Systematic was the most time-efficient, however, it was only 4 minutes faster than random.
- Systematic= 12 hours 35 minutes
- Random= 12 hours 39 minutes
- Haphazard=13 hours 9 minutes
| Species | Systematic %error | Random %error | Haphazard %error |
| Common 1- Eastern Hemlock | 26.8% | 10.0% | 5.6% |
| Common 2- Red Maple | 9.2% | 15.9% | 0.9% |
| Rare 1- White Pine
|
4.8% | 396.4% | 52.4% |
| Rare 2- Stripped Maple
|
31.4% | 90.3% | 77.1% |
| Average % Error
|
18.1% | 128.2% | 34% |
Figure 1: Percentage (%) error found for different sampling methods for most common and most rare species in virtual Snyder-Middleswarth Natural Area. Data used to perform these calculations (actual and data density) can be seen by clicking the following hyperlink Reid Marriott Sample Data
- For both common species, the haphazard sampling strategy was the most accurate with a % error of 5.6% for Eastern Hemlock and 0.9% for Red Maple.
- For both rare species, the systematic sampling strategy was the most accurate with a % error of 4.8% for White Pine and 31.4% for Stripped Maple.
- The accuracy declined for the rare species except for white pine systematic as that sampling method only had a % error of 4.8%. However, all of the other rare species error % was 31.4% or larger; reaching 396.4% in the case of White Pine random sampling. The accuracy went down when the species were less abundant.
- The Systematic sampling strategy was the most accurate as it had the lowest average % error of 18.1%.
- 24 sample points were able to capture the number of species in this community. However, I do not think that 24 sample points were enough sample points to accurately estimate the abundance of the species found. This is evident by the high error % in the rare species as they were either greatly under or overestimated based on the 24 samples. Furthermore, there was even a 26.8% error with respect to a systematic sampling of the Eastern Hemlock which is the most abundant species in the community; so under the 24 sample conditions not all collection methods were accurate with the most abundant species. I believe that 50 would have been a better amount of samples as this would have allowed for 10 samples to be taken from each of the southern ridge top, north-facing slope, bottomland, south-facing slope and norther ridge top.
All the best,
Reid Marriott (T00716013)