Tunnels are washed regularly to maintain road safety and increase their life span. During a tunnel wash event, highly contaminated tunnel wash water is generated and released to the environment. Tunnel wash water may be led to sedimentation ponds where particles and particle bound contaminants are allowed to settle. Still, the removal of contaminants is only partial, and release of contaminated water to local recipients is of concern. Further, a growth reduction has previously been observed in fish sampled downstream of where discharge water from Vassum sedimentation pond is released to the stream Årungenelva. This reduction may be related to release of tunnel wash water from the pond to the stream. The main aim of this thesis was to investigate sub-lethal effects caused by exposure to tunnel wash water using juvenile brown trout (Salmo trutta) as a model species. Brown trout was exposed to filtered (1.2 μm) tunnel wash water in a laboratory study for 25 days. In addition, fish was sampled in the stream Årungenelva downstream and upstream (reference) from where water from Vassum sedimentation pond is discharged into the river. In fish from the laboratory study, the results revealed an increased concentration of several three-ring polycyclic aromatic hydrocarbon (PAH) metabolites in bile of fish exposed to tunnel wash water. This was however not observed for metabolites of the four-ring PAH pyrene or the five-ring PAH benzo[a]pyrene. In addition, an effect on the phase I enzyme Cytochrome P450 1A (CYP1A) was observed. Elevated activity of this enzyme, (measured as 7-ethoxyresorufn O-deethylase (EROD) activity) in gills and liver as well as elevated CYP1A protein in liver was observed in fish exposed to tunnel wash water. This indicates uptake of bioavailable contaminants of fish exposed to filtered tunnel wash water. In fish sampled downstream of the sedimentation pond in Årungenelva the biliary concentrations of PAH metabolites was lower while the EROD activity in liver was higher compared to responses observed in fish sampled upstream from the sedimentation pond. No differences were observed in EROD activity in gills or in CYP1A protein in liver between fish sampled at the two locations in the stream. Effects observed in fish sampled upstream of the pond may be explained by the close proximity between the upstream location and the highway. The biomarker responses in fish from Årungenelva may thus indicate that both locations in the stream is affected by road related contaminants Exposure to lead was assessed by quantifying the δ-aminolevulinic acid dehydratase (ALA-D) activity in red blood cells of fish. No inhibition of enzymatic activity was observed in tunnel wash water exposed fish in the laboratory study and no difference was observed between fish sampled at the two locations in Årungenelva. The results the ALA-D biomarker indicated that trout were not exposed to lead at any extent. In the laboratory study, tunnel wash water from two tunnels, the Granfoss tunnel and the Nordby tunnel, was included. Stronger effects were observed in several of the investigated biomarkers in fish exposed to Nordby compared to fish exposed to Granfoss tunnel wash water. The two tunnels have similar annual average daily traffic (AADT), but the Granfoss tunnel is washed with a higher frequency. Washing frequencies may thus affect concentrations and the toxicity of road-related contaminants in tunnel wash water. In fish sampled in Årungenelva, it could not be concluded that fish sampled downstream from the sedimentation pond have experienced a higher exposure to road-related contaminants compared to fish sampled upstream from the pond. The findings of the current study could thus not relate the growth reduction previously observed in Årungenelva to the exposure of road-related contaminants. Due to severe rain the sampling in Årungenelva was postponed several times. Sampling closer to a tunnel wash event might have revealed a different pattern in the biomarkers investigated in fish from the stream.