Influence Of Agricultural Landscape On Riverine Dissolved Organic Carbon Loading Rates During Storm Events

Dissolved Organic Carbon(DOC) is the largest organic pool in the water environment and the most active and sensitive index in carbon cycling. The river is the strong link of terrestrial and ocean ecosystems. The riverine carbon cycling characterizes the terrestrial activities including natural process and human activity. In addition, many researches focus on the riverine DOC concentration, flux and their variations, which is vital for the knowledge of regional and global carbon cycling.In this study, DOC concentrations were measured over 2 years from March 2013 to October 2015 in Chippewa River, Michigan, United States, by choosing 6 monitoring sections when there were rainfall events. At the same time, we discussed the variations of DOC concentrations in spring and autumn during rainfall events in different scales and analyzed the influence factors. Apart from this, we modeled based on Weibull Distribution to estimate DOC concentration when there were not enough sampling data. We calculated DOC flux in different rainfall events. The main conclusions are as follows:(1) The study revealed differences of DOC concentrations in spring and autumn seasons. The DOC concentrations in spring were higher than that in autumn. The range of DOC concentration was from 4.0 to 9.48mg/L during the study period. Specifically, the range of DOC concentrations were 5.0~9.48mg/L and 4.0~7.0mg/L in spring and autumn, respectively. The results showed that rainfall related discharge was an important factor to direct the DOC, however, the agricultural residues, corn straw covering and ploughing-free practice, and rich deciduous fallen leaves also played a substantial role during this process.(2) The relationship between DOC concentration and discharge presented seasonal trend during storm events. The DOC concentration reached its maximum value before the peak of discharge in spring but it showed the opposite trend in autumn. Snow-melting was the key characteristic during this hydrological process.(3) A semi-analytical model was applied to estimate DOC concentrations when the sampling data were not enough. The model had been verified by loads of field sampling data, and the range of R2 was from 0.56 to 0.9. Using this model, DOC flux was estimated. The results showed that 70% annual DOC flux was exported during storm events. About 78.9% of DOC flux exported by rainfall was occurred in spring and 74.2% presented in autumn. In addition, DOC flux also conducted seasonal tendency: DOC flux exported by storm events in spring was twice than that in autumn. If the precipitation became stronger, the difference value would be more severe. Furthermore, we estimated the DOC flux which assumed if the biggest storm event of the study area would occur in spring and autumn. The result indicated that the exported DOC in spring would be 2.3 times in autumn if this assumption would occur.