Photo-chemical reactivity of dissolved organic carbon along a gradient of subarctic lakes with different color

Abstract

Lake-atmosphere carbon exchanges can be significantly affected by photochemical dissolved organic matter (DOM) mineralization. However, there is an incomplete understanding of how shifting optical characteristics of DOM due to increasing allochthonous organic carbon input affect the photo-reactivity of the DOM per unit of absorbed incoming light. Here, we measured the absorption of ultraviolet (UV) light and subsequent photochemical DOM decay in 148 lakes within the subarctic region of Abisko. These lakes range from brown-water lakes with allochthonous inputs connected to some mires to tundra clear-water lakes with relatively more autochthonous inputs. Fluorescence excitation-emission matrix analysis was used to assess the chemical composition of the DOM. The aim was to see how increasing colored DOM (CDOM) affects the photo-mineralization. We found that the photo decay rates in absolute values were positively correlated to CDOM. However, the photo decay per unit of absorbed light energy did not increase with increasing CDOM; rather, it showed a weak decreasing trend. Fluorescence analyses helped explain these patterns, as humic-like fluorescent DOM of presumable terrestrial origin was associated with high absolute photo decay rates, but not generally with higher photo-reactivity per unit of absorbed light energy than other types of DOM. The results suggest that even though increasing inputs of terrestrial substances mean a higher abundance of photo-degradable materials, it does not necessarily mean that CO₂ emissions increase in lakes where browning limits the ability of light to penetrate deeper water.

Key Words:

• Photo-reactivity
• DOM
• subarctic lakes
• EEM-PARAFAC
• Brownification

Disclaimer

As a service to authors and researchers we are providing this version of an accepted manuscript (AM). Copyediting, typesetting, and review of the resulting proofs will be undertaken on this manuscript before final publication of the Version of Record (VoR). During production and pre-press, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal relate to these versions also.

Acknowledgment

We thank Francois Guillemette for doing the PARAFAC modelling and analyses, and Cristian Gudasz for his valuable contribution and for the basic water chemistry measurements and for the map in figure 1. We thank the Royal Physiographic Society of Lund ’’fysiografen’’ and Helge Ax:son Johnsons stiftelse for funding this work. M.B has received a grant from the Swedish Research Council VR.

Fig. 1 The geographical distribution of the studied lakes.

Fig. 2. Initial and final incubation values for the dissolved organic matter (DOM) in mg carbon per L, colored dissolved organic matter (CDOM), fluorescent dissolved organic (FDOM), and ratios between them during 24 hours of laboratory UV light exposure of water from 148 subarctic lakes.

Fig. 3. The percentage of the photo-degraded DOM_lost plotted against CDOM on a log-log scale (n=146).

Fig. 4. The relationship between the photo decay in absolute values and the CDOM of subarctic lakes near Abisko, northern Sweden (n = 146).

Fig. 5. The relationship between the photo decay rate per absorbed energy and colored dissolved organic matter (CDOM). n=144.

Fig. 6. The change rate of the different components of the fluorescence in 24 hours of incubation. All components except C3 are significantly different from zero according to student two tails t-test for two samples with unequal variance. The number of samples n=145-146.

Table 1. Pearson r correlation linear values of the different measured variables in this study where. The ΔC is the difference between the initial and the final values (after 24 hours of UV light irradiance). Strongly positively skewed variables (Fisher’s skew > 2) have been log-transformed, and are shown in bold. The none-detected negative values of PD were excempted. n.s stands for not significant. * not significant correlation (p>0.1) ,** marignally significant correlation (0.05 < p > 0-1),*** the high significant correlation (p≤0.05).