A brief overview of the Stink Bug- the Green stink bug, Acrosternum hilare is a member of the order Hemiptera and as stated in early lab blogs they have piercing/sucking mouthparts. They are pests on a variety of plants such as cotton, soybean and corn when full grown the Stink bugs are nearly 18mm, they go through simple metamorphosis.
For this weeks lab we went to the Ashland Bottoms Research Farm. Where we performed sampling for Stink bugs in a production soybean field using a sweep net. We were asked to take fifteen samples (ten sweeps per sample) and record the densities we found of both the adult and nymph Stink bugs in each sample. As well as record the amount of time it took to collect each sample. From the data collected we were then to calculate the optimum number of samples needed to estimate the stink bug densities at 95 and 99% confidence intervals for the field. The Calculations are shown below.
When interpreting the values I had for my sweep net samples I used this website- http://easycalculation.com/statistics/standard-deviation.php This website does a very good job at simplifying the data, and all that needs to be inserted is the amount of stink bugs that were caught in each sample. The calculator does the rest.
As you can see from the above page I posted, that for a 95% confidence interval 5719 sample would need to be taken. For a 99% confidence interval 321,701 samples need to be taken to achieve this accuracy with the data set I sampled. Please note that theses numbers are not realistic to perform and that my dataset is somewhat skewed with large values like the 24 stink bugs I recorded for sample 5. This variability increases the number of samples required for accuracy and the time it would take, for the fifteen samples that I took it took over 20 minutes, so it would take much longer to obtain 321,701 samples. When obtaining my samples I performed a siz-zag pattern making a horseshoe thru the majority of the field. The sampling pattern matters because I want to be representative of the whole field not just a small area, to know if there is enough bugs present to justify treatment or not. The amount of time presence/absence sampling could save is a good amount. I didn’t have an actual stopwatch out in the field last week and had to rely on my cell phone to record the times. So it is hard for me to put an exact time amount on it, but presence/absence would be faster.
Overall this lab was more fun than the last couple weeks; we finally got out into the fields and were able to do some real scouting which I enjoy. The topic covered today and the lab itself showed how sampling efficiency, accuracy and precision were all related to taking samples in the field. Being efficient was important because the van rolled at 4pm and we had to be done by then. Accuracy and precision are necessary to establishing the actual densities that are present. My data had some outliers and thus my confidence intervals were skewed but having consistent data would have helped to nail down an accurate population.
For this weeks lab we went to the Ashland Bottoms Research Farm. Where we performed sampling for Stink bugs in a production soybean field using a sweep net. We were asked to take fifteen samples (ten sweeps per sample) and record the densities we found of both the adult and nymph Stink bugs in each sample. As well as record the amount of time it took to collect each sample. From the data collected we were then to calculate the optimum number of samples needed to estimate the stink bug densities at 95 and 99% confidence intervals for the field. The Calculations are shown below.
When interpreting the values I had for my sweep net samples I used this website- http://easycalculation.com/statistics/standard-deviation.php This website does a very good job at simplifying the data, and all that needs to be inserted is the amount of stink bugs that were caught in each sample. The calculator does the rest.
As you can see from the above page I posted, that for a 95% confidence interval 5719 sample would need to be taken. For a 99% confidence interval 321,701 samples need to be taken to achieve this accuracy with the data set I sampled. Please note that theses numbers are not realistic to perform and that my dataset is somewhat skewed with large values like the 24 stink bugs I recorded for sample 5. This variability increases the number of samples required for accuracy and the time it would take, for the fifteen samples that I took it took over 20 minutes, so it would take much longer to obtain 321,701 samples. When obtaining my samples I performed a siz-zag pattern making a horseshoe thru the majority of the field. The sampling pattern matters because I want to be representative of the whole field not just a small area, to know if there is enough bugs present to justify treatment or not. The amount of time presence/absence sampling could save is a good amount. I didn’t have an actual stopwatch out in the field last week and had to rely on my cell phone to record the times. So it is hard for me to put an exact time amount on it, but presence/absence would be faster.
Overall this lab was more fun than the last couple weeks; we finally got out into the fields and were able to do some real scouting which I enjoy. The topic covered today and the lab itself showed how sampling efficiency, accuracy and precision were all related to taking samples in the field. Being efficient was important because the van rolled at 4pm and we had to be done by then. Accuracy and precision are necessary to establishing the actual densities that are present. My data had some outliers and thus my confidence intervals were skewed but having consistent data would have helped to nail down an accurate population.
http://langlade.uwex.edu/ag/images/green_stink_bug_nymph_000.jpg
