DEPARTMENT OF MATHEMATICAL & STATISTICAL SCIENCES

UNIVERSITY OF ALBERTA

**MATHEMATICAL
BIOLOGY SEMINAR**

MONDAY, NOVEMBER 18, 2002

3:00 PM

CAB
657

**Dr. Brian Leung**

**Department of Biological Sciences**

**University of Notre
Dame**

Biological Invasions: Predictions, Risk, and Action

Ecological forecasting and risk analysis are
exciting intellectual frontiers where the science of ecology meets critical
public policy needs. As the recent National Invasive Species Management Plan
(National Invasive Species Council 2001) and other reports (CENR 1999) have
recognized, ecological guidance and quantitative risk analyses are crucial to
respond to many environmental challenges, including those posed by
nonindigenous species (NIS). Nonindigenous species are one of the top causes of
biodiversity loss in many ecosystems. In addition, they can cause large
ecosystem changes, and interact strongly with many other drivers of global
environmental change. The financial cost of nonindigenous species to the US
alone has been roughly estimated at $137 billion/yr. Clearly, it is important
to determine which areas are at risk of invasion to focus prevention effort. It
is likewise important to determine reasonable actions, both in terms of
prevention before and control after an invasion. I have developed a model to
forecast the location of new invasions, using propagule pressure, gravity
models, and population dynamics in terms of Allee effects. Then, I use
bioeconomic risk analysis for decision formulation, with stochastic dynamic
programming as its mathematical basis. These models are applied to zebra mussel
invasion of inland lakes. For forecasting invasions, I demonstrate that Allee
effects are present and that, using a validation data set, we can correctly
predict up to nine times as many invasions as the null (random) model. Further,
the Allee model assigned average probabilities 4.5 times higher for lakes that
became invaded compared to uninvaded lakes, whereas the non-Allee model only
predicted probabilities two times as high. Thus, this model improves
predictions of future invasions, and demonstrates the importance of considering
the Allee effect. For determining appropriate actions, I use cost-benefit
analyses and show that society would benefit by spending up to $324,000 USD/yr
to reduce invasions of zebra mussels into each lake with a power plant. In
sharp contrast, US Fish & Wildlife Services spent $825,000 in FY2001 to
manage *all* aquatic invaders for *all* lakes. Thus, greater investment in
prevention appears warranted.