UV-visible absorbance spectroscopy as a proxy for peatland dissolved organic carbon (DOC) quantity and quality: considerations on wavelength and absorbance degradation

Absorbance in the UV or visible spectrum (UV-vis) is commonly used as a proxy for DOC concentrations in waters draining upland catchments. To determine the appropriateness of different UV-vis measurements we used surface and pore water samples from two Welsh peatlands in four different experiments: (i) an assessment of single wavelength proxies (1 nm increments between 230-800 nm) for DOC concentration demonstrated that 254 nm was more accurate than 400 nm. The highest R-2 values between absorbance and DOC concentration were generated using 263 nm for one sample set (R-2 = 0.91), and 230 nm for the other three sample sets (respective R-2 values of 0.86, 0.81, and 0.93). (ii) A comparison of different DOC concentration proxies, including single wavelength proxies, a two wavelength model, a proxy using phenolic concentration, and a proxy using the area under a UV spectrum at 250-350 nm. It was found that both a single wavelength proxy (<= 263 nm) and a two wavelength model performed well for both pore water and surface water. (iii) An evaluation of the E2 : E3, E2 : E4, E4 : E6 ratios, and SUVA (absorbance at 254 nm normalised to DOC concentration) as indicators of DOC quality showed that the E4 : E6 ratio was subject to extensive variation over time, and was highly correlated between surface water and pore water, suggesting that it is a useful metric to determine temporal changes in DOC quality. (iv) A repeated weekly analysis over twelve weeks showed no consistent change in UV-vis absorbance, and therefore an inferred lack of degradation of total DOC in samples that were filtered and stored in the dark at 4 degrees C.