our data with a data-set obtained in a large European study (chapter 6). The results indicate that warmer lakes have substantially higher probabilities of being dominated by cyanobacteria than colder lakes of the same nutrient levels.

In chapter 7 I address the question how climate affects the carbon metabolism of lakes. Whether an aquatic ecosystem acts as a net source or sink of atmospheric carbon depends on the balance between organic carbon burial and carbon dioxide (CO2) evasion, both of which are influenced by gross primary production, carbon fluxes from terrestrial ecosystems and ecosystem respiration. These processes and many other aspects of lake functioning and community composition are affected by temperature, it is therefore unclear a priori how a warm lake’s metabolism may differ from that of a colder lake. Our measurements revealed that the warmer lakes were generally more supersaturated with CO2 than the colder ones, while there was no underlying latitudinal gradient in dissolved organic carbon (DOC) or in the trophic state of the lakes. At similar algal biomass:DOC ratios, the colder (South Argentinean) lakes were undersaturated, whereas the warmer (Mid and North Brazilian) lakes were oversaturated. Correcting for the phenomenon that gasses dissolve better in colder water, our results indicate that going from colder to warmer lakes the CO2 releasing processes accelerated more than CO2 absorbing processes did, making warm lakes a more intense source of atmospheric CO2 than comparable cooler ones. Since aquatic ecosystems are important components of the global carbon circle these differences can have notable implications for regional carbon balances in a warming world: cool lakes may start to emit more CO2 when they warm up.

Promotor: Prof.Dr. Marten Scheffer