To assess whether there were temporal trends in water quality measures, the data were divided 
into zones and seasons to minimize biases associated with differences in sampling (spatial and 
temporal). For chlorophyll a , there was insufficient wet season data available, so trend analysis was 
only performed for the dry season. A Mann Kendall trend test was used to test whether there were 
significant trends within a zone and a season. If there were significant seasonal patterns within a zone, 
then the Seasonal Kendall test was used to determine if there was a significant increasing or decreasing 
trend. The Seasonal Kendall test performs the Mann Kendall test for each season and then combines 
the results of these into one overall test for whether there is a consistent monotonic trend over time 
(Helsel et al., 2006). For the Seasonal Kendall test all of the data (within a zone and season) was used. 
For the Mann Kendall trend test, there can only be one observation for each date, so multiple 
observations (either multiple stations or sampling events) on a single day were averaged. For all trend 
tests, p values less than 0.05 were considered significant. In addition to the trend analysis, we divided 
the data into historical and recent groups and tested whether there were significant differences in 
median values using the Mann-Whitney Rank Sum test. 
Mann-Whitney Rank Sum test and Kruskal-Wallis one way ANOVA were performed using 
SigmaStat software package (version 3.5, Systat Software, Inc., San Jose, CA), while trend analysis 
(Mann Kendall and Seasonal Kendall) were performed using a Windows Program written by the U.S. 
Geological Survey (Helsel et al., 2006). 
33 
1 ^ 
