Aerosols change the Earth's energy balance. Black carbon (BC) aerosols are a product of incomplete combustion of fossil fuels and biomass burning and cause a net warming through aerosol radiation interactions (ari) and aerosol cloud interactions (aci). BC aerosols have potentially strong implications on the Arctic climate, yet the net global climate effect of BC is very uncertain. Best estimates assume a net warming effect, roughly half to that of CO2. However, the time scales during which CO2 emissions affect the global climate are on the order of hundreds of years, while BC is a short-lived climate pollutant (SLCP) with atmospheric life times of days to weeks. Climate models or atmospheric transport models struggle to emulate the seasonality and amplitude of BC concentrations in the Arctic, which are low in summer and high in winter/spring during the so called Arctic haze season. The high uncertainties regarding BC's climate impact are not only related to ari and aci, but also due to model parameterizations of BC lifetime and transport, and the highly uncertain estimates of global and regional BC emissions. Given the high uncertainties in technology-based emission inventories (EI), there is a need for an observation-based assessment of sources of BC in the atmosphere. We study short-term and long-term observations of elemental carbon (EC), the mass-based analog of optically-defined BC. EC aerosol concentrations and carbon-isotope-based (δ13C and ∆14C) sources were constrained (top-down) for three Arctic receptor sites in Abisko (northern Sweden), Tiksi (East Siberian Russia), and Zeppelin (on Svalbard, Norway). The radiocarbon (∆14C) signature allows to draw conclusion on the EC sources (fossil fuels vs. biomass burning) with high accuracy (&lt;5% variation). Stable carbon isotopic fingerprints (δ13C) give qualitative information of the consumed fuel type, i.e. coal, C3-plants (wood), liquid fossil fuels (diesel) or gas flaring (methane and non-methane hydrocarbons). These fingerprints can be used in conjunction with Bayesian statistics, to estimate quantitative source contributions of the sources. Finally, our observations were compared to predictions from a state of the art atmospheric transport model (coupled to BC emissions), conducted by our collaborators at NILU (Norwegian Institute for Air Research). Observed BC concentrations showed a high seasonality throughout the year, with elevated concentrations in the winter, at all sites. The highest concentrations were measured on Svalbard during a short campaign (Jan-Mar 2009) focusing on BC pollution events. Long-term observations showed that Svalbard (2013) had overall the lowest annual BC concentrations, followed by Abisko (2012) and Tiksi (2013). Isotope constraints on BC combustion sources exhibited a high seasonality and big amplitude all across the Eurasian Arctic. Uniform seasonal trends were observed in all three year-round studies, showing fractions of biomass burning of 60-70% in summer and 10-40% in winter. Europe was the major source region (&gt;80%) for BC emissions arriving at Abisko and the main sources were liquid fossil fuels and biomass burning (wood). The model agreed very well with the Abisko observations, showing good model skill and relatively well constrained sources in the European regions of the EI. However, for the Svalbard and East Siberian Arctic observatories the model-observation agreement was not as good. Here, Russia, Europe and China were the major contributors to the mostly liquid fossil and biomass burning BC emissions. This showed that the EI still needs to be improved, especially in regions where emissions are high but observations are scarce (low ratio of observations to emitted pollutant quantity). Strategies for BC mitigation in the (Eurasian) Arctic are probably most efficient, if fossil fuel (diesel) emissions are tackled during winter and spring periods, all across Eurasia. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 2: Manuscript. Paper 3: Manuscript.</p>