Impact of pedestrian traffic on UW Prexy's Pasture

-- James Burford, Shay Horton, Thad Reger, Julia Vold

Do you ever wonder how much damage you are doing by walking on the grass instead of a path? Our experiment showcased how much effect humans actually had on the grass in Prexy’s pasture (Figure below).  We decided that testing the phenology of both severely disturbed and minimally disturbed grass areas would be an ideal experiment.

In testing our samples we collected and measured spectral reflectance of both the severely disturbed and minimally disturbed grass samples. We were able to take 8 sets of data from September 23rd, 2013 to October 17th, 2013 from this data we were able to infer that minimally disturbed grass does not lose its vigor as fast as severely disturbed grass does. Our hypothesis was confirmed in that minimally disturbed grass displayed higher spectral reflectance values than severely disturbed grass.

We made a graph based on spectral reflectance for each grass type as well as an NDVI graph for each (Figure 2). NDVI stands for Normalized Difference Vegetation Index and is calculated using the following formula: (NIR - RED)/(NIR + RED). NIR and RED are the near infrared and red wavelength values as read by the spectrometer.

The minimally disturbed grass showed very little change in NDVI over time (Figure 2A), whereas the severely disturbed grass displayed decreasing values (Figure 2B) that coincided with the temperature changes seen throughout the experiment. We fitted trend lines through the NDVI values and the difference between severely disturbed and minimally disturbed grasses based on their NDVI values. The linear graphs of NDVI values are shown below. Based on these graphs we can see that the rate of NDVI decline was higher in severely disturbed grass in comparison to the minimally disturbed grass.

Figure 2: NDVI values of minimally disturbed (A) and severely disturbed (B) grasses in Prexy’s Pasture

A source of error that could have caused outliers in our data sets could be from dirt that clung to the roots of the samples that we took.  This project went smoothly, and was very informative to all of our group members. We saw results close to what we expected to when we began the project in late October.