Experimental Design Lab 1 - Field Data

Design Preparation

To observe the effects of habitat type, and functional diversity on species richness at York University, a total of six locations were chosen. Three of which were grasslands and three of which were woodlots. The six locations chosen were: Stong Grassland, Danby Grassland, Boynton Grassland, Boyer Woodlot, Danby Woodlot and Boynton Woodlot. Each location was divided into a grid where all the quadrats were numbered. The size of each quadrat was 5 m x 5 m and every location had three quadrats that were randomly selected (with a random number generator) for every data collection period. The study occurred weekly over a course of 3 weeks and each field data collection period was consistently at the same three hour time period every week (2:30 pm – 5:30 pm). To gather data from all six locations within the three hour period, the group was divided into two pairs and both pairs collected data from three locations (out of six), both of which were randomly selected before each data collection period. Each pair carried brought with them 1-2 field guides and 1-2 transect measures and for every location, the pair had to count using steps or the transect measures to go to the selected quadrat.

Data Collection

To measure species richness and abundance, each pair formed a 5 m x 5 m quadrat using 1 or 2 of the transect measures taken and recorded the species that were observed as well as the number of each observed species within the quadrat. The data was recorded on a data sheet that was prepared beforehand. This process was repeated three times at each of the three locations. For those species that could not be identified, photographic evidence and/or a description of the unknown species was taken for later identification. Once all the species were identified, the functional group (forbs, shrubs, trees, grasses and herbs) for every observed species were identified and recorded.

Statistical Methods

To determine if the hypothesis and predications were supported or rejected, two statistical methods were used: t-test and correlation test. A t-test was used on the average species richness of all the grassland and woodlot locations to calculate the p-value. The p-value indicated if our results had been statistically significant (p-value < 0.05) or not (p-value > 0.05) i.e. it helped us determine if our results supported our prediction that grasslands had more species richness than woodlots. A correlation test was used to test if there was a correlation between species richness and functional groups. The test also indicated what type of correlation i.e. it indicated whether there was a positive correlation (0 < correlation coefficient < +1), or a negative correlation (-1 < correlation coefficient < 0) or no relationship at all (correlation coefficient equals approximately 0). So this test helped determine if number of tree species affected species richness negatively or if at all.