Previous Efforts of Modeling Recreational Boat Traffic

  The literature on recreational boat traffic modeling is either very sparse or very difficult to find. Only a handful of studies were located that made reference to a boat traffic model or simulation. One study (Boating Research Center, 1991) has already been mentioned in section 1.2.2. That study did produce some spatially explicit cartographic output of simulated boat traffic. However,the description of the simulation's assumptions and implementation are very sparse, and replication of that effort is impossible with their given description.

There are examples of boat traffic simulations that take temporal characteristics into account. Berg-Andreassen and Prokopowicz (1992) used queuing theory (QT) to stochastically model arrival and residence time of boats in lower Mississippi River anchorages, including recreational boats. That study was presented as a development of the theoretical framework of a QT model to investigate (among other things) conflicts between recreational and commercial boat traffic. Dehghani et al. (1993) used a queuing based model to explore the effects of bridge opening timing on both marine and automobile traffic for a bridge in Fort Lauderdale, and Bandy (1991) used a queuing model translated into SLAMM II computer code to examine the effects of a proposed boat lift on boat traffic in a lock separating waters from Lake Michigan and a river in northern Indiana. This is a common enough method for modeling commercial marine traffic and potential uses of port facilities. Even though these studies were rather limited on spatial aspects of their simulation, in principle it would not be too difficult to create a spatially explicit version of this model using the current state of the art in computer technology.

Boudreau et al. (1992) adapted the ``level of service'' (LOS) concept widely used by traffic engineers to model demand by recreational boats for slip facilities in Marina del Rey, California. This was done to explore the feasibility of adding new slips to an existing facility to mitigate excessive demand for services in that area. This model concentrated on available facilities, and treated demand (boats arriving to the marina) as an external factor.

In two reports that were part of a series on recreational boat use in North Carolina, Hammon et al. (1974) developed a flowchart-like model of recreational boat activity that strongly resembles a Forrester Diagram, and Cordell et al. (1975) presented the methodology of how that model was implemented in a specific lake in their state. One problem cited in their study(s) was the difficulty of capturing the subtleties of complex interactions between individuals in a recreational environment using models that aggregate data to such a degree as do Forrester models.