Multiple organic substances present in agricultural and natural environments are fluorescent, including proteins, bacteria, pigments, oxidized aromatic substances such as humic-like material, phenols or phenolic acids. Soils and sediments, surface and groundwater, wastewater and aerosols, plant materials and food products are known to contain diverse fluorescent substances. Depending on specific environment, they play important roles, e.g., revealing microbial activity, buffering and controlling light penetration in water, affecting soil structure and stability, influencing distribution of agrochemicals and contaminants in soil-plant-water-air continuum or taste, smell and color of food products. A traditional spectroscopy intended to characterize the presence of fluorescent components had developed from measuring excitation, emission or synchronous spectra to obtaining three-dimensional excitation-emission matrices (EEMs) analyzed by parallel factor analysis (PARAFAC). The result of this development is a possibility to perform “a mathematical chromatography”, i.e., chemometric separation of EEMs into major chemically meaningful families of fluorescent substances, without any assumptions on shape of spectra. The presentation brings a series of demonstrations on how this methodology is working, by considering the extracts from soil irrigated by either tap or wastewater, or amended by external organic matter sources, examining natural water, such as sampled in Kinneret lake and Kishon river, treated wastewater and food product (olive oil). One strong advantage of this methodology is that it allows not only organic matter fingerprinting but also helps to eliminate overlapping of signals from different components. The meaningful result, as distinct from other chemometric methods, is due to the fact that mathematic structure of PARAFAC analysis coincides with a physical trilinear effect of light excitation, emission and concentration of the components. Thus, when and if the role of fluorescent components is important, EEM+PARAFAC might be a powerful solution giving us components to correlate with the processes of interest.