Category: Post 4: Sampling Strategies

Blog Post 4: Sampling Methods

TRU Open Learning Leave a comment

In the virtual forest I chose to use Area Based sampling. No sampling method showed much efficiency over the other in terms of time spent.

The most accurate method overall was the random sampling method. For the most common species random sampling was the most accurate. For the least common species random sampling was somewhat accurate, although random sampling failed to return any samples of the least common tress, striped maple. Accuracy was better for the common species of trees due to lower percentage errors.

I would want to sample more areas in a random fashion to lower % error rates.

Tree Species Actual Density Area Systematic % error Area Random % error Area Haphazard % error
Most Common Eastern Hemlock 469.9 320.0 31.9 341.7 27.3 550.0 17.0
2nd Most Common Red Maple 118.9 84 29.4 137.5 15.6 162.5 36.7
2nd Least Common Chestnut Oak 87.5 36.0 58.9 58.3 33.4 41.7 52.3
Least Common Striped Maple 17.5 52.0 197 0 N/A 29.2 66.9
Time to Sample 12h37m 12h47m 12h31m

Blog post #4 Sampling strategies

TRU Open Learning Leave a comment

After trying out the sample methods tutorial I was able to gain some insight in to the pros and cons of each sample method. Haphazard sampling was the fastest method, but as I could see when comparing accuracy it was low preforming in that area. This makes sense since I chose my sections based on areas with high tree density. The one that took the longest was the systematic survey this seems accurate since I chose to check every 7th grid. This created extra work and more grids were surveyed then the others types.

 

The percent error of the two most common trees based on the three sampling strategies the random sampling was the most accurate, for the two most common trees, and the haphazard was the least accurate.As for the least populous trees only the random sampling actually accounted for both trees in the survey. The haphazard sample picked up neither, and systemic survey only found one of the two.

 

From the result over view it appears that the random sample survey did a more accurate job that the other two types. In terms of hours spent on surveys it was in the middle. This is important to consider when I am  looking at collecting my data for my project. More time invested in to systematic survey with a increased sample size my not give you the most accurate results. On the other hand random samples fall to chance and a large are of the survey grid could be missed. This is something to consider when looking at the randomly chosen quadrants and collecting your data.

 

Haphazard while convenient and more efficient and economical has to be critiqued  for a large bias in selecting quadrants or zones for surveying, or risk an invalid data set and wasted time.

Blog Post 4: Sampling Strategies

TRU Open Learning Leave a comment

In the Virtual Forest tutorial, of the three sampling strategies that I used, the haphazard or subjective sampling technique had the fastest estimated sampling time at 4 hours and 26 minutes, followed by random sampling at 4 hours and 54 minutes and then systematic sampling along a topographic gradient at 12 hours and 37 minutes.

Of the two most common species, Eastern Hemlock  (EH) and Sweet Birch (SB), the systematic sampling had a percentage error of -22% for EH and -14.9% for SB. With random sampling, the percentage error was 2.5% for EH and 21% for SB. With haphazard or subjective sampling, the percentage error was 32% for EH and 80% for SB. Therefore for these two common species, it would appear that random sampling had the lowest percentage error for EH and systematic sampling had the lowest percentage error for SB. If you average the percentage error of the two sample techniques, random sampling presented the least amount of percentage error for the common species of trees.

Of the two rarest species, Striped Maple (SM) and White Pine (WP), the percentage error for the systematic sampling technique was 128% and -100% respectively. Random sampling had a percentage error of 25% for SM and -100% for WP. For haphazard or subjective sampling, the percentage error for SM was 141% and -100%. All three sampling techniques failed to record any occurrences of White Pine trees. Random sampling had the smallest percentage error for SM making it the most effective sampling technique for this rare species.

Accuracy for all species was relatively consistent with the 3 sampling techniques except for the most rare species of White Pine which was undetected in all methods. The most abundant species were more accurate although percentage error was high amongst all recorded species.

Based on the results, all three sampling techniques showed fairly consistent results although random sampling appeared to be the most effective for the most abundant tree type. However, for the Red Maple, a more rare tree, the systematic sampling using a topographic gradient method was the most effective with only a 0.9% percentage error. However, this method took the longest so may not be the most feasible method to use in this case. For the Yellow Birch tree, a more common tree, the haphazard or subjective sampling method had the least percentage error at only 3.7%. This method was the fastest method but it could pose problems of bias amongst the researcher in the field.

Blog Post 4 – Sampling Strategies

TRU Open Learning Leave a comment

The results of the three sampling strategies which are used are systematic, haphazard, and random. In this tutorial systematic/random and haphazard sampling are compaired.

The haphazard sampling had the fastest estimated sampling time because there was less travel time between the samples points.

When looking at the percent error, the overall accuracy of the species which were more common was greater than the species which were less common.

The systematic/random sampling strategy was the most accurate because the areas being analysed do not overlap.

Post 4: Sampling Strategies

TRU Open Learning Leave a comment

Systematic, random, and haphazard sampling strategies were used in a virtual forest tutorial. The systematic sampling had the fastest sampling time (4 hours and 5 minutes), with haphazard being second (4 hours and 30 minutes), and random being the slowest (4 hours and 39 minutes). For the two most common species, Eastern Hemlock and Red Maple, the percentage errors varied. For Eastern Hemlock, the random sampling technique had the lowest percentage error (32.4%), haphazard (32.9%) with the second lowest, and systematic with the highest percentage error (37.5%). For Red Maple, haphazard sampling had the lowest percentage error (19.9%), random sampling had the second lowest (21.9%), and systematic with the highest percentage error (38.3%). The two most rare tree species, Striped Maple (40.0%) and White Pine (94.0%), percentage error was the highest for the systematic sampling strategy. Haphazard sampling had the lowest percentage error for Striped Maple (14.3%), and random sampling had the lowest for White Pine (8.3%). Random sampling for Striped Maple had the second lowest percentage error for Striped Maple (56.0%), and haphazard sampling for White Pine (78.6%). Accuracy decreased as species abundance decreased. the percentage error for the rare species ranged from 8.3-94.0% for White Pine, and 14.3-56.0% for  Striped Maple. For the common species, the percentage error ranged from 32.4-37.5% for Eastern Hemlock, and 5.4-36.2% for Red Maple. Systematic sampling was the least accurate sampling strategy. Both random and haphazard sampling were approximately the same amount of accuracy. If all the seven species sampled were included, random was more accurate 43% of the time as well as haphazard, with systematic being accurate 14% of the time.

Post Four: Sampling Strategies

TRU Open Learning Leave a comment

In the virtual forest I chose to use Distance Based sampling.

Systematic sampling was the most efficient in terms of time spent sampling, but only by about 15 and 30 minutes respectively.

The actual densities varied widely with my estimated data from sampling. Haphazard sampling was the most accurate sampling strategy for common species, with an average error margin of 12.5%. The most accurate for rare species was Systematic sampling, with an average error margin of only 2.3%. My results showed that accuracy was better for more common species, which was not surprising. Although not recorded in the table below, I noticed that systematic sampling showed closer results for the species that were neither common nor rare. Because of wide ranges in error margins, it would be ideal to sample more than 24 points for a more accurate estimate.

 

Tree Species Actual Density Distance Systematic % error Distance Random % error Distance Haphazard % error
Most Common Eastern Hemlock 469.9 516.5 9.9 368.2 21.6 399.3 15
2nd Most Common Sweet Birch 117.5 42.3 64 108.3 7.8 105.9 9.9
Least Common White Pine 8.4 8.5 1.2 21.7 158 0 100
2nd Least Common Striped Maple 17.5 16.9 3.4 32.5 85.7 8.1 53.7
Estimated Time 4h15m 4h29m 4h44m

Blog Post 4: Sampling Strategies

TRU Open Learning Leave a comment

I used the distance-based sampling methods to measure tree species abundance in the tutorial. Table 1 below illustrates the data gathered.

 

Table 1 shows that the estimated times to perform the studies were somewhat similar, but that the fastest method to sample 24 plots was the systematic approach by approximately 20 minutes.

The comparison of measurements of abundance showed that the most precise technique for the most abundant species of tree (Eastern Hemlock) was the haphazard technique. With a % error of only 3.21, it surpasses the random sampling method in accuracy by less than 3%. Both the random and haphazard techniques were far more accurate than the systematic method, which lend a terrible % error of 40.99%. As for the second most abundant species of tree in the area (Red Maple), the best sampling method was systematic, with a % error of 24.05.

For the rarest species of tree (White Pine), the most accurate sampling method was found to be the systematic approach. With a relatively high 53.57% error, it is far more accurate than the two other methods, who have both surpassed 100% error. The abundance of the second least abundant tree (Striped Maple) was very accurately measured by the haphazard method. It scored a 6.86% error, comparatively to 64.57% and 157.71% for random and systematic.

Generally speaking, the accuracy of all three methods seemed to have diminished as species abundance became lower. Except for a few odd data points, the majority of % error was inversely proportional to the actual species abundance, no matter the sampling strategy.

Finally, the average % error of the haphazard method (28.96%) was lower than for random and systematic methods (36.46% and 53.39%). Therefore, I conclude that the distance-based haphazard method is the most accurate for sampling abundance in a forest.

Post 4: Sampling Strategies

TRU Open Learning Leave a comment

Kevin Ostapowich

March 27, 2019

The sampling methods I chose were distance-based.  See the table below for a summary of the results:

Sampling techniques results

The method with the least amount of error is the Haphazard method.  The Random method takes the longest to sample and the Systematic takes the shortest amount of time to sample.

The error is lower for the most common species (eastern hemlock, sweet birch, red maple) across all methods compared to the two rarest species (white pine and striped maple) which have very high errors across all methods.  The more abundant the species, the more accurate any sampling method is.  Given the results from this trial I would probably choose the Haphazard method as it has the lowest errors (in this study) and enables the user more flexibility in choosing sample sites.

Blog Post 4: Sampling Strategies

TRU Open Learning Leave a comment

The systematic sampling technique had the fastest estimated time to sample of 10 hours and 27 minutes. The haphazard sampling technique has the slowest estimated time to sample of 10 hours and 44 minutes. The random sampling technique had an estimated time to sample of 10 hours and 42 minutes. I can conclude that the accuracy decreased completely with a decrease in species abundancy. I would say that random and haphazard are equally accurate and systematic is a little less accurate than those.

 

Percentage Error – Most Common Species:

Random: 4.6% for Red Maple and 20.8% for White Oak

Haphazard: 4.0% for Red Maple and 34.2% for White Oak

Systematic: 5.3% for Red Maple and 14.1% for White Oak

 

Percentage Error – Least Common Species:

Random: 100% for Yellow Birch and White Ash

Haphazard: 100% for Yellow Birch and 525% for White Ash

Systematic: 100% for Yellow Birch and White Ash

Blog Post 4: Sampling Strategies

TRU Open Learning Leave a comment

1. Which technique had the fastest estimated sampling time?

The technique with the fastest sampling times was systematic with 12 hours and 35 minutes.

2. Compare the percentage error of the different strategies for the two most common and two rarest species.

Common / Rare Sampling Technique % Error
Common (Eastern Hemlock) Random 28.6%
Haphazard 20.6%
Systematic 17.4%
Common (Sweet Birch) Random 4.3%
Haphazard 4.3%
Systematic 26.0%
Rare (White Pine) Random 197.6%
Haphazard 48.8%
Systematic 138.1%
Rare (Red Maple) Random 8.9%
Haphazard 19.2%
Systematic 14.4%

3. Was one sampling strategy more accurate than another?

Based on the information presented in the above table, no single sampling technique was more accurate than another.