A handheld XRF-analyzer specially calibrated for measurements of metals on plastic boat hulls has been used on leisure boats in Denmark (DK), Finland (FI), and Germany (DE). The results on tin and copper are presented as μgmetal/cm . Tin is a proxy for the occurrence of organotin compounds on the boat. Two or three sites were visited in each country and between 25 and 90 boats were measured at each site. Every boat was measured at six to eight places, and the results are presented both as mean and median 2 2 values. Linear regression of mean to median values of the 377 data pairs shows high relationship with R =0.9566 for tin and R of 0.9724 for copper and thus both ways of calculation may be used. However, for regulative use, it is suggested that all individual measurements on each boat should be presented and used for decisions of removal or sealing of boat hulls. The results are compared with published data from different parts of Sweden, i.e., boats in fresh water, brackish water, and salt water. The results show that tin with mean values > 50 μgSn/cm is still found on 42, 24, and 23% of the boats in DK, FI, and DE, respectively. The corresponding percentages based on median values are 38, 22, and 18% for DK, FI, and DE, respectively. The variation among boats is high with a maximum mean value of 2000 μgSn/cm . As comparison, one layer of an old TBTantifouling paint Hempels Hard racing 2 2 superior, corresponds to 300 μgSn/cm . The percentage of boats with tin > 400 μgSn/cm content based on mean values was 10% in DK, 5% in FI, and 1% in DE. The corresponding median values were 9, 6, and 1% for DK, FI, and DE. Copper, > 100 μgCu/cm ,was detected on all measured boats in DK and in DE and on all but 3% of the FI boats. One layer of Hempels MilleXtra corresponds to 4000 μgCu/cm . The recommendation on the can is to apply two layers. The proportion of boats with higher mean copper values than 8000 μgCu/cm was 51, 56, and 61 for boats in DK, FI, and DE, respectively. The proportion based on median values > 8000 μgCu/ cm was 50, 54, and 61% for DK, FI, and DE. The conclusion is that many leisure boats around the Baltic Sea still display or possess antifouling paints containing organotin compounds and that more than half of the boats have more copper than needed for one boat season according to the paint producers. Much of these known toxic compounds will probably be released into the environment and harm the biota. The calibrated XRF-method, intended for area measurements on boat hulls, is an easy and cheap way to detect boats with organotin compounds and high copper content. We recommend environmental authorities to use this method for identification of such boats and to use the results for requesting measures to minimize further leakage to the environment. . . . Keywords TBT Copper Concentrations on boat hulls XRF-measurements Introduction The Baltic Sea is a polluted sea with many hazardous sub- One of the substances of high concern is the occurrence stances at elevated levels in both water, sediment, and biota. of organotin compounds where tributyltin (TBT) is the most important substance. In spite of prohibition of TBT since 1989 for use on leisure boats shorter than 25 m Responsible editor: Philippe Garrigues (Council Directive 89/677/EEC) and on ships from 2008 (Gipperth 2009), TBT is still detected in water, * Britta Eklund email@example.com; https://www.aces.su.se biota, and sediments in particular boat harbors and in shipping lanes. High content of TBT is one reason why many sediments do not reach BGood Environmental Department of Environmental Science and Analytical Chemistry, ACES, Stockholm University, Stockholm, Sweden Status^ according to the Marine Strategy Framework Directive (2008/56/EC) adopted by EU in 2008. The Laboratory for Aquatic Research, Limnomar, Hamburg, Germany 14596 Environ Sci Pollut Res (2018) 25:14595–14605 occurrence of organotin compounds is still high in sediments Formerly, the only way to know the amount of toxic sub- around the world (e.g., Hoch 2001; Viglino et al. 2004;Antizar- stances on the boat hull was to scrape off paint and send to a Ladislao 2008;Eklund et al. 2008, 2016; Cornelissen et al. laboratory for chemical analysis. With the calibrated method 2008). Within the European Water Framework Directive, the developed by Ytreberg et al. (2015) intended for area mea- organotin substances are prioritized and should be phased out as surements on plastic boat hulls, the concentration of metals on quickly as possible (Council Directive 2000/60/EC). the boat hulls can be screened. In this method, a handheld X- Nevertheless, chemical and biological monitoring data show ray fluorescence (XRF) analyzer was used for determining the persistence and bioavailability of organotin compounds on a quantity of metals in the antifouling paint on boat hulls. The low but significant level around the Baltic Sea (Eklund et al. advantage with this method is that it is calibrated to give the 2008, 2010, 2016;HELCOM 2009;Strand 2009;Nybergetal. results in μg metals/cm instead of in percent as in other com- 2014). The highest values are found in harbors but even deep monly used XRF methods. The results in metals per area water sediments are considered having Bbad status^ with levels enable comparisons between boats and within the same boat higher than the threshold value of > 50 μg/kg ww (HELCOM at various times. The method is a cost efficient, non-destruc- 2010). One important source of TBT is waste from maintenance tive, and rapid screening method, and for the first time boats work on boat hulls with old coatings on leisure boats (Eklund with tin (TBT) on the hull can be identified. This makes it and Eklund 2014;Eklundet al. 2014). Elevated concentrations possible for environmental authorities to require measures to of TBT in sediments in leisure boat harbors have been observed be taken and by this stop further leakage to the environment. in a number of studies such as by Maguire (2000), Hoch (2001), The method has until now, only been used in one study for andEklundetal. (2008, 2010). The concentrations were much screening of ca 700 boat hulls in Sweden (Ytreberg et al. above a predicted no-effect concentration (PNEC) value for 2016). The highest concentrations of both tin and copper were TBT in sediment of 0.02 μg/kg dw (Anon 2005)and detected on boats moored in salt water compared to the boats Swedish limit value of 1.6 μg TBT/kg dry sediment (Anon in brackish water or fresh water. The question is whether this 2015a). These high concentrations in sediments cannot ex- hold true for a larger region like the Baltic Sea and how this clusively be explained by historical inputs but by persistent complies with regulation in different areas and countries. This leaching out of old TBT paints that is enhanced when hulls is investigated in this study. are treated with high-pressure hosing (Eklund et al. 2008). The backbone polymers of eroding or self-polishing biocidal antifouling paints build no water barrier and hydrolyze at the Aim water contact including penetration by water. Thus an over- laying copper paints with the same matrix property are pen- The main object was to investigate the persistence of tin, on etrated by water and even the under laying old TBT paint boat hulls in Finland, Denmark, and Germany and to compare layer. It can be expected that the leaching rate of TBT from the results with Swedish data. Another aim was to measure the an old paint layer beneath is lower than the top layer but amount of copper on boat hulls in the countries around the trials with high pressure hosing show that even below cop- Baltic Sea and relate to the compliance with existing regula- per paint essential amounts of TBT of old layer can be tion in the countries. Furthermore, the readings delivered data found in the waste water (Ytreberg 2012). Thus leaching on regional and national practice in boat maintenance. of existing TBT paints occurs. These observations imply that the obligation to remove or to seal existing TBT paints accordingtothe IMOAF-Convention(IMO 2008)has not Methods been effected on leisure boats around the Baltic Sea. Today, antifouling paints based mainly on copper, in some A handheld X-ray Fluorescence (XRF) analyzer calibrated for countries boosted by organic co-biocides, are the most com- area measurements of tin, copper, and zinc in antifouling monly used coatings (Brooks and Waldock 2009). The general paints coated on plastic boat hulls (Ytreberg 2012) has been habit in most countries is that before launching, the hull is coat- used to measure metals on leisure boat hulls in Finland (FI), ed with new antifouling paint each year, and most boat owners Denmark (DK), and Germany (DE) in the period April 2015 are following the instructions on the can where often two layers to February 2017. The Olympus XRF-analyzer (Delta-50, are recommended to be applied by brush or roller. However, it is Innov-X) is equipped with a 4 W, 50 kV X-ray tube, which commonly observed that the paint thickness on the boat hulls is able to excite and detect heavy elements such as the K-lines increases each year, and that not all paint is leaking out during of tin. The penetration depth differs depending on the element the boat season. This practice indicates the excess use of anti- itself and the combination of other elements in the sample. fouling paints. With better control of the amount of paint applied With lower total metal content measurement of tin has been on the boat hull each year, the performance could be just as shown to be linear up to ca 120 μm dry film and with higher good with less toxic compounds distributed into the sea. total metal content, it was linear up to 55 μmdry film Environ Sci Pollut Res (2018) 25:14595–14605 14597 (Ytreberg et al. 2017b). With thicker samples and higher total In all, 377 boats were investigated. The relationship between metal content, it is increasingly underestimating the concen- the mean and the median values for tin and copper tration of metals up to a thickness maximum of 600 μmdry mesurements is illustrated in Figs 1 and 2. The results show film (Ytreberg et al. 2017b). a linear relationship with R for tin 0.9735 but when one boat Three sites were visited in Finland along the southern coast, with very high values of around 2000 was excluded, the R and the number of measured leisure boats was 52, 53, and 51, was 0.9566 (Fig. 1). For copper, the relationship was even respectively. The salinity was around 4–5 practical salinity better with a R of 0.9724 (Fig. 2). unit (PSU) at all these sites. In Denmark, three boat clubs near The mean and median values of 6–8 measurements per Helsingör at the Öre sound were measured with 25, 28, and 28 boat for tin and copper are shown in Figs. 3 and 4, and a boats at the respective sites. The salinity in Öre sound is in the summary of the data are presented in Table 1. All boats where range 12 to 15 PSU. In Germany, two boat clubs in the Kiel tin was detected, also contained copper. bay were visited, one with 50 boats and a storage hall with 90 The results of the quality procedure were for the means of boats from five different harbors. The salinity ranges between all means of the tin sample 599; the means of all mean stan- 12 and 20 PSU. The selection of boats was either that all boats dard deviations was 9 and the coefficient of variations was 1.4. at the visited site were measured or a random selection based For copper, the mean of all means was 4199; the mean of the on every third boat or equivalent was measured. The permis- standard deviations was 65 and the coefficient of variation sion to measure the boats was provided by the harbor masters was 1.6. of the individual boat clubs. The results of one layer of antifouling paints on plastic Each boat hull was measured at six to eight places. The pieces from the tin paint Hard Racing Super gave a measured places were always measured in the order starboard back, value of approximately 300 μg Sn/cm , and the measurement starboard middle, starboard front, portside front, portside mid- of the copper antifouling paint MilleXtra gave a value of ap- dle, portside back, and two measuring points on the rudder or proximately 4000 μg Cu/cm . stern. The position of the measuring points on sailboat hulls was ca 20 cm below the water line. On motor boats, the bot- tom of the hull was measured on starboard and portside and on Tin the stern at the side ca 10 cm below the waterline. The mea- suring time at each place was 10 s where the 50 kV beam of The limit of quantification (LOQ) for field measured tin was the instrument was used for the entire period. In the calcula- estimated to 50 μg Sn/cm . The majority of boats in all four tion, both the mean and the median values of all measure- countries had tin below 50 μgSn/cm , i.e., 58, 76, and 77% as ments per boats have been compared. mean value and 62, 78, and 82% as median values for all The data has been compared to existing data from Sweden measured sites in DK, FI, and DE, respectively (Fig. 3 and performed in 2015 at the West coast of Sweden (20 PSU), in Table 1). Mean values higher than 100 μg Sn/cm were found the Stockholm area (5–6 PSU), and in freshwater (Ytreberg on 25, 13, and 9% of the boats, and higher than 400 μgSn/cm et al. 2016). In this study only one place per boat was were found on 10, 5, and 1% of the boats in DK, FI, and DE, measured. respectively. The corresponding percent of median values The performance of the instrument was checked at least higher than 100 μg Sn/cm was 21, 12, and 6.5, and higher twice each day of measurement by shooting at two reference than 400 μgSn/cm was 8.7, 6, and 1 for boats in DK, FI, and samples where one contained tin and the other copper. Four DE, respectively. In Sweden, the mean value for all measured measurements were conducted on each reference sample. boats with less than 50 μgSn/cm was 63%, and the respec- The data has also been compared to measurements on one tive percentage for fresh water, brackish water, and salt water layer of antifouling paints applied by a roller on a plastic piece was 65, 76, and 48. The Swedish values for boats higher than from a boat hull. For tin, an old but newly opened can from 100 μg Sn/cm were 26, 16, and 29 for fresh water, brackish Hempel (Hard Racing Super, no 4182) containing bis (TBTO) water, and saltwater, and for boats with higher values than and tributyltin fluoride) was used and for copper, the com- 400 μg Sn/cm , the corresponding figures were 10, 6, and monly used antifouling paint Hempel MilleXtra (71100) con- 16% (Ytreberg et al. 2016). The variation among boats is high, taining 34% copper-containing (di copper oxide) paint) was and a maximum mean value of 2000 μg Sn/cm was found on used. a Finnish boat, which corresponded to the maximum median of 1900 μgSn/cm . The maximum means at the other Finnish sites were 640 and 680 μgSn/cm , and the maximum medians Results were 650 and 550 μg Sn/cm . In Denmark, the maximum values were 690, 690, and 800 μgSn/cm at the respective At all investigated sites in all three countries, the majority of three sites, and the median values were respectively 720, 730, boats were coated with biocide containing antifouling paints. and 880 μgSn/cm . The lowest maximum values were found 14598 Environ Sci Pollut Res (2018) 25:14595–14605 Fig. 1 Relationship between Tin - Relaonship mean and median values (n=376) mean and median values of tin from 376 measured boat hulls (6– 8 measurements per boat hull) in 900 y = 0.9498x - 2.1274 Denmark, Finland, and Germany. R² = 0.9566 (One very high value with mean and median around 2000 was excluded in the graph) 0 100 200 300 400 500 600 700 800 900 Mean values, μg Sn/cm in Germany with 530 μg Sn/cm at one site and 260 μgSn/ water, respectively. The Swedish proportion of boats with 2 2 cm at the other site. The German corresponding median higher values than 8000 μgCu/cm were 6, 17, and 55% for values were 500 and 280 μg Sn/cm . The highest maximum fresh water, brackish water, and saltwater. Also for copper, the values were found in Sweden with 3000 μg Sn/cm on one variation among boats is large. The maximum mean value of 2 2 boat at the West coast, 1700 μgSn/cm on a boat sailing in 37,100 μgCu/cm was found at one of the Finnish boat clubs, 2 2 Brackish water, and 2100 μgSn/cm on boat at a storage area and the maximum median value was 38,700 μg Cu/cm .The near the lake Mälaren. mean maximum copper values at the other two sites were 24,800 and 34,100 μg Cu/cm , and the median maximum were 25,900 and 36,200 μg Cu/cm . The German boat sites had similar maximum mean values of 32,100 and Copper 33,000 μgCu/cm with median maximum of 29,700 and 34,500 μg Cu/cm . The highest mean value in Denmark was The limit of quantification (LOQ) for field measured copper 26,300 μgCu/cm , and the maximum at the other sites were was estimated to 100 μgCu/cm . The majority of boats had a 16,000 and 19,500 μgCu/cm , respectively. Based on the me- mean copper value higher than 100 μgCu/cm with 100% of dian values, the corresponding maximum for the Danish boats the boats in Denmark and Germany and 97% in Finland (Fig. were 28,500, 16,200, and 21,100. The Swedish maximum 4 and Table 2). When based on median values, the percentage values were again at the top with 58,000 μgCu/cm for a boat boats with < 100 μgCu/cm was 99, 96, and 100 in DK, FI, at the West coast, 46,200 μgCu/cm at the brackish water site, and DE. The proportion of boats with higher mean copper and 25,500 μgCu/cm for boats in the lake Mälaren. values than 8000 μg/cm was 51, 56, and 61% for boats in The content of copper differs among the countries and the DK, FI, and DE, respectively, and the corresponding median areas. As a summary, the highest median concentrations of values were 50, 54, and 61% for DK, FI, and DE (Table 2). In copper were found on boats at the West coast of Sweden Sweden, the percentage boats with less than 100 μgCu/cm 2 2 (9500 μg/cm ) > Helsingör at the Sound (8000 μg/cm )> were 29, 11, and 1% for fresh water, brackish water, and salt Fig. 2 Relationship between Copper - aRelaonship mean and median (n=377) mean and median values of copper from 377 measured boat y = 0.9874x - 77.342 hulls (6–8 measurements per boat 40000 R² = 0.9724 hull) in Denmark, Finland, and Germany 0 5000 10000 15000 20000 25000 30000 35000 40000 Mean values, μg Cu/cm 2 2 Median values, μg Cu/cm Median values, μg Sn/cm Environ Sci Pollut Res (2018) 25:14595–14605 14599 Tin on boat hulls in Denmark, mean values Tin on boat hulls in Denmark, median values 100 100 DK 1 (n=25) DK 1 (n=25) DK 2 (n=28) DK 2 (n=28) 80 80 DK 3 (n=28) DK 3 (n=28) 60 60 <50 50-99 100-199 200-399 400-799 800-1599 >1600 <50 50-99 100-199 200-399 400-799 800-1599 >1600 μg Sn/cm μg Sn/cm Tin on boat hulls in Finland, mean values Tin on boat hulls in Finland, median values FI 1 (n=52) FI 1 (n=52) FI 2 (n=53) FI 2 (n=53) FI 3 (n=51) FI 3 (n=51) 40 40 20 20 <50 50-99 100-199 200-399 400-799 800-1599 >1600 <50 50-99 100-199 200-399 400-799 800-1599 >1600 μg Sn/cm μg Sn/cm Tin on boat hulls in Germany, mean values Tin on boat hulls in Germany, median values 100 100 DE 1 (n=50) DE 1 (n=50) 90 90 DE 2 (n=90) DE 2 (n=90) <50 50-99 100-199 200-399 400-799 800-1599 >1600 <50 50-99 100-199 200-399 400-799 800-1599 >1600 μg Sn/cm μg Sn/cm Tin on boats in Sweden (one measurement per boat hull) Fresh water (n=228) Brackish water (n=268) 50 Salt water (n=202) <50 50-99 100-199 200-399 400-799 800-1599 >1600 μg Sn/cm Fig. 3 Proportion of tin on boat hulls in Denmark, Finland, and Germany (7–8 measurements per boat hull) in comparison to Swedish data (one measurement per boat hull) from Ytreberg et al. 2016 Helsinki in the Finnish Bay (7500 μg/cm )>Stockholm area assumption that tin corresponds to the amount of organotin 2 2 (2400 μg/cm ) and lowest in freshwater (900 μg/cm ). compounds has been shown by Lagerström et al. (2017)who found a linear relationship (R = 0.934) between the total con- tent of tin and the sum of chemically detected organotin com- pounds in layers of antifouling paints scraped from painted Discussion hulls of leisure boats. Therefore, higher tin values are consid- ered to pose a risk for leakage of organotin compounds to the The XRF-method, calibrated for area measurements of metals environment. on boat hulls (Ytreberg et al. 2016), used in this study has been In this study, eight measurements per boat hull have been shown to effectively identify those boats with antifouling measured, and both the mean and the median values have paints containing tin and copper. The validity of the Percent Percent Percent Percent Percent Percent Percent 14600 Environ Sci Pollut Res (2018) 25:14595–14605 Copper on boat hulls in Denmark, mean values Copper on boat hulls in Denmark, median values DK 1 (n=25) DK 1 (n=25) DK 2 (n=28) DK 2 (n=28) DK 3 (n=28) DK 3 (n=28) 40 40 <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 > 32000 <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 > 32000 μg Cu/cm μg Cu/cm Copper on boat hulls in Finland, median values Copper on boat hulls in Finland, mean values 60 60 FI 1 (n=52) FI 1 (n=52) FI 2 (n=53) FI 2 (n=53) 50 50 FI 3 (n=51) FI 3 (n=51) 40 40 30 30 20 20 10 10 <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 > 32000 <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 > 32000 2 2 μg Cu/cm μg Cu/cm Copper on boat hulls in Germany, mean values Copper on boat hulls in Germany, median values DE 1 (n=50) DE 1 (n=50) 50 50 DE 2 (n=90) DE 2 (n=90) 40 40 30 30 20 20 10 10 <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 > 32000 <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 > 32000 μg Cu/cm μg Cu/cm Copper on boats in Sweden (one measurement per boat hull) Fresh water (n=228) Brackish water (n=268) Salt water (n=202) <100 100-999 1000-1999 2000-3999 4000-7999 8000-16000 16000-31999 μg Cu/cm Fig. 4 Proportion of copper on boat hulls in Denmark, Finland, and Germany (6–8 measurements per boat hull) in comparison to Swedish data (one measurement per boat hull) from Ytreberg et al. 2016. Both mean and median values are shown been used to compare the results from the different countries. suggested that when the XRF method is used for regulative The high relationships for both tin and copper show that both screening of boats, all individual data should be reported and mean and median values may be used for comparison between not only the mean or median value. the different countries and different sites. It also indicates that The penetration depth of the signal depends on the thick- usually the measured contents on the boat hull are rather even. ness of the paint layer and on the composition of other metals The reason for the few discrepancies is the deviation of one or (Ytreberg et al. 2017b). For these reasons, the method may two data on the boat. Usually, this is often the case for the underestimate the amount of metals at thicker samples and in rudder/stern, which may differ from the rest of the boat hull, particular at high concentrations of several metals. Therefore, either much higher metal content or much lower. This is im- the calibrated XRF-method has to be regarded as a screening portant information to the boat owner, and it is therefore tool. An advantage of the method is the low cost, where a total Percent Percent Percent Percent Percent Percent Percent Environ Sci Pollut Res (2018) 25:14595–14605 14601 Table 1 Percentage of boats with mean and median values (6–8 measurements) of tin on plastic boat hulls Country and Mean values of tin Median values of tin No. of sites <50 μg >100 μg >400 μg >800 μg <50 μg >100 μg >400 μg >800 μg 2 2 2 2 2 2 2 2 Sn/cm Sn/cm Sn/cm Sn/cm Sn/cm Sn/cm Sn/cm Sn/cm St d e v i n St d e v i n St d e v i n St d e v i n St de v i n St d e v i n St d e v i n St d e v i n brackets brackets brackets brackets brackets brackets brackets brackets Denmark 1 (n = 25) 56 24 8 0 64 16 8 0 Denmark 2 (n = 28) 57 21 11 0 61 21 7 0 Denmark 3 (n = 28) 61 29 11 3.6 61 25 11 3.6 Mean DK (n= 3) 58 (2.5) 25 (4) 10 (1.6) 1.2 (2.1) 62 (2) 21 (4.5) 8.7 (2.1) 1.2 (2.1) Finland 1 (n = 52) 81 13 8 0 85 12 8 0 Finland 2 (n = 53) 74 11 4 0 76 11 6 0 Finland 3 (n = 51) 73 16 4 2 73 14 4 2 Mean FI (n= 3) 76 (4.5) 13 (2.5) 5 (2.3) 0.7 (1.2) 78 (6.2) 12 (1.5) 6 (2) 0.7 (1.2) Germany 1 (n = 50) 70 16 2 0 74 12 2 0 Germany 2 (n=90) 84 20 089 100 Mean DE (n= 2) 77 (10) 9 (9.9) 1 (1.4) 0 (0) 82 (11) 6.5 (7.8) 1 (1.4) 0 (0) SE, freshwater 65 26 10 4 65 26 10 4 (n =202) SE, brackish water 76 16 6 3 76 16 6 3 (n =268) SE, marine water 48 29 16 7 48 29 16 7 (n = 202) Mean Sweden (n= 3) 63 (14) 24 (6.5) 11 (5.0) 5 (2.4) 63 (14) 24 (6.5) 11 (5.0) 5 (2.4) Data for Sweden are from Ytreberg et al. 2016 and based on a single measurement per boat LOQ = 50 μg Sn/cm The mean values are presented in italic with the standard deviation in brackets of eight measurements of organotin compounds and copper on TBT one boat is much less than the cost of just one sample of the corresponding chemical analysis. The calibrated XRF-method The differences among the four countries were large when it is easy to use and could be a useful tool in regulative work by comes to the proportion of boats with higher tin values. One environmental authorities. layer of TBT paint is around 300 μg/cm .Meanvaluesof tin In accordance with the results from Sweden (Ytreberg et al. >400 μg/cm was found on 10% of the Danish boats and in 2016), there are no indications of freshly applied TBT-paints 16% of the Swedish boats measured at the West coast, where- by illegal sale and application. Most likely the measured tin as in Finland, only 5% and in Germany, only 1% of the boats concentrations on the leisure boat hulls indicate the persis- had higher tin content than 400 μg/cm . The median values tence of old TBT-paints probably as remnants in under laying were almost the same (Table 1). Apparently, the enforcement paint layers after incomplete removal (Lagerström et al. 2017). of the TBT regulation has differed in the investigated Even if many boat clubs have rules for protection of the countries. ground, most boat clubs do not enforce this regulation. In a Even if the overall results show that the majority of the compilation of results from 34 investigated boat storage areas measured boats in all four countries had concentrations below in Sweden, organotin compounds have been shown to be the LOQ of 50 μgSn/cm , still a high proportion of the boats spread and accumulated to the unprotected ground along with have remnants on their boats with > 50 μg Sn/cm , i.e., on 42, scraping and sanding during maintenance of the boat hulls 24, 23, and 37%, in DK, FI, DE, and SE, respectively (Eklund and Eklund 2014). The risk is evident that these bio- (Table 1). It was not possible to get the building year of the cides from antifouling paints on the boats may be spread to boats but most likely, the tin and organotin compounds are adjacent waters in connection with rain and run-offs, which is found in under-lying paint layers on boats built before the illustrated by results in soil and sediment at a boatyard studied prohibition of organotin compounds in antifouling paints by Lagerström et al. (2016). (Council Directive 89/677/EEC). However, it cannot be 14602 Environ Sci Pollut Res (2018) 25:14595–14605 Table 2 Percentage of boats with mean and median values (6–8 measurements) of copper on plastic boat hulls Country and no. of sites Mean values of copper Median values of copper % boats in different ranges % boats in different ranges <100 μg >4000 μg >8000 μg <100 μg >4000 μg >8000 μg 2 2 2 2 2 2 Cu/cm Cu/cm Cu/cm Cu/cm Cu/cm Cu/cm St dev in brackets St dev in brackets St dev in brackets St dev in brackets St dev in brackets St dev in brackets Denmark 1 (n = 25) 0 88 56 0 92 56 Denmark 2 (n = 28) 0 75 39 0 79 39 Denmark 3 (n = 28) 0 61 57 4 64 54 Mean DK (n= 3) 075 (14) 51 (10) 1.3 (2.3) 78 (14) 50 (9.3) Finland 1 (n=52) 8 7348 1069 48 Finland 2 (n = 53) 0 79 66 0 79 62 Finland 3 (n = 51) 0 75 53 2 71 51 Mean FI (n= 3) 3 (4.6) 76 (3.2) 56 (9.3) 4 (5.3) 73 (5.4) 54 (7.5) Germany 1 (n = 50) 0 88 68 0 88 66 Germany 2 (n = 90) 0 89 54 0 89 56 Mean DE (n =2) 088 (0.6) 61 (9.6) 088 (0.6) 61 (7.4) SE, freshwater (n = 202) 29 18 6 29 18 6 SE, brackish water (n =268) 11 37 17 11 37 17 SE, marine water (n = 202) 1 72 55 1 72 55 Mean Sweden (n =3) 14 (14) 42 (27) 26 (26) 14 (14) 42 (27) 26 (26) Data for Sweden are from Ytreberg et al. 2016 and based on a single measurement per boat LOQ = 100 μgCu/cm The mean values are presented in italic with the standard deviation in brackets excluded that a few newer boats may have been imported from from wash pads at the East coast (Ytreberg 2012, Ytreberg et al. countries outside Europe with less restriction of TBT paint. 2015). These data show that TBT is still discharged in connec- Overall, the pattern of tin occurrence on boat hulls in all tion with washing of the hull. High-pressure washing is com- four countries was alike (Fig. 1). Even in Sweden, there were mon practice in all four countries albeit antifouling paints ex- no large difference between the different areas, and in fact, the cept hard coatings are designed to erode or polish and cannot highest measured tin concentration of 3000 μgSn/cm was on stand hp-washing without damage or removal of upper paint a boat in fresh water. The little difference between the coun- layers. tries may be due to that almost all boats built before 1989, The national laws around the Baltic Sea are quite differing. when the prohibition of TBT was enforced in EU (Council In Germany, according to the Federal Water Act, it is illegal to Directive 89/677/EEC), were coated with TBT paints and thus clean boat hulls with tap or high-pressure water outside wash- the proportion of older boats with such paints is similar. down areas or tarps as collection systems and is only allowed High inputs derived from high pressure washing (hp) after on wash-down areas with collection and filter systems. In a lifting the boats in autumn has been determined. In a study few harbors at the German Baltic coast, hp-washing is 2007 of sediments in a gradient towards the slipway in a small prohibited. In Sweden, there is a national recommendation harbor for recreational boats, the concentration of organotin not to clean boats outside areas with filtering and collection compounds increased the closer to the uptake area (Eklund systems (Anon 2015a, b). The national authority leaves it to et al. 2008). In a Finnish study performed in 2015 and 2016 the municipality to make the final decision about the regula- of wash water from a wash down area, both TBT and tion in their respective community. A similar approach is triphenyltin (TPhT) were detected. The values of TBT were found in Denmark where the local harbors can decide on reg- 0.023 and 0.75 μg/L and for TPhT 0.085 and 1.4 μg/L ulations for the boats in their harbor (personal communication (Haaksi and Gustafsson 2016). Measurements in wash pads H. Anker, Professor at Copenhagen University). Many munic- in Sweden of wash water and precipitated waste from several ipalities in Finland have enforced local regulations for boat wash pads show high amounts of biocides in the sediment washing refereeing to the national environmental law stating including TBT with median values of 69,000 μgTBT/kg dry that the ground and water should not be polluted (personal sediment at the West coast and 11,000 μgTBT/kg drysediment communication with Tomas Kull, Environmental inspector Environ Sci Pollut Res (2018) 25:14595–14605 14603 in Pargas City). In practice, it does not exist specific regula- boat is thus: a boat with a hull area of 20 m and a copper tions for most of the small boat harbors in the investigated concentration of 8000 μgCu/cm will result in 1.6 kg copper countries. and 3.2 kg copper with a concentration of 16,000 μg Cu/cm . However, even if there is a global ban for use of TBT The leakage rate varies depending on the formula in the paint containing antifouling products, such products are still on sale but also on the salinity of the water. In more saline waters, the in some areas, and the contamination is not decreasing (Bargar leakage rate is faster than in less salty waters and in freshwater et al. 2013; Turner and Glegg 2014, https://www. (Ytreberg et al. 2017a; Lagerström et al. 2018). The conclu- seahawkpaints.com/antifouling-bottom-paint/visited at 22- sion is still that much copper each year may leak into the 08-2017). environment from leisure boats. Much of the copper may be A boat hull with an average 100 or 400 μg TBT/cm means taken up by fouling organisms attached to the hull and may on a hull of 10 m a total amount of 10 or 40 g. In view that as reach concentrations up to 28,000 mg copper/kg dw (Bighiu low concentrations as 1 ng/L are sufficient to affect the endo- et al. 2017). When this material in connection with scraping crine system of molluscs (Bryan et al. 1986), and low concen- and sandpapering enters the ground, it exceeds the guidance trations of TBT are harmful to other organisms (references in values for least sensitive land use of 200 mg copper/kg dw in review by Antizar-Ladislao 2008), the amount of organotin Sweden by factors up to 140 and thus adds to the contamina- compounds that potentially could be distributed to the envi- tion of the soil. ronment from boat hulls is considerable. In Sweden, the pattern is different depending if the boat is On the background of persisting concentrations of tin on moored in salt water at the West coast, in brackish water in the boat hulls, we suggest the following options to be enforced by Stockholm area, or in freshwater where biocidal antifouling environmental authorities, harbor operators, boat clubs, etc.: paints are not allowed (Ytreberg et al. 2016). The boats with Certificate for boats built before 2008 on the existing paint the higher copper values are found on the saltwater boats and layers. In case of existing paints with illegal compounds like the lowest in freshwater (Table 2). The salinity along the organotin, diuron, or irgarol, these boats should be strictly Finnish coast is comparable to the Stockholm water area and excluded from hp-washing outside wash-down areas. the fouling pressure is similar. However, the use of copper in Cleaning with a sponge should be performed with a tarp to Finland is as high as in the sites with saltwater, i.e., in protect the ground and collection of the wash water. Germany, Denmark, and Swedish West coast. This difference Or obligation to remove all paint layers down to the primer between Sweden and Finland is probably due to the different in case no certificate can be delivered regulations in the two countries. In large, Finnish boat owners Prohibition of hp-washing for boats from other harbors, are allowed to use the same antifouling paints as in Germany, resp. guest boats unless presenting a certificate. Denmark, and the West coast of Sweden, and this is apparent- In harbors where the erection of a wash-down area with ly what they do. In Sweden, a differentiated regulation was collection and filter systems is too expensive, as maintenance enforced already in 1993 (Chemical Agency in Sweden practice should be recommended to clean the hulls after lifting 1993). Based on a risk benefit/analysis, the decision was taken in autumn with a soft sponge and sand them softly, if neces- that no biocide-containing antifouling paints were approved sary, before launching in the next spring. These maintenance for use in freshwater and the Bothnian Bay, lower copper works should include protection measurements of the ground. content was approved for the Swedish brackish water, and higher copper content was approved for use in marine waters (West coast of Sweden). The Swedish data show that the lower Copper salinity, the higher proportion of boats with less copper. This finding supports that the regulations have had a restrictive The results show similar pattern of range intervals of copper in effect. However, even if copper in antifouling paints has been Denmark, Finland, Germany, and the Swedish West coast prohibited for use on boats in freshwater for more than 2 2 with most boats in the interval 8000 to 16,000 μg Cu/cm , 20 years in Sweden, copper > 100 μgCu/cm was still mea- and between 50 and 61% of these boats had more than sured on 71% of the surveyed boats along the lake Mälaren 8000 μg Cu/cm both calculated as means or as median values (Ytreberg et al. 2016). One explanation is that boaters belong- (Fig. 4, Table 2). This can be related to the fact that one layer ing to boat clubs in Lake Mälaren used their boats both in the of the commonly used copper-based paint Hempel MilleXtra. lake and in the brackish environment and interpreted the reg- corresponds to ca 4000 Cu/cm . Thus, two paint layers, which ulation that they then were allowed to use copper paints. are recommended by the paint producers to last for a boat However, the writing of the regulation has the last year be- season, is roughly 8000 μg Cu/cm . This means that most come stricter, and today it is prohibited to use antifouling paint boats have more paint layers on their boats, which must be based on copper in fresh water. The XRF-method may be a regarded as an excess of what is needed. The potential amount good tool for the environmental authorities to control this of copper that may leak into the environment from a single regulation. 14604 Environ Sci Pollut Res (2018) 25:14595–14605 The high concentrations of copper on leisure boat hulls in Kiel The data from all countries show that an excess of antifoul- may be explained by several factors. Due to the high fouling ing copper is used in all countries. Approximately 50% of the pressure with barnacles and blue mussels along the German measured boats in DK, DE, and FI, and at the West Coast of Baltic coast, German boat owners repaint the hull every year SE had higher concentrations on their boat hulls than two irrespective of a fail in performance of the antifouling paint in layers of the most common copper-containing antifouling the previous season. Paint manufacturers recommend paints with paint with 34% copper. This excess of copper use should be copper contents of 20–25%. It is uncommon to repaint exclu- stopped. sively those areas of the hull which have been fouled in the previous season, and it is recommended but uncommon, to apply each paint layer in a different colors to control the polishing or Open Access This article is distributed under the terms of the Creative erosion rate. Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, The new application of a handheld X-ray analyzer facilitates distribution, and reproduction in any medium, provided you give appro- identification of boats with high contents of toxic heavy metals priate credit to the original author(s) and the source, provide a link to the on plastic boat hulls. Work on calibrations for hulls made by Creative Commons license, and indicate if changes were made. steel, aluminum, and wood is on its way. The calibration in metals/cm gives figures, which can be compared with other References measurements in space and time in contrast to ordinary XRF- analyzers which only provides figures in percent (Ytreberg Anon (2005) Common implementation strategy for the water framework 2012). In particular boats with high tin concentrations, indicating directive. Environmental quality standards (EQS). Substance data organotin compounds, can then be recommended for removal or sheet. Priority substance no. 30 tributyltin compounds (TBT-ion) sealing of the old paint in accordance with the Water Framework CAS-No. 688–73-3 (36643–28-4). Final version, Brussels, 15 January 2005. https://circabc.europa.eu/sd/d/899759c1-af89-4de4- Directive in the EU (2000/60/EC). In addition, violation of the 81bf-488c949887c8/30_Tributyltin_EQS datasheet_150105.pdf. ban of copper-based paints in freshwater areas can be detected Accessed 4 Dec 2015 and identified. This would be a fast way to eliminate further Anon (2015a) Regulations of the Swedish Agency for Marine and Water discharges of these substances to the environment. Under this Management on revision of the regulations of the Swedish Agency for Marine and Water Management (HVMFS 2013:19) on classifi- aspect, the XRF method can be regarded as a powerful tool for cation and environmental quality standards of surface waters. the survey and monitoring of antifouling practice. HVMFS 2015:4. Swedish Agency for Marine and Water According to the implementation of the EU-BPR and the Management, Gothenburg, Sweden (in Swedish) transformation into national law, it can be expected that several Anon (2015b) Boat bottom washing of recreational boats. Guidance, copper compounds will be incorporated in authorized antifouling Revised version 2015 (In Swedish, Båtbottentvättning av fritidsbåtar. Riktlinjer, reviderad upplaga 2015) https://www. products of the future. Despite initiatives to reduce the copper havochvatten.se/hav/vagledning–lagar/vagledningar/ovriga- content of antifouling paints and to phase-out biocidal antifouling vagledningar/batbottentvatt-av-fritidsbatar.html paints by 2030, there will be a need for monitoring, especially in Antizar-Ladislao B (2008) Environmental levels, toxicity and human ex- freshwater areas and the Bothnian Bay where actually no biocid- posure to tributyltin (TBT) contaminated marine environment. Environ Int 34:292–308 al paints are allowed at the Swedish side. Bargar TA, Garrison VH, Alvarez DA, Echols KR (2013) Contaminants assessment in the coral reefs of Virgin Islands National Park and Virgin Islands Coral Reef National Monument. Mar Pollut Bull 70: Conclusions 281–288 Bighiu MA, Eriksson-Wiklund A-K, Eklund B (2017) Biofouling of leisure boats as a source of metal pollution. Environ Sci Pollut Res In comparison to chemical analyses, the XRF-method to iden- 24:997–1006 tify tin and copper on boat hulls is a non-destructible, cheap Brooks SJ, Waldock M (2009) Copper biocides in the marine environ- and fast method. The calibrated screening method is recom- ment. In: Arai T, Harino H, Ohji M, Langston WJ (eds) mended to be used in regulative works to provide data on Ecotoxicology of antifouling biocides. Springer, Tokyo, pp 413–428 which regulatory authorities can rely in determining which Bryan GW, Gibbs PE, Hummerstone LG, Burt GR (1986) The decline of the gastropod Nucella lapillus around south-west England: evidence boats need to remove tin paint or have excess of copper paint. for the effect of tributyltin from antifouling paints. J Mar Biol Assoc The need is illustrated by the fact that 23–42% of the in total UK 66:611–640 377 measured boats in Denmark, Finland, and Germany still Chemical Agency in Sweden (1993) Debourg, C, Johnson, A, Lye, C, have amount of tin (organotin compounds) on their hulls. This Törnqvist, L, Unger, U Antifouling Products – Pleasure boats, com- mercial vessels, nets, fish cages and other underwater equipment is a source to leakage to the environment and is one cause to KEMI Report No 2/93 elevate organotin concentrations in the sediments along the Cornelissen G, Pettersen A, Nesse E, Eek E, Helland A, Breedveld GD Baltic shores and a reason why good status according to the (2008) The contribution of urban runoff to organic contaminant Marine Strategy Framework Directive (2008/56/EC) is not levels in harbour sediments near two Norwegian cities. Mar Pollut Bull 56:565–573 achieved. Environ Sci Pollut Res (2018) 25:14595–14605 14605 Council Directive 2000/60/EC of the European Parliament and of the Coatings from Ships, including TBT hull paints Submitted by the United Kingdom. LC 30/9, 10pp Council of 23 October 2000 establishing a framework for Community action in the field of water policy - The Water Lagerström M, Norling M, Eklund B (2016) Metal contamination at Framework Directive recreational boatyards linked to the use of antifouling paints – in- Council Directive 89/677/EEC of 21 December 1989 amending for the vestigation of soil and sediment with a field portable XRF. Environ eighth time Directive 76/769/EEC Sci Pollut Res 23:10146–10157 http://link.springer.com/article/10. Council Directive 89/677/EEC of 21 December 1989 amending for the 1007/s11356-016-6241-0 eighth time Directive 76/769/EEC on the approximation of the laws, Lagerström M, Strand J, Eklund B, Ytreberg E (2017) Total tin and regulations and administrative provisions of the Member States re- organic speciation in historic layers of antifouling paint on leisure lating to restrictions on the marketing and use of certain dangerous boat hulls. Environ Pollut 220:1333–1341 substances and preparations Lagerström M, Lindgren JF, Holmqvist A, Dahlström M, Ytreberg E Eklund B, Eklund D (2014) Pleasure boat yard soils are often highly (2018) In situ release rates of Cu and Zn from commercial antifoul- contaminated. Environ Manag 53:930–946 http://www. ing paints at different salinities. Mar Pollut Bull 127:289–296 springerlink.com/openurl.asp?genre=article&id=doi:10.1007/ Maguire RJ (2000) Review of the persistence, bioaccumulation and tox- s00267-014-0249-3 icity of tributyltin in aquatic environments in relation to Canada’s Eklund B, Elfström M, Borg H (2008) Tributyltin originates from plea- toxic substances management policy. Water Qual Res J Can 35:633– sure boats in Sweden in spite of firm restrictions. Open Environ Sci 679 2:124–132 Marine Strategy Framework Directive (2008/56/EC) Eklund B, Elfström M, Gallego I, Bengtsson B-E, Breitholtz M (2010) Nyberg E, Poikane R, Strand J, Larsen MM, Danielsson S, Bignert A Biological and chemical characterization of harbour sediments from (2014) Tributyltin (TBT) and imposex HELCOM Core Indicator of the Stockholm area. Soil Sed Pollut 10:127–141 Hazardous Substances Tributyltin (TBT) and imposex. 17 pp. Eklund B, Johansson L, Ytreberg E (2014) Characterization and risk HELCOM, CORESET assessment of a boatyard for pleasure boats. J Soil Sed 14:955– Strand J (2009) Coupling marine monitoring and environmental risk as- 967 http://www.springerlink.com/openurl.asp?genre=article&id= sessment of TBT: a case study using the contamination of organotin doi:10.1007/s11368-013-0828-6 compounds in the Danish marine environment. VDM Verlag Dr, Eklund B, Hansson T, Bengtsson H, Eriksson Wiklund A-K (2016) Toxic Müller 84 pp effects on the red alga Ceramium tenuicorne of polluted sediment Turner A, Glegg G (2014) TBT-based antifouling paints remain on sale. from natural harbor and small boat harbors at the west coast of Mar Pollut Bull 88:398–400 Sweden. Arch Environ Contam Toxicol 70:594 http://link.springer. Viglino L, Pelletier E, St-Louis R (2004) Highly persistent butyltins in com/article/10.1007/s00244-016-0262-z northern marine sediments: a long-term threat for the Saguenay Gipperth L (2009) The legal design of the international and European Fjord (Canada). Environ Toxicol Chem 23:2673–2681 Union ban on tributyltin antifouling paint: direct and indirect effects. Ytreberg E (2012) Dispersion of biocides from boats – Investigation of J Environ Manag 90:S86–S95 different sources and their contribution. (Spridning av biocider från Haaksi H, Gustafsson J (2016) What is discharged from a boat bottom? båtar – undersökning av olika källor och dess bidrah). ITM report Investigation about advantages with a boat wash pad. (VAD 215, Stockholm University (In Swedish) LÖSGÖRS FRÅN BÅTENS BOTTEN? – UTREDNING OM Ytreberg E, Lundgren L, Bighiu MA, Eklund B (2015) New analytical MILJÖFÖRDELAR MED EN BÅTBOTTENTVÄTTPLATS), (In application for metal determination in antifouling paints. Talanta Swedish). Report from Keep the Arcipelago Tidy (KAT), Finland 143:121–126 HELCOM (2009) Biodiversity in the Baltic Sea – an integrated thematic Ytreberg E, Bighiu MA, Lundgren L, Eklund B (2016) XRF measure- assessment on biodiversity and nature conservation in the Baltic Sea ments of tin, copper and zinc in antifouling paints coated on leisure Balt Sea Environ Proc No 116B. Available at: www.helcom.fi boats. Environ Pollut 213:594–599 http://www.sciencedirect.com/ HELCOM (2010) Hazardous substances in the Baltic Sea – an integrated science/article/pii/S026974911630210X thematic assessment of hazardous substances in the Baltic Sea. In: Ytreberg E, Lagerström M, Holmqvist A, Eklund B, Elwing H, Baltic Sea Environment Proceedings, 120B Dahlström M, Dahl P, Dahlström M (2017a) A novel XRF method Hoch M (2001) Organotin compounds in the environment: an overview. to measure environmental release of copper and zinc from antifoul- Appl Geochem 16:719–753 ing paints. Environ Pollut 225:490–496 IMO (2008) Interpretation of the London Convention and Protocol: Ytreberg E, Lagerström M, Yngsell D, Eklund B (2017b) Occurrence of Collaboration with MEPC on BBoundary^ Issues. Draft Guidance TBT on ships and leisure boats (in Swedish). Report to the Swedish on Best Management Practices for Removal of Anti-Fouling Transport Agency December 2017
Environmental Science and Pollution Research – Springer Journals
Published: Mar 12, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera