Forest Clearing in the Gray's River Watershed, 1905-1996

This study explores the patterns of land clearing in the 80,000 acre Gray's River watershed in southwest Washington between the years 1905 and 1996. The exploration consists of three parts. First, a description is presented of how the natural forest ecosystem functions, including succession, the role of old-growth and second-growth in the forest ecosystem, and natural and man-made disturbances to natural growth cycles.

Second, this study presents original data documenting the patterns of forest cutting within the Gray's River watershed. In 1905 there were an estimated 65,000 acres of old-growth forest, but by 1996 only 1,476 acres, 2% of the old-growth remained in the watershed. Harvest levels reached a climax in the late 1970's when, on average, 2,258 acres of forest were removed each year.

Third, this study assesses the current and future forest by considering three different likely scenarios for timber harvesting. This work concludes that disruptions of the forest ecology and the healthy functioning of the watershed are likely to be permanent conditions in the Gray's River watershed.

Comparison Between Snapshot & Composite Change Data

This study evaluates how spatial and temporal data obtained from historic map evidence function in two change detection applications that focus on the snapshot and composite data models.

The snapshot method is not conducive since it is only temporal coincidence that is being queried, not causation as defined in the composite model with a temporal topology that defines begin and end events for database states. The capacity of the snapshot model is extenuated with the use of temporal joining. This study demonstrates the use of temporal joining with an example describing changes to surface water in Portland, Oregon between 1880 and 2000.

The composite method integrates time data and space data into a database that functions with various object states separated by events. The example demonstrating this capability is the Gray's River watershed forest cover. Even though the composite model demonstrated uses cartographic time in a static rollback form, it is useful for determining how events changed object states based on map evidence.