Justifying transect size...

As I combed through the literature looking for suggestions of how to implement my vegetation surveys, I struggled with the size of my transects. Most of the literature did not reveal transect size, but transect area. Also, the spatial scale of much of the research in the literature was greater than I would be able to accomplish in one semester. After talking with Megan, she suggested that I visit the reference site of Slaughter Creek, Bull Creek at St. Edward's Park, and establish a transect size there that will be manageable, but also include an appropriate diversity of environmental factors that normally influence riparian habitat integrity. Then I will normalize this area to my controlled (WQPL) and impacted (upstream and downstream) sites along Slaughter Creek. I think this is a practical solution because the width of the riparian zone, from the stream outward (or upland) will differ at each site. If I were to choose transects with fixed dimensions, I believe that I would not have the wealth of data I would need to appropriately apply statistical analysis on my data.
Below is an image of my reference site Bull Creek at St. Edwards Park that I created from Google Earth: