Hypothesis: The wettability of complex fluids on surfaces usually depends on the adsorption of solutes to any of the constituting interfaces. Controlling such interfacial processes by varying the composition of a phase enables the design of smart responsive systems. Our goal is to demonstrate that 3D Confocal Raman Microscopy (CRM) can reveal the mechanistic details of such processes by allowing to simultaneously monitor the contact angle variation and redistribution of the chemical species involved. Experiments: Motivated by the enhanced oil recovery process of low salinity water flooding, we studied the response of picolitre oil drops on mineral substrates upon varying the ambient brine salinity. The substrates were pre-coated with thin layers of deuterated-stearic acid (surfactant) that display salinitydependent stability. Findings: 3D CRM imaging using a recently proposed faster 'ai' (algorithm-improved) mode reveals that the surfactant layer is stable at high salinities, leading to preferential oil wetting. Upon reducing the ambient brine salinity, this layer decomposes and the investigated surfaces of mica and - somewhat less pronounced - silica become more water wet. Eventually, the surfactant is found to partly dissolve in the oil and partly precipitate at the oil-water interface. We anticipate that ai-3D-CRM will prove useful to holistically study similar systems displaying reactive wetting. (C) 2020 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license.