Master of Environmental Studies Capstone Projects

Document Type

Thesis or dissertation

Date of this Version



Presented to the Faculties of the University of Pennsylvania in Partial Fulfillment of the Requirements for the Degree of Master of Environmental Studies 2011.


The Unionville Serpentine Barrens are a part of Natural Lands Trust’s ChesLen Preserve in Unionville, Chester County, Pennsylvania. This area of the Preserve is underlain with serpentine rock (serpentinite), known for its high magnesium and low calcium content, which inhibit growth in most plants. However, certain plants are tolerant of these conditions, and a very few have evolved a set of traits that confine them almost exclusively to this soil.1 The Unionville Barrens exist today in large part because of disturbances such as corundum mining, wildfires, and grazing that have prohibited plant succession from grassland to forest. The Unionville Barrens are home to at least 15 plant species that are categorized as endangered, threatened, or rare.2 The inspection of a series of aerial photographs taken from 1937 to 2002 clearly shows the decline in area of serpentine grassland from 58.1 acres in 1937 to just 8.9 acres in 2002. In order to sustain the rare plant communities, the grassland must be conserved and, where possible, forested areas converted back to grassland. In order to prioritize grassland restoration areas to make the most efficient use of limited land management resources, it is necessary first to understand if, and how, the rate of succession from grassland to forest is associated with soil conditions or other measurable features of the landscape. This research analyzes the status of current plant communities and their associated soil depths and other characteristics as they relate to the aerial photographs. A grid of 105 points placed 65 meters apart was overlain on top of the Unionville Barrens restoration study area. Soil horizon depth and color, soil depth to bedrock, and surrounding plant communities were recorded at each of these points. Statistical and spatial analyses of these results determined that certain variables best retrospectively “predicted” the likelihood of a quick succession from grassland to woods, when examining this data against a chronosequence of aerial photographs from 1937 to 2010. It was assumed that areas that underwent rapid succession from grassland to woods would require too costly a disturbance regime to sustain the grasslands long-term. Therefore, the restoration priority areas will be selected based on areas where succession must have occurred recently, or where measurements indicated a slow succession rate.



Date Posted: 15 November 2011