Time versus flow composite water sampling for regulatory purposes in the Everglades agricultural area

By legislative mandate, all entities discharging agricultural drainage water to area canals in the Everglades Agricultural Area (EAA) of South Florida are required to participate in an extensive total-phosphorus (TP) concentration and load monitoring program. This means that all farms must purchase, install, and operate instruments for measuring pump station discharge volumes and for collecting water samples from the drainage stream. Acceptable methods for both flow measurement and water sampling are varied, and the selection of appropriate methods is often dependent on the characteristics of the sampled water, the types of pumps used, and the end use of the resulting data. In the EAA, drainage water TP concentrations are highly variable, pump stations are unique and difficult to accurately calibrate and monitor, and the end use of the data is for determining regulatory program compliance. Of great interest is the possibility of reducing monitoring costs by using time composite sampling techniques rather than flow compositing to determine drainage water TP concentrations and loads. To address this question,flow and time composite sample data pairs from 820 sampling events over a three-yr period, from 10 farms representative of agricultural conditions in the EAA, were compared. Pump station average operating flow rates ranged from 240 Us to 1860 Us, flow composite sample trigger volumes ranged from 0.92 ML to 7.2 ML and time composite sample intervals were 30 and 60 min. The non-parametric Kolmogorov-Smirnov statistical test showed that the flow and time composite data sets, for both TP concentrations and loads, respectively, were not significantly different. Linear regression analyses yielded flow/time multiplicative factors of 0.99 for TP concentrations and 1.07 for TP loads. However, field and laboratory errors intrinsic to agricultural drainage water sampling and analysis were deemed to be great enough so as to make the application of the 1% and 7% corrections to the data unnecessary. Furthermore, the sample collection interval substantially influenced the time composite event TP concentration and load values, indicating that proper selection of sampling intervals could significantly alter differences beftveen the time and flow composite sampling strategies. Hence, for the purposes of the regulatory monitoring program in the EAA, and constrained by the test conditions above, a simple 1:1 correspondence between flow and time composite water samples is acceptable provided that the time composite sampling period is set between 60 and 180 min.