Mineral dust is ubiquitous in the atmosphere and represents a dominant type of particulate matter by mass. Despite its well-recognized importance, assessments of dust impacts on clouds and climate remain highly uncertain. This thesis addresses the role of dust as cloud condensation nuclei (CCN) and giant CCN (GCCN) with the goal of improving our understanding of dust-warm cloud interactions and their representation in climate models. We investigate the CCN-relevant properties of mineral dust samples representative of major regional dust sources experimentally in the laboratory conditions to determine their respective affinity to water. Based on the experimental exponent derived from the dependence of critical supersaturation with particle dry diameter, we determine the dominant physics of activation (i.e., adsorption activation theory (AT) or traditional Köhler theory (KT)) for dust particles from different global regions. Results from experimental measurements are used to support the development of a new parameterization of cloud droplet formation from dust CCN for climate models based on adsorption activation mechanism. The potential role of dust GCCN activating by AT within warm stratocumulus and convective clouds is also evaluated.