TREE SPECIES COMPOSITION ACROSS TWO STREAMS AT TOTOKET MOUNTAIN, NORTH BRANFORD, CT

 

Page prepared by: Sally Atkins, Todd Forrest, Jamie James, Peter Price-Thomas, and Anne St. John

 

Introduction

This research was undertaken as part of Dr. Tom Siccama’s course YFES 519b Methods of Ecosystem Analysis. Research was conducted in February-April 1998 at Totoket Mountain, located in North Branford, CT. The property is both owned and managed by the Regional Water Authority. The greater research area is characterized by mixed hardwood forests with some hemlock stands, and possesses both year-round and seasonally flowing streams.
 

Objectives

The purpose of our research was to examine the composition of tree species along a stream edge. We were interested in finding out how the species composition changed as one moves away from the stream. Additionally, we analyzed the difference in species composition along the stream as compared to the north-south and east-west transects laid down by the class.
 

Research Site

We chose to examine two streams for our project. Stream 1 runs down a southeast facing slope; this stream is very shallow in most places and runs over and through large amounts of exposed rock. The overall area of Stream 1 could be characterized as xeric and the forest floor is very rocky. In addition, the sparse canopy here permits grass to cover the forest floor. Stream 2 is located at the bottom of a steep, northwest facing slope; this stream runs from north to south. The west side of this stream is a relatively large floodplain area, characterized by many water-loving herbaceous species, while the steep slope on the east side is a large hemlock stand of questionable health. Stream 2 is both wider and deeper than Stream 1, and it clearly flows year-round.
 

Methodology

We began by laying a 30 meter measuring tape perpendicularly across the stream. Placing the 15 m mark in the middle of the stream, we traveled along the line 15 m in each direction. Moving out from the center of the stream, we measured every tree within 3 m on either side of the line having diameter at breast height (dbh) greater than 4.6 cm. In addition to dbh, we recorded the species, vigor, location along the 30 m tape (using the middle of the stream as 0), and stream width. Moving up the stream, we placed plots (30 m by 6 m) approximately 10 m apart. The methodology was identical for both Stream 1 and Stream 2.


 

Results and Discussion

MACRO-SCALE EFFECTS:

At the macro-level, when data from Stream1 and Stream 2 were combined and compared to the class transect data, it is evident that hemlock (TSCA) and yellow birch (BEAL) were significant species.  However, this is because the transect on Stream 2 went through a hemlock stand, with numerous yellow birch on the stream’s western floodplain. The Stream 2 data accounts for the relatively large difference between our streams and the class’ transect data. In addition, the large proportion of yellow birch and hemlock on Stream 2 may be attributed to the fact that this stream is wider and deeper than Stream 1; the transects on Stream 1 are dominated by sugar maple (ACSA).


 
 


 


MICRO-SCALE EFFECTS:

At the micro-level on Stream 1, there was little difference in the species composition, regardless of distance from the middle of the stream.  For Stream 2, hemlock and sugar maple are more abundant further away from the stream center. The increase in hemlock is clearly due to the fact that our transects cross into a relatively large hemlock stand. The increase in sugar maple corresponding to greater distance from the stream center may be attributed to the fact that this species prefers more xeric locations.


 
 
 


 
 Species Area Curves

We also examined the relationship between the number of plots assessed and the number of species recorded and compared the numbers for the stream plots with the numbers from the main transect.

Here are the curves for the stream transects:

 

and here are the curves for the uphill and contour transects:


 

It took ten stream plots to find 15 species while it took only about 5 of the "dryland" plots to reach the same number.

Recommendations for Future Research

A larger sample size would be more readily comparable with the overall forest data set, which would allow statistical comparison. Also, in future study, it would be advisable to use more than 10 m distance between plots so that they do not overlap (as some of our plots did).
 

Go to YFES 519b main page