Date of Award

May 2018

Degree Type


Degree Name

Doctor of Philosophy


Freshwater Sciences

First Advisor

Harvey A Bootsma

Committee Members

val Klump, James Waples, Sandra McLellan, John Berges


African Great Lakes, Biogeochemistry, Carbon dioxide exchange, Carbon dynamics, Lake Malawi, Tropical


Large lakes of the world play a vital role in the global carbon cycle as they act both as conduits and sinks of terrestrially and atmospherically derived carbon. Lake Malawi, lying at the extreme southernmost end of the East African Rift Valley is one of the largest, deepest and most ancient of the African Great Lakes. In this study, the spatial and seasonal variation of direct measurements of air and water pCO2 were taken for a period of one annual cycle using a vessel of opportunity along the north-south axis of Lake Malawi. These data, together with limnological and meteorological variables, were used to estimate the annual net CO2 flux at the air-water interface.

The data reveal distinct spatial and temporal variation in pCO2 and CO2 flux that is related to hydrodynamic and meteorological conditions that drive nutrient dynamics and phytoplankton productivity. Contemporaneous measurements of lake temperature profiles, nutrients, weather conditions, phytoplankton biomass and seston δ13C suggest that increased nutrient supply due to vertical mixing and allochthonous inputs promotes high phytoplankton growth rates and CO2 uptake during the cool, mixing season and the hot, rainy season. Spatially, the southernmost region of the lake which is the most nutrient-rich and hence most productive was distinct from the rest of the lake. High CO2 efflux to the atmosphere was observed in this region at the onset of the cool, mixing season probably due to the physical resupply of dissolved inorganic carbon (DIC) from deep waters during upwelling. Seasonally, almost the entire lake was CO2 undersaturated with respect to the atmosphere during the wet, hot season (December to April) and the cool, mixing season (July to September), periods when nutrient supply from river inputs and vertical mixing that promote phytoplankton photosynthesis are high. By contrast, during the hot, stratified season (October and November), CO2 evasion to the atmosphere was observed, possibly driven by high respiration to photosynthesis ratios.

The experiments conducted to determine the influence of river water loading and vertical exchange on the metabolism of Lake Malawi using Linthipe River water and hypolimnetic water from Lake Malawi shows distinct differences. River loading results in CO2 supersaturation implying high respiration rates while hypolimnetic water showed a net consumption of carbon dioxide. Low phytoplankton biomass and particulate organic carbon production were observed in incubation bottles spiked with river water. In contrast, bottles spiked with hypolimnetic water showed high phytoplankton biomass and particulate organic carbon. The high OC: DP ratio compared to lake seston stoichiometry in Linthipe River is responsible for the observed heterotrophy while autotrophy by hypolimnetic water was sustained by the relatively low OC: DP from vertical flux.

Autochthonous primary production constitutes the major source of organic carbon in the lake and although concentrations of DOC and POC are relatively low compared to other lakes, the internal organic carbon inventory is large. The vertical exchange is an important source of DIC to the upper 200 m of the lake and it appears the recycling rate of carbon decreases with depth. A comparison of carbon sedimentation rates and DIC vertical flux rates among different strata in the lake suggests that the carbon recycling efficiency within the epilimnion is 73%, while it is 33% within the anoxic hypolimnion. If carbon is selectively retained while P is efficiently recycled into the epilimnion, CO2 fixation in the epilimnion will be enhanced leading to autotrophy.

Several studies have indicated that the majority of oligotrophic inland waters are net sources of CO2 to the atmosphere. Results from the present study indicate that this paradigm may not apply to large tropical lakes. On an annual basis, Lake Malawi is a net CO2 sink and hence net autotrophic, absorbing 209 to 320 mmol C/m/yr from the atmosphere. Using the Lake Malawi C:P stoichiometry requirements and the carbon mass balance approach, we still determined that the lake is a net CO2 sink. The data further suggest that surface pCO2 variability is driven primarily by biological processes and vertical mixing, with seasonal temperature fluctuations playing a minor role. The variability in CO2 underscores the importance of making measurements with high spatial and temporal resolution to accurately determine air-water gas fluxes in large lakes.