Environ Sci Pollut Res 
£) Springer 
Content courtesy of Springer Nature, terms of use apply. Rights reserved. 
Fig. 7 Time trend of selected mircopollutants in the Baltic Sea from 2001 to 2014. a Perfluoroalkyl substances, b Triazine herbicides, c Urea herbicides, 
d Phenoxyherbicides. Data: see Tab. SI 1, S12 
slight upward trends can be observed at most stations. TERB 
is possibly replacing ATR and SIM, although its concentra 
tions are much lower. Yet, its concentration range (0.2-5.2 ng/ 
L) is in agreement with literature values (not detected-11 ng/ 
L) (Bester and Huhnerfiiss 1993; Pempkowiak et al. 2000; 
Nodler et al. 2013; Orlikowska et al. 2015). IRG shows a 
slight downward trend at the western stations TF010 to 
TF030, but merely any in the Arkona Basin (TF113 & 
TF109) and is generally decreasing (Fig. 7b, Tab. S11). 
High summer concentrations have been observed at the most 
western stations KB2 and MB3. Unfortunately, only data of 
concentration in summer between 2001 and 2005 are avail 
able, which do not allow a trend detection. However, all IRG 
concentrations are below the maximum concentration of 16 
ng/L listed by the Water Framework Directive since 2013 
(Union 2013). Furthermore, IRG has decreased since 1997, 
where it was detected in the Baltic Sea in a much higher 
concentration range (90—440 ng/L) (Biselli et al. 2000). The 
downward trend of IRG in the Baltic Sea coincides with the 
results of Orlikowska et al. (2015) who did not detect IRG in 
the Baltic Sea in 2014. The same was observed for the 
phenylurea DIU (Fig. 7c, Tab. S11). Due to its elevated con 
centrations in summer and local input sources, no general 
trends can be identified. Generally, slightly elevated 
concentrations for ISO were observed in summer, which 
makes the evaluation less reliable. Both concentrations of 
DIU and ISO are lower than the reported median 
concentration of Nodler et al. (2014) in 2009. All observed 
concentrations did not exceed the environmental quality stan- 
dardof0.2 qg/L and 0.3 qg/L, respectively (Union 2013). The 
concentration of CHLO is uniform at all stations showing no 
trends (Fig. 7c, Tab. S11). 
The phenoxyacetic acid 2,4-D has a high variability at 
some local hot spots and is probably influenced by summer 
inputs (Fig. 7d, Tab. S11). Therefore, no reliable trends can be 
assigned, except for a decreasing tendency at the western sta 
tions. Similarly, DCPP and MCPA show a tendency to slight 
ly decreasing trends, which are interfered by seasonal but not 
local effects (Fig. 7d, Tab. SI 1). The concentration of MCPP 
is below the reported median concentration (7.8 ng/L) by 
Nodler et al. (2014) in 2009. 
METOLA and META show a clear upward trend at all 
stations (Tab. SI 1). Although CHL was detected at high con 
centrations, no trend can be identified, because of its high 
variability and short sampling period (since 2005 only) (Tab. 
Sll). Also, trend evaluations of the pharmaceuticals and the 
complex former BENZTRI are limited by the short monitor 
ing period (since 2008) (Tab. Sll). For CARB, an upward 
trend seems to be indicated. 
As was shown in the “Spatial distribution” section, the 
concentrations are homogenously distributed throughout the 
Baltic Sea, except for the semi-enclosed most western stations