Abstract
In the last decade, there has been increasing use of optical measurements to gain insight into dissolved organic matter (DOM) composition, and more specifically to identify DOM source in aquatic systems. However, there are few controlled studies which examine the effects of environmental processing on different sources of DOM. Here, DOM optical properties of five endmember sources–including peat soil, plant and algae leachates–were investigated following biological and photochemical degradation during a three-month incubation period. As microbial processing of DOM occurs independent of being in the photic zone, the effects of photoexposure were examined at various points along the biodegradation curve to simulate photodegradation occurring as microorganisms consumed and transformed the bioavailable DOM. Samples were analyzed for dissolved organic carbon (DOC) concentration, absorbance, and fluorescence. Optical parameters commonly used to determine DOM composition were then analyzed to determine their effectiveness in discriminating original DOM source and in distinguishing between microbial and photochemical alteration. The qualitative optical parameters included DOC-normalized absorbance values (SpA254, SpA280, SpA350, SpA370 SpA412, SpA440, SpA488, SpA510, SpA532, SpA555); absorbance spectra slopes (S275-295, S290-350 and S350-400) and the UV slope ratio (SR); DOC-normalized fluorescence peaks (SpA, SpC, SpM, SpD, SpB, SpT, SpN, SpZ), fluorescence peak ratios (C:M, C:T, C:A, A:T); fluorescence indices (FI, HIX, β:α, BIX); and five components determined by Parallel Factor Analysis (PARAFAC; %C1-5). While there was little change in DOC concentration in the soil leachate over the study period, DOC concentrations in plant and algae leachates decreased by over 70% within the first three days of biodegradation. This rapid loss of DOC in the plant and algal leachates suggests the majority of DOM leached from these materials is unlikely to persist in the environment, and thus is unlikely to make up a significant fraction of the DOM pool in most natural samples. This emphasizes the need to use the signature of processed DOM to identify original source material. Individual qualitative optical parameters changed extensively as DOM composition was altered following both biodegradation and photodegradation, particularly in the plant and algal leachates. These changes frequently resulted in overlapping optical parameter values which made it impossible to identify original source material. In particular, the sometimes opposing effects of biological and photochemically-driven changes on DOM optical signature can confound source identification; for example, this effect of one degradation process masking the signal from the other was notably apparent for SUVA where values increased with biological degradation and decreased following photoexposure, suggesting that using this parameter alone can generate inconsistent and disparate conclusions about DOM composition and source. In addition to examination of parameters individually, multivariate statistical analyses were used to determine whether used in combination, these parameters could identify unique optical signatures that could be linked to original DOM source even after exposed to biological and photochemical alteration. PCA demonstrated that when the suite of 30 parameters were combined, the optical signature of the materials did not fall out clearly by source and environmental processing; as was seen when examining the individual parameters, optical signatures of the different sources overlapped over time, with the effects of biodegradation and photodegradation often acting in opposition. The trajectory in PCA space did however generally follow what is expected as DOM undergoes degradation: a shift from fresh-like to humic-like material. Discriminant analysis (DA) was used to identify which qualitative indicators are the most promising for distinguishing DOM source and processing. Of the 30 qualitative indicators evaluated, 17 were quantitatively determined by DA to be the most significant (p < 0.05): absorbance parameters included SUVA, SpA350, SpA412, S275-295, and S290-350, while fluorescence parameters included humic-like (SpC, SpM, SpD, SpZ) and fresh-like (SpB, SpT, SpN) DOC-normalized fluorescence peaks as well as peak ratios (C:A, C:M) and indices (FI, HIX, β:α). The classification of DOM source (soil, rice, cattail, tule, algae) was influenced most heavily by SpC, SpM, C:M, and HI, while the classification of DOM processing (biodegradation versus photoexposure) was influenced most by SpN, SpD, SpT and SpA350. This dataset highlights the challenge of using optical properties to identify DOM source material because the effects of biodegradation and photodegradation, which in the natural environment can occur simultaneously, can lead to confounding results. Moreover, samples collected from the environment typically contain a mixture of DOM sources which have undergone different degrees of processing. In natural systems, multiple parameters and careful consideration should be taken when using optical properties to characterize DOM source.