In 1999 the AWI established the HAUSGARTEN observatory, to assess the impact of climate change on Arctic ecosystems in Fram Strait (Arctic), which included repeated camera transects to assess changes on the deep Arctic seafloor. A first analysis of the footage highlighted that marine debris increased over time. Plastic debris was also sighted during sea surface observations for seabird surveys. This prompted us to add a pollution observatory to the ongoing research programme FRAM, aiming to quantify plastic pollution in different ecosystem compartments to identify hidden sinks. Here, we summarise the results of this work encompassing matrices such as snow, sea ice, surface waters, water column, deep seafloor, biota and Arctic beaches. Images from the deep seafloor taken since 2002 showed a marine debris concentration of 4,571 ± 1,628 items km-2, which is in range with polluted oceanic regions. Visual surveys of floating debris from the same region revealed 500 times lower concentrations (9 items km-2), showing that the deep Arctic seafloor constitutes a sink for marine debris. Quantities of 9–483 g m-2 were reported from 15 beach surveys on Svalbard by citizen scientists. Plastics accounted for >80% of the mass, primarily from fisheries. Microplastics in samples from the sea surface, water column, sediment, sea ice and snow were analysed by combining state-of–the-art sampling technology with µFT-IR analyses. Using the same analysis for samples from different ecosystem compartments enabled us to determine the vertical distribution of microplastics, as sea ice entrains extremely high microplastic concentrations, which are released to the underlying waters during ice melts. In-situ pump-filtrations throughout the water column revealed that microplastics prevail at all depths in Fram Strait (0–1,287 items m–3). Microplastic concentrations in sediments ranged from 239–13,331 N kg–1. Highest microplastics concentrations in sediments and the water column were measured close to the marginal ice zone and polymer compositions indicated a sea ice origin for most particles found in the deep waters of East Greenland, indicating sea ice as a temporal sink. Indeed, the highest concentration (1.2 ± 1.4) ×107 items m-3) was recorded in an ice core from pack ice of Fram Strait. The presence of microplastic in snow samples from ice floes indicates atmospheric deposition of microplastics. Recent research shows that resident zooplankton ingests microplastics, which were also found in the ice algae Melosira arctica. The data indicate that the seafloor and sea ice constitute (temporal) sinks of plastic pollution and that pollution levels are high, despite of the distance to sources. The receding sea ice has already led to increased anthropogenic pressure in the Fram Strait, which is likely to become a major shipping lane during summer. The number of fishers operating around Svalbard and of ship calls to Longyearbyen has already increased significantly. In addition, the prevailing hydrography promotes the transport of plastic pollutants from distant sources, mostly from the Atlantic Ocean, but also from the Central Arctic via the Transpolar Drift. Long-range atmospheric transport and deposition likely adds to this.