Category: Post 4: Sampling Strategies

Below are the results from the sampling strategies:

The most accurate sampling technique for the most common species, the Eastern Hemlock, was the random sampling. The error percentage is roughly half that of the other techniques used. For the rarest species, the white pine, haphazard was the most accurate. The most efficient sampling technique in terms of time was systemic which took 12 hours and 34 minutes to complete. The others were not far behind that with random taking 12 hours and 48 minutes and haphazard taking 13 hours and 8 minutes to complete. My data showed that the abundance of trees, in general, impacted the accuracy of all methods. The less dense the trees, the more likely it is to have variations in the data. Twenty four points of sampling is not enough for this location. The trees spread out in a such a way that the results can be misleading. One area may have way more of one type of tree then other locations. To get a more accurate reading more sampling is needed and spread out in a way that all types of trees can be accounted for.

The results of the three sampling strategies used in the virtual forest tutorial.

Based on the time spent sampling, the Distance Random sampling was the most efficient with 4 hours and 43 minutes, Area Systematic sampling took 12 hours and 37 minutes and Haphazard sampling took 12 hours and 48 minutes. For the common species, the most accurate sampling strategy was the Systematic sampling. The most accurate for rare species was the Random sampling strategy. The accuracy declined for the rare species for the haphazard sampling strategy. 24 was not a sufficient number of sample points to capture the number of species in this community.  No it was not enough sample points to accurately estimate the abundance of these species. The accuracy for the haphazard sampling strategy changed for the common species and rare species. The systematic sampling strategy was more accurate compared to the 2 other strategies.

Below are the results of the 3 sampling strategies used in the virtual forest tutorial:

1. Based on the information provided, the fastest estimated sampling time was the random sampling method estimated at 12 hours and 19 minutes.
2. The 2 most common species are the Eastern Hemlock and Sweet Birch. As you can see from the data presented above, the random sampling method yielded the lowest percentage error. For the 2 rarest species – Striped maple and White pine –  the random sampling method yielded the lowest percentage error for only 1 of them (Striped maple), whereas the White pine’s lowest percentage error was in the hap hazardous sampling method. The accuracy did seem to change with species abundance generally speaking.
3. After taking the mean percentage error of each sampling strategy these were my results – Systematic (44.5%), Random (29.1%) and Hap Hazardous (26.5%). This would be indicative that the Hap Hazardous Sampling Strategy is the most accurate out of all 3 strategies.

Virtual Forest Tutorial: Snyder- Middleswarth Natural Area.

Time Spent Sampling:

Of the three sampling strategies, systematic, random, and haphazard, they were all very similar in the amount of time spent sampling, with haphazard being slower than the other two by 2 minutes.

Accuracy of each Method of Sampling:

The most accurate sampling method for the more common tree species was the systematic sampling method.  The most accurate sampling method for the rare species, Striped Maple and White Pine was the haphazard method. The accuracy for the systematic and random method declined for the rare species compared to common, due to the fact that they missed one rare species all together each. Whereas the haphazard method stayed about the same for both common and rare species. Perhaps with more sample points and another location for systematic would have been more accurate.

In the Virtual Forest tutorial, using Distance-based sampling methods:

Table 1. Comparison of three distance-based sampling strategies; actual, random, and haphazard used to calculate the abundance of seven tree species (Eastern Hemlock, Sweet Birch, Yellow Birch, Chestnut Oak, Red Maple, Striped Maple, and White Pine) in the Snyder- Middleswarth Natural Area.

Based on findings from Table 1:

Which technique had the fastest estimated sampling time?

The times between the 3 sampling strategies were somewhat similar, with the fastest strategy by ~20 minutes being the systematic approach in regards to sampling 24 plots.

Comparing the % error of the different strategies for the two most common and two rarest species. For each species the most accurate strategy was as follows:

• Eastern Hemlock (most common species) most accurate strategy:   Haphazard 

• • It surpasses the random strategy with an % error of 3.21.
  • Than random & haphazard strategies were more accurate than the systematic, as the systematic had an % error of 40.99.

• Red Maple (second most abundant species) most accurate strategy: Systematic 

• • This strategy had a 24.05% error.

• White Pine (rarest species) most accurate strategy: Systematic .
  • Even though it had a 53.57% error, it is far more accurate than the random & haphazard strategies with % errors of 100, and 122.6, respectively.
• Striped Maple (second least abundant species) most accurate strategy: Haphazard.

• • This strategy had a 6.86 % error which is very low, especially when compared to 64.57% and 157.71% errors for Random and Systematic, respectively.

Did the accuracy change with species abundance?

When comparing all three strategies regardless of tree species, accuracy decreased as species abundance lowered. Therefore, I recommend that more samples than 24 need to be taken to increase accuracy. The majority of % error was inversely proportional to the actual species abundance (save for a few data points), regardless of the sampling strategy.

Was one sampling strategy more accurate than another?

Haphazard strategy had an average % error of 28.96, which is lower than that for Random & Systematic (36.46% & 53.39%, respectively) – there by being the most accurate sampling strategy in a forest terrain. The random strategy being the least accurate.

% error of area densities for systematic, random and haphazard sampling types in the Synder-Middlesworth Natural Area:

Systematic common species % error:

Eastern Hemlock: 20% error

Sweet Birch: 9.8% error

Systematic rare species % error:

Striped Maple: 18.9%

White Pine: 48.8%

Random common species % error:

Eastern Hemlock: 31.7%

Sweet Birch: 14.9%

Random rare species % error:

Striped Maple: 281.1%

White Pine: 50%

Haphazard common species % error:

Eastern Hemlock: 43.9%

Sweet Birch: 53.2%

Haphazard rare species % error:

Striped Maple: 151.4%

White Pine: 100%

The systematic sampling had the lowest sample time with 12hrs 4mins. It was also the most accurate for both common and rare species. Next was random sampling that would take approximately 12hrs 53mins. The slowest and least accurate was the haphazard sampling method at 13hrs 11mins. The results overall seemed to be more accurate throughout all three sampling techniques in the more abundant species such as Eastern Hemlock and Sweet Birch. As well, the overall most accurate sampling method appears to be the systematic method.