Black carbon (BC) has short lifetime but large spatio-temporal variability in the atmosphere. It is important to have accurate global 3-D distribution of BC to fully represent and/or assess its radiative and microphysical effects in climate models. Many climate models, including the Community Atmosphere Model version 5 (CAM5), have large biases in the prediction of global 3-D distribution of BC, particularly over high latitudes and in the upper troposphere. Many factors can affect the amount and impact of BC in a specific region such as the Arctic, among which the global distribution of emissions is of primary importance. We recently improved the treatment of aerosol convective transport and wet scavenging processes in CAM5, which largely reduces these biases and improves global spatial distributions of BC in the atmosphere and in snow over land. There are still many uncertainties in global BC emissions that also change over time. To better understand the response of climate to BC emission uncertainties, we further introduced an explicit BC source tagging technique in CAM5 to establish quantitative BC source-receptor relationships, characterize source-to-receptor transport pathways, and attribute BC loading and radiative forcing to various regional/sectoral BC sources. We found that BC lifetime, forcing-per-emission and forcing-per-loading have strong dependence on source regions. With the tagging tool we also quantified the sensitivity of BC in receptor regions to uncertainties or changes in various regional sources in a single simulation without perturbing the BC emissions. The established source-receptor relationships and quantified burden/forcing sensitivities will be used to assess potential climate impacts of uncertainties in BC emissions from various sources.