Aerosol climate effects are intimately tied to interactions with water. Here we combine hygroscopicity measurements with direct observations about the phase of secondary organic aerosol (SOA) particles to show that water uptake by slightly oxygenated SOA is an adsorption-dominated process under subsaturated conditions, where low solubility inhibits water uptake until the humidity is high enough for dissolution to occur. This reconciles reported discrepancies in previous hygroscopicity closure studies. We demonstrate that the difference in SOA hygroscopic behavior in subsaturated and supersaturated conditions can lead to an effect up to about 30% in the direct aerosol forcing—highlighting the need to implement correct descriptions of these processes in atmospheric models. Obtaining closure across the water saturation point is therefore a critical issue for accurate climate modeling. Aerosols affect climate in two ways: aerosol-radiation interactions (ARI) (the direct effect) and aerosol-cloud interactions (ACI) (the indirect effect). While the indirect effect has received far more attention recently, the current Intergovernmental Panel on Climate Change Assessment Report 5 suggests nearly equal magnitudes and uncertainties for both effects (roughly 0.5 ± 0.5 W m!2 cooling in each case) [Boucher et al., 2013]. Secondary Organic Aerosol (SOA) often dominates aerosol mass in remote areas [Jimenez et al., 2009; Hallquist et al., 2009] and consists of a highly complex mixture of sometimes very viscous and sparingly soluble compounds [Petters et al., 2009; Renbaum-Wolff et al., 2013; Virtanen et al., 2010]. It is very likely that SOA was even more dominant in preindustrial times when water-soluble sulfates and inorganic nitrates were much less abundant. Understanding the direct and indirect effects of SOA is critical to narrowing the uncertainties in aerosol-climate interactions, which remain stubbornly the largest sources of uncertainty in climate forcing. ARI is due to light scattering and absorption by particles: Scattering depends on particle size, which increases dramatically when particles swell with water on a humid day or as they approach cloud base. ACI is due to light scattering by cloud droplets: Cloud-droplet number is controlled by the number of particles that activate as air rises through the condensation level and relative humidity (RH) exceeds 100%. Both subsaturated swelling and supersaturated activation are controlled by particle hygroscopicity, but the thermodynamic regimes and measurement methods are very different. Here we focus on the critical transition region linking these two effects, with important implications for pristine environments dominated by organic particulate matter and especially highly uncertain preindustrial conditions. That transition is the swelling of particles via water uptake as RH rises toward 100% and the subsequent growth to cloud droplets of a subset of those particles (the cloud condensation nuclei (CCN)). Especially below 100% RH, relevant to ARI, highly nonideal behavior by different SOA types strongly influences particle swelling (hygroscopic growth), with substantial implications for climate forcing. If solutions were ideal (with water activity equal to its mole fraction in the particle phase and also its vapor saturation ratio), subsaturated and supersaturated water uptake would be fairly straightforward, though