h i g h l i g h t sWe observed 1:1 M conversion of DMS to DMSO in estuarine waters. This suggests that DMS photo-oxidation occurred via the CDOM sensitised 1 O 2 pathway. Photochemical rate constants decreased~10-fold from river to seawater. Rate constants were strongly correlated with CDOM absorption coefficients (a 350 ). a 350 -normalised rate constants increased~10-fold from river to seawater. a r t i c l e i n f o a b s t r a c tDimethylsulphide (DMS) photo-oxidation and dimethylsulphoxide (DMSO) photoproduction were estimated in 26 laboratory irradiations of coastal samples from NE England (Tyne estuary) and W Scotland (Loch Linnhe and River Nant at Taynuilt). Pseudo-first order rate constants of DMS photo-oxidation (0.038 h À1 to 0.345 h À1 ) and DMSO photo-production (0.017 h À1 to 0.283 h À1 ) varied by one order of magnitude and were lowest in the coastal North Sea. Estuarine samples (salinity S < 30) had a mean DMSO yield of 96 ± 16% (n ¼ 14), consistent with 1:1 M conversion via photosensitised oxidation by singlet oxygen. Photochemical rate constants were strongly correlated with coloured dissolved organic matter (CDOM) absorption coefficients at 350 nm, a 350 . Variations in a 350 explained 61% (R 2 ¼ 0.61, n ¼ 26) and 73% (R 2 ¼ 0.73, n ¼ 17) of the variability in DMS photo-oxidation and DMSO production, respectively. However, CDOM normalised photochemical rate constants increased strongly towards coastal waters exhibiting lowest CDOM absorbance, indicating water samples of marine character (S > 30) to be most reactive with respect to DMS photo-oxidation. Estimates of water column averaged DMS photo-oxidation rate constants, obtained by scaling to mean daily irradiance (July, NE England) and mid-UV underwater irradiance, were 0.012 d À1 , 0.019 d À1 , and 0.017 d À1 for upper estuary (S < 20), lower estuary (20 < S < 30) and coastal waters (S > 30), at the lower end of previous observations. Comparing our water column averaged DMS photo-oxidation rate constants with estimated DMS losses via air-sea gas exchange and previously reported biological consumption implies that DMS photochemical removal is of only minor importance in our study area.