To gather further data about metal mobility in accumulated sediments in stormwater treatment facilities, metal mobility in sediments from sedimentation tanks, gully pot sediment traps, and sedimentation ponds was investigated using the sequential extraction procedure. This method allows distinguishing the metal speciation between ion-exchangeable, carbonate-associated, reducible, organic matter/sulfide-associated, and residual fractions. The metal fractionation reveals that, for all treatment facilities, the majority of Cd, Cu, Pb, and Zn and a significant amount of Ni were in potentially mobile forms. Cd, Pb, and Zn are, to a large extent, associated with Fe-Mn oxides while Cu is commonly present as Cu–organic matter complexes. The metals in these potentially mobile fractions may represent a potential environmental hazard, e.g., due to release during maintenance (sediment removal) when the chemical phase distribution might change.

Sedimentation is a widely used technique in structural best management practices to remove pollutants from stormwater. However, concerns have been expressed about the environmental impacts that may be exerted by the trapped pollutants. This study has concentrated on stormwater ponds and sedimentation tanks and reports on the accumulated metal concentrations (Cd, Cr, Ni, Pb, and Zn) and the associated toxicity to the bacteria Vibrio fischeri. The metal concentrations are compared with guidelines and the toxicity results are assessed in relation to samples for which metal concentrations either exceed or conform to these values. The water phase metal concentrations were highest in the ponds whereas the sedimentation tanks exhibited a distinct decrease towards the outlet. However, none of the water samples demonstrated toxicity even though the concentrations of Cu, Pb, and Zn exceeded the threshold values for the compared guidelines. The facilities with higher traffic intensities had elevated sediment concentrations of Cr, Cu, Ni, and Zn which increased towards the outlet for the sedimentation tanks in agreement with the highest percentage of fine particles. The sediments in both treatment facilities exhibited the expected toxic responses in line with their affinity for heavy metals but the role of organic carbon content is highlighted.

Temporal trends in stormwater pond sediment quality were evaluated by conducting field sediment surveys in 1998 and 2006. A sediment removal process in 2006 was also assessed. Results show that, for both years, the sediment closest to the inlet had the lowest concentrations of metals. The results from 1998 showed no significant differences in sediment heavy metal concentrations depth while the results from 2006 showed a decreasing trend with depth. A significant difference between the two years could be seen for Cr, Cu, and Pb. The water phase in the pond showed a difference before and after the removal of sediment. The water samples after emptying showed higher concentrations of metals as a result of disturbing and re-suspending the bottom sediment. No significant difference could be found in the samples taken from the bottom sediment before the removal of the sediment and from the sediment pile formed during the removal process. The total heavy metal concentration in water exceeded the threshold limits for the selected quality guidelines and, according to soil quality guidelines, the sediment would be classified moderately serious to serious.

The overall objective of this thesis is to increase the knowledge of both the pollutants within stormwater treatment facilities and the maintenance of these facilities. This objective has been achieved by studying water and sediment with respect to metal concentrations, fractionation, and toxicity from ponds, sedimentation tanks, and gully pots with different catchment characteristics as well as from the sediment removal process.The results showed that, in all facilities, the metals in the standing water were predominantly attached to particles. In the sediment, the ponds showed the highest concentrations of metals and percentages of fine particles followed by sedimentation tanks and gully pots. Two of the ponds and one sedimentation tank had elevated concentrations of Cu, Ni, and Zn in the sediment compared to the other studied facilities, concentrations that are consistent with these facilities being influenced by the highest traffic loads. In general, the metal fractionation between the facilities differs where the ponds have more metals in the mobile fraction than the sedimentation tanks and gully pots. All facilities showed that Cr was the least available while Cd was the most available in the ponds, Zn in the sedimentation tanks, and Cu in the gully pots. The results showed that for most metals, the oxidizing and reducing conditions would be most critical with respect to the release of metals from sediments. However, the results from the metal fractionation showed that, for all treatment facilities, between 45-96% of the metals (except Cr and Ni) were in potentially mobile form and could, therefore, be a threat to the environment if the facilities are not maintained properly. The results showed that facilities with similar catchment areas have similar metal concentrations, particle-size distributions, and fractionation in the sediment. The toxicity test showed that of all the facilities, the standing water in only one gully pot in the residential area showed a toxic response to the bacteria, Vibrio fischeri. The sediment showed toxic response in all facilities where the ponds had the highest toxicity followed by the sedimentation tanks and gully pots. Both water and sediment from all of the studied facilities exceed one or more of the compared guidelines. In general, the metal concentrations and guidelines for water and sediments were not consistent with the results from the toxicity test. Therefore, to accurately evaluate the environmental impact of pollutants trapped in stormwater treatment facilities, it is important to carry out both chemical analyses and toxicity test. These stormwater treatment facilities need to be maintained in order to function properly; therefore, the sediment must be removed. During the sediment removal from the gully pots and the pond, the concentration in the water phase increased and became more hazardous after the sediment removal compared to before. The gully pots in the residential area had higher metal concentrations in the water phase than the gully pots receiving only road runoff. The sediment phase, however, did not show a large difference in concentrations between before and after the sediment was removed. Due to the high metal and PAH concentration in the removed water, the water should not be poured back to the gully pot and based on the results, an annual cleaning of the gully pot is recommended. The identification of sources in gully pot water and sediment shows that the metals Cu, Pb, Zn, Ni, Cd and Hg and PAHs were enriched in the gully pots, which indicates that they probably come from anthropogenic sources.

Syftet med denna avhandling är att öka kunskapen om föroreningar i olika reningsanläggningar för dagvatten samt underhåll av dessa anläggningar. Detta har gjorts genom att studera metallkoncentrationer, fraktionering och toxicitet i vatten och sediment från dagvattendammar, magasin och rännstensbrunnar i olika områden samt under tömningen av flera rännstensbrunnar och en damm. Den högsta koncentrationen av metaller i vattenfasen hittades i rännstensbrunnarna jämfört med dammar och magasin. Resultatet visade att metallerna var till största delen associerade med partiklar. Dammarna visade de högsta metallkoncentrationerna i sedimentet och den högsta andelen små partiklar jämfört med magasinen och rännstensbrunnarna. Två dammar och ett magasin visade förhöjda halter av Cu, Ni och Zn i sedimentet jämfört med de andra undersökta anläggningarna vilket stämmer överens med att dessa anläggningar har högre trafikbelastning. Generellt så skiljer sig fraktioneringen av metaller mellan de olika anläggningarna där dammarna har högre procent av rörliga metaller jämfört med magasinen och brunnarna. Alla anläggningar visar att Cr är minst rörlig medan Cd är mest tillgänglig i dammarna medan Zn är mest tillgänglig i magasinen och Cu är mest tillgänglig i rännstensbrunnarna. De anläggningar som har hög metallkoncentration har också hög procent mobila metaller. Resultatet visade att för de flesta metaller var den oxiderade och reducerande fraktionen viktigast. Resultatet från fraktioneringen visar att mellan 45-96% av alla metaller (utom Cr och Ni) är rörliga i anläggningarna och kan därför bli hot mot miljön om anläggningarna inte sköts om på ett lämpligt sätt. Det visade sig att det bara var vatten från en rännstensbrunn som var toxiskt för bakterien, Vibrio fisheri, medan sedimenten i alla anläggningar visade sig vara toxiskt. Dammarna visade den högsta toxicitet därefter magasinen och sist rännstensbrunnarna. Både metallkoncentrationerna i vattnet och sedimenten från anläggningarna överskred en eller flera av de riktlinjer som jämfördes. Generellt så stämde inte metallkoncentrationerna och riktlinjerna överens med resultatet från toxicitetstesterna. För att göra en noggrann och riktigt utvärdering av anläggningarna är det därför viktigt att både utföra kemiska analyser och toxicitetstester. Alla anläggningar behöver underhållas för att de ska fungera på rätt sätt och därför är det viktigt att avlägsna det sediment som ansamlas. Under tömningen av rännstensbrunnar och dammen ökade koncentrationerna av metaller och PAH i vattenfasen vilket gjorde att den blev farligare efter tömningen än före. Däremot så förändrades inte koncentrationerna i sedimenten. Rännstensbrunnarna från bostadsområdet visade sig ha högre metallkoncentrationer i vattenfasen än rännstensbrunnarna i centrum. På grund av den höga halten av metaller och PAHer i vattenfasen som bildas under tömningen, ska detta vatten inte tömmas tillbaka i brunnarna utan den måste bli renad. Med avseende på resultaten så rekommenderas en årlig tömning av rännstensbrunnarna. Källidentifieringen visade att metallerna Cu, Pb, Zn, Ni, Cd and Hg och PAH var anrikad i rännstensbrunnarna vilket indikerade på att de kommer från antropogena källor.

A mixture of sediment and water is formed during the cleansing of catch basins. This paper discusses the concentration levels and distribution of numerous metals and organic carbon (OC) in the water phase of this mixture. The results show that due to the high concentrations of metals in the water phase, the catch basin mixture should be treated before it reaches a recipient. Three sites with different types of area and traffic intensity were sampled. Four fractions were analyzed: unfiltered, dissolved (<0.2 µm), colloidal (0.22 µm to 1 kD [kilodalton]), and truly dissolved (<1 kD). The results of the unfiltered fraction show high concentrations of metals and OC in the catch basin mixture. A comparison of Canadian and Swedish Environmental Protection Agency guidelines and the catch basin mixtures shows that concentrations exceeded the threshold values for As, Cd, Cr, Cu, Ni, Pb, and Zn. Compared with samples from a reference lake in the area, the unfiltered fraction showed high concentrations of all elements. OC seems to have a large impact on the overall speciation of trace metals in the catch basin mixture. To trace the sources of the particulate fraction in the unfiltered samples, Al-normalization was used. Al-normalization indicated that Ca, K, Mg, Na, Mn, Ba, Co, and Cr concentrations could be explained by mineral particles used as traction control. Furthermore, the trace elements As, Cu, Pb, Zn, and Ni were all enriched in the catch basin mixture.

A gully pot is often cleaned with the help of an eductor truck, which uses hydrodynamic pressure and a vacuum to loosen and remove the solids and standing liquid from a gully pot. This paper considers the polycyclic aromatic hydrocarbons (PAH) content in the gully pot mixture (water and sediment) after it has been discharged from the eductor truck. The results show that most PAH was attached to particles, and the dissolved phase represented approximately 22% of the total water concentration. No significant difference was found for the water phase between a housing area and a road, whereas a significant difference was found for NAP, ACE, FL, ANT, FLR, PYR, BaF, and BPY in the sediment at a 95% confidence level. Source identification showed that the PAH in the gully pot mixture came from mixed sources. Both the water and sediment phase exceed all or some of the compared guidelines. The result from this paper shows that not only the sediment needs to be discussed, but also the water phase created during the maintenance of different BMPs.

When eductor trucks are used to maintain catch basins, the mixture of water and sediment that is produced has to be disposed of in some way. This paper considers the quality of this mixture after it has been discharged from the eductor truck. The results show that the metal concentrations varied more in the water phase than in the sediment phase and that most of the metals were attached to particles. No significant difference could be found in the water between a housing area and a road, whereas a significant difference could be found for Cr, Cu, Ni, and Zn in the sediment at a 95% confidence level. The smallest fraction (<63 µm) in the sediment had the largest concentration of metals. Sampling procedure, time, and traffic intensity had an impact on the particle-size distribution. The water exceeded the guidelines for all studied metals; however, the concentration for the water will be affected by the wash water volume, whereas, for the sediment, only copper exceeded the guidelines.

The choice of stormwater procedures will affect the pathways of the water and its pollutants, since different stormwater facilities reduce different types of substances and fractions to varying degrees. Traditional stormwater handling has mainly been emphasised to transport the water from the cities. However, during the last year, the interest in environmental aspects has increased. Different ecological solutions, such as infiltration, wetlands, and swales will affect the path of water and its substances. The main objective of this thesis is to characterize and study the pathways of pollutants (heavy metal and organic substances) in stormwater systems. The main focus has been on gully pot water and sediment after it has been removed from the grit chamber. In order to study the pathways of pollutants, a substance flow analysis was done in Hammarby Sjöstad. Since no measurements were possible in the area, theoretical calculations were performed. The results show that, in a separate system with treatment, most of the pollutants will end up in the sediment or soil, depending on which BMP is used. The largest amount of pollutants will be discharged to receiving waters with a separate system without treatment. The samples of gully pot water and sediment were taken during the autumn 2004 in Luleå, Sweden. Three different sites were chosen due to different types of area and traffic loads: a residential area with 500 v/d and two sites in the city with different traffic intensities, 13,800 v/d and 25,500 v/d, respectively. The water samples were analysed for particle-size distribution, suspended solids (SS), heavy metals, PAH, phthalate, and nonylphenols, while the sediment samples were analysed for particle-size distribution and heavy metals. The metals in the water phase were analysed for six fractions: truly dissolved (<1kD), colloids, <0.2 µm, <0.45 µm, and two different extracted fractions. The results showed that the gully pot mixture (water and sediment) contained a wide range of substances. For the different fractions in the water phase, the highest concentration of metals varied between the three sampling sites. All analysed metals were found in the mixture and, out of 29 organic substances, 18 were detected. Most of the metals and PAH were attached to particles in the water. The concentration of PAH was highest in the city with 13,800 v/d. The only phthalates substance detected in all three areas was di-(2-etylhexyl) phthalate and the highest total concentration was found in the residential area. The concentration of total 4-nonylphenol was detected only in the residential area. The sediment's smallest fraction (<63 µm) had the highest concentration of metals. The leaching test indicates that the metals in the sediment are harder bound in the city. However, in the water, the metals are harder bound in the residential area. The water exceeds the guidelines for freshwater for most of the substances, while the sediment only exceeds the guidelines for soil for a few substances. This indicates that the gully pot mixture must undergo treatment before it can be discharged to the environment.

Few BMPs have been investigated with the regards to the problem that pollutants change speciation during the emptying process. This type of sediments that is a mixture of water can be called "wet sediments". These wet sediments can be found in different BMPs like pond, wetland, gully pots, magazines or sediment deposited in the sewer system. In this paper these wet sediments are exemplified by the water phase from a gully pot mix (sediment and water) after the emptying process. The objective of this paper considers the quality of the water phase of this mixture due to the metals cadmium, copper, nickel, lead and zinc in order to make a first characterization of wet sediments. The variation between the gully pots within the two sites is very high. The residential area has higher concentration of metals than the city for both the total and dissolved phase. A first screening was done in order to see if the pollutants change speciation during the cleaning process and the results show that the concentration is much higher in the mixture than in the water stored in the gully pot. The load to the gully pots during one year is higher in the city than in the residential area. This shows that the cleaning frequency is important for the quality of the water phase.

During the cleaning of the gully pot, water has to be added to the gully pot to remove all the sediment. This mixture of gully pot water and sediment, as well as the "wash-water," has to be taken care of. This paper considers the quality of the water phase of this mixture, due to a wide spectrum of metals and organic substances. Three sampling sites were selected according to type of area and traffic intensity. The samples were analysed for suspended solids (SS), heavy metals, PAH, DEHP, and NPEO. The highest total concentration of metals was found either in the housing area or in the city with high traffic, depending on the type of metal. The highest concentration of PAH was found in the city with medium traffic. Most of the metals and PAH were attached to particles. Even if the concentrations were higher in the housing area, one will get higher amounts of metals from the city during the same period of time. The comparison between guidelines and the blend showed that the concentration in the blend exceeds the discharge criteria for most of the substances.