12/11/2018 - Laserfiche WebLink

Home Browse Search

4.3 RESULTS Both the FDEP IWR Run 55 dataset and the SJRWMD data were queried and combined into a single dataset to be used for analysis. 5 TASK 3 - TREND ANALYSIS 5.1 METHODS The validated data gathered under Task 1 for the WBIDs surrounding BCL provided the source data for the trends analysis. The WBIDs included: ■ Blue Cypress Lake (2893V) ■ Blue Cypress Marsh (28938) ■ Padgett Branch (Class I) (3152C) ■ Blue Cypress Creek (3133) ■ Blue Cypress Lake Drain (2893V1) ■ Padgett Branch (3152B) Within each of these WBIDs several agencies sampled numerous sites at different times over the period of record. SJRWMD has been sampling several sites continually for long enough that trends could be determined at those sites. Figure 2 shows the locations of the WBIDs and the stations selected for trend analysis. 5.2 ANALYTICAL METHODS The seasonal Kendall -Tau trend test was used to determine whether significant temporal trends were detected in the data. The seasonal Kendall Tau is a nonparametric test that is useful for detecting trends in environmental and water quality data when the underlying distribution is not known (Gilbert, 1987; Loftis et a/., 1989; Helsel and Hirsch, 1992). This test has also been applied to trend detection for water quality variables in numerous studies. This first step of the seasonal Kendall Tau test is to develop a time series plot of the raw data for the period of record being analyzed (Figure 3). The time series is plotted with the predicted values to give a sense of the overall trend of the data. As can be seen in the example plot in Figure 3, an increasing trend is apparent in the data being evaluated during the period of record. In the second step of the analysis, the intra -annual variability in the data is explored by plotting the monthly univariate statistics in the form of a box -and -whisker plot (Figure 4 provides an example). The box represents the inter -quartile range, while the top and bottom of the whiskers represent the maximum and minimum values, respectively. Also included in the plot is the mean represented by a red dot. The notches in the boxes represent the 95 percent confidence interval of the medians. If the confidence limits around the medians for any pair do not overlap, the medians of that pair are significantly different at the oc = 0.05 level. 08620-002-01 4 September 2018 Task 1 — Data Collection and Task 3 — Trend Analysis