Date of Award:


Document Type:


Degree Name:

Master of Science (MS)



Committee Chair(s)

Andrew Kulmatiski


Andrew Kulmatiski


Karen H. Beard


Bruce Bugbee


Stomata are the gateway between the lithosphere, the biosphere, and the atmosphere. Because of photosynthesis, plants inevitably lose water through their stomata. The rate at which water moves through stomata is stomatal conductance. As stomatal conductance increases, the rate of CO2 assimilation increases, therefore, plants must reach a balance between acquiring CO2 and losing H2O. Plants achieve this balance by adjusting stomatal aperture. Therefore, modeling stomatal conductance is important to global circulation models and land surface models, as well as for predicting how changing climate conditions affect water use efficiency and plant productivity, and has implications for agriculture and natural resource management.

Here a large dataset of field measurements was used to describe stomatal conductance for Kruger National Park, South Africa and develop statistical models of landscape-level stomatal conductance. Then models were used to estimate stomatal conductance across the region over several growing seasons. Over 8,000 measurements of stomatal conductance were made in four sites that represented a range of precipitation regimes and soil types within Kruger National Park from 2007-2012. Known environmental drivers of stomatal conductance, such as soil moisture, temperature, and shortwave radiation, were also measured during this period.

Observed mean daytime stomatal conductance for the park was 75 ± 1 and 155 ± 2 mmol m-2 sec1 for grasses and woody plants, respectively. When statistical models were used to produce three years of continuous estimates of gs from environmental data, average daytime stomatal conductance was estimated as 67 and 158 mmol m-2 sec-1 for grasses and woody plants, respectively. The Random Forest statistical models that were used to produce continuous estimates of gs indicated that soil moisture, particularly at shallow depths, and plant species identity are primary drivers of landscape-scale stomatal conductance for Kruger National Park. However, results indicate that there is still a large amount of landscape-scale variability in stomatal conductance that the environmental drivers investigated here were unable to explain.

Results provide a rare example of landscape-level estimates of stomatal conductance based on direct measurements. The models give insight into the relative importance of environmental drivers and the nature of their effect on stomatal conductance in savanna ecosystems. Because the measurements were collected over a range of species and soil conditions, the models should provide inference for many deciduous, sub-tropical savannas of southern Africa.