MERIS chlorophyll-a observations for improved costal monitoring
Harvey, T1; Kratzer, S1; Philipson, P2
1SU, SWEDEN; 2BG, SWEDEN

The Baltic Sea is one of the most polluted and eutrophicated seas in the world and therefore in the need of effective ways of monitoring the changes in water quality. Little has been published about validation of coastal monitoring systems based on satellite data. The aim of this study is to evaluate the use of operational satellite data in comparison to ship-based coastal monitoring data by analysing monthly means of chl-a concentrations measured during the productive season of 2008 and 2010. Full resolution (300 m x 300 m) images from the ENVISAT / MERIS (Medium Resolution Imaging Spectrophotometer) sensor were processed with the FUB Case-2 water properties processor and were corrected for adjacency effects with ICOL (Improved Contrast between Ocean and Land) processor. Remote sensing is an important tool for monitoring, as the same methods are applied on a basin-wide scale. Our results show that the satellite-based monitoring is comparable to the ship-based monitoring, and is better in capturing spatial dynamics. The increased amount of data (when adding the observations from MERIS) results in an 1.5-fold time's higher frequency of observations compared to using only ship-based monitoring. The difference between the two methods regarding the chl-a concentrations was analysed by Spearman's rank correlation and a Wilcoxon signed rank test, comparing the pooled monthly means. The correlation between measurements close in time (0-3 days) was significant (p-value < 1.49 e-05), with a correlation coefficient rs= 0.60. The Wilcoxon signed rank test shows no significant differences between the monthly means of the two methods, p-value = 0.881. The improved data frequency and the spatial coverage by remote sensing will be important for future work on the status classification of the water basins according to European Council's Water Framework Directives.