Predicting lakefront housing growth and changing bluegill growth rates using a linked economic-ecological model

Background/Question/Methods

Lakeshore development causes lake ecosystem changes, such as the removal of sunken logs serving as aquatic habitat, reduced growth rates for sport-fish, and increased nutrient loadings. Land development constraints (LDCs) such as zoning and open space preservation represent community’s attempts to mitigate these effects. However, it is often unclear how effective these policies are in limiting ecological impacts. Our goal here was to first quantify the effects of LDCs on the frequency and size of subdivisions along lakefront lots in Vilas County, Wisconsin, and second link these effects directly to estimated growth rates of bluegills (Lepomis macrochirus). An econometric model was estimated on a time series dataset from the years 1974-1998. This dataset included over 1,500 unique observations of parcels on 140 lakes as well as information on the physical features of each lake and parcel and the LDCs that influenced the parcel’s development potential. The model predicted both the probability of a parcel subdividing in any given time period, and the number of new parcels created for each subdivision. This model is then linked to an ecological model which predicts the growth rates of bluegills based on the housing density of a lake. To address the ecological effects of different landscape policies, we simulated development patterns under counterfactual policy choices. Specifically, we changed the minimum frontage zoning from its current level (either 100ft or 200ft depending on the lake) to the state minimum standard of 100ft. We then predict changes in bluegill growth rates given the new landscape.

Results/Conclusions

The subdivision/new parcel model predicted 98% of all subdivisions and 96% of all new parcels from1974-1998 correctly. In the counterfactual simulation, the number of subdivisions increases by over 20% and the total number of new lots increases by nearly 50% over the factual simulations. Using the simulated counterfactual landscape to generate housing density values, the bluegill growth rate model predicted bluegill growth rates to be 0-10% lower for the counterfactual model, compared to the original levels. This suggests that a change to state-mandated minimum frontage zoning would comprise bluegill growth substantially. Our approach highlights the virtues of a linked economic-ecological framework, where human actions are modeled on a landscape, and the results of these actions are quantified in ecological terms.