The storm also caused huge areas of deposition – in areas that were previously water – of sand and gravel. There were noticeably fewer Darter fish in the stream this week. I fear that the shifting silt and sand buried many of these fish as they were hiding beneath the rocks during the storm.
Here is a "Before" and "After" film clip which shows how much Turkey Branch's morpology has changed as a result of this Summer's storms:
Success in Removing Invasive Plants and Restoring Native Species
As in the past two weeks I selected a number of areas, measuring 3 feet by 3 feet, in places that I weeded 1-8 weeks ago. As in the earlier weeks, I uprooted and counted every invasive plant. I also counted the native plants that were left in the ground. I selected areas that I weeded from between 1 week and 8 weeks earlier, comparing the average number of plants/square yard with my control [non-weeded areas'] average.
I used the Student t Test:
to determine if the difference in the means of the study and areas the control areas is statistically significant.
X1 = the average number of invasive plants/square yard in areas I have not weeded
X2 = the average number of invasive plants/square yard in areas I weeded
s1 is that variance of untreated areas and s2 the variance of the weeded areas and n1 and n2 are the number of samples.
For my anaylsis, I started with my "control" group which consists of 20 areas [3 feet by 3 feet] which I did not weed. I selected these areas by selecting the squares, every 10 feet, in a cross section beginning at the stream and ending at the trail. I recorded the various invasive plants that I found in each sample area and calculated "TOTAL Invasives". I calculated an average number of invasive plants/square yard [180.7] by adding the "TOTAL Invasives" for each area and dividing that number by the number of areas [20].
I subtracted the average from each of the "TOTAL Invasive" sample values (x - mean) to calculate the variance. I then squared each variance and added all of the squared variances and divided this number by the number of samples [20].
I made the same calculations [for the mean, variance and variance squared] for the areas that I weeded - these areas were classified according to how many weeks ago they were weeded. I subtracted the sample mean from the control mean and divided that value by the square root of the sum the squared variances of the "control" and sample areas. This t value is compared to the value in the t Test Table, according to the "degrees of freedom" [sum of the number of "control" plus sample areas minus 2] at the 99% confidence interval.
My Null Hypothesis is that there is no difference between the mean number of invasive plants/square yard of the unweeded ["control" areas] and the weeded areas.
My Research Hypothesis is that there is a difference between these means. If my calculated t value is greater than the t Test Table, I accept the Research Hypothesis and reject the Null Hypothesis.
Here is the "Control" samples for Native Plants/Square Yard:
7-8 Weeks Since Last Weeded
At the 99% confidence interval, with 38 degrees of freedom ((n1+n2) – 2) the table value is 2.7, while my t-test value is 3.3 which indicates that the average number of invasive plants/square yard are different [significantly less] - statistically - in areas that were weeded than in the control area.
There appears to be no statistically significant difference between the numbers of native plants per square yard between weeded and control areas.
4-6 Weeks Since Last Weeded
At the 99% confidence interval, with 39 degrees of freedom ((n1+n2) – 2) the table value is 2.7, while my t-test value is 3.3 which indicates that the average number of invasive plants/square yard are different [significantly less] - statistically - in areas that were weeded than in the control area.
There appears to be no statistically significant difference between the numbers of native plants per square yard between weeded and control areas.
1-3 Weeks Since Last Weeded
At the 99% confidence interval, with 33 degrees of freedom ((n1+n2) – 2) the table value is 2.75, while my t-test value is 3.5 which indicates that the average number of invasive plants/square yard are different [significantly less] - statistically - in areas that were weeded than in the control area.
In spite of the fact that weeded does not appear to encourage the growth and re-growth of native plants, some bright spots can be found, such as this Western Sunflower [Helianthus occidentalis]:
A Graphical Analysis of the Effectiveness of Weeding Invasive Plants
It is clear that the number of invasive plants per square yard increases as the time since the area was last weeded becomes greater.
Here is a scatter-graph showing no relationship between native plants/square yard and weeks since last weeded:
Stream Cleanup
The storm also swept an unusual amount of litter [about 80 pounds] into the stream including a shopping cart from a store nearly two miles away!
No comments:
Post a Comment