Post #4

Systematic, random and haphazard sampling techniques were compared in the virtual forest tutorial.

Systematic had the fastest estimated sampling time (12 hours and 7 minutes). Haphazard sampling was second fastest (12 hours 34 minutes) and random was the slowest (12 hours and 42 minutes).

The percent error is summarized in table 1. Systematic and haphazard produced similar percent error values for the two most common species (eastern hemlock and red maple).  Systematic sampling produced errors of 7.3% for eastern hemlock 15.6% for red maple. Random sampling yielded percent error of 1.57% for eastern hemlock and 51% for red maple.

White pine and striped maple were the least common. However, no method sampled either of these trees. The next least common were yellow birch and chestnut oak. Haphazard yielded the smallest percent error (4.3% for yellow birch and 5.13% for chestnut oak). Systematic was second best (14.8% for yellow birch and 15% for chestnut oak). Random had the highest percent error (33.9% for yellow birch and 61.5.0% for chestnut oak).

It appeared that systematic sampling became more slightly more inaccurate as species abundance decreases. Haphazard sampling was more stable, however it randomly had a very high error for sweet birch. Random produced the smallest error in entire tutorial for the most common species (eastern hemlock 1.57%), but produced more inaccurate results for all other less abundant species.

Table 1. Percent error produced by systematic, random and haphazard sampling in a virtual forest tutorial