Century-long increasing trend and variability of dissolved organic carbon export from the Mississippi River basin driven by natural and anthropogenic forcing
Metadata
Show full item recordAbstract
There has been considerable debate as to how natural forcing and anthropogenic activities alter the timing and magnitude of the delivery of dissolved organic carbon (DOC) to the coastal ocean, which has ramifications for the ocean carbon budget, land-ocean interactions, and coastal life. Here we present an analysis of DOC export from the Mississippi River to the Gulf of Mexico during 1901-2010 as influenced by changes in climate, land use and management practices, atmospheric CO2, and nitrogen deposition, through the integration of observational data with a coupled hydrologic/biogeochemical land model. Model simulations show that DOC export in the 2000s increased more than 40% since the 1900s. For the recent three decades (1981-2010), however, our simulated DOC export did not show a significant increasing trend, which is consistent with observations by U.S. Geological Survey. Our factorial analyses suggest that land use and land cover change, including land management practices (LMPs: i.e., fertilization, irrigation, tillage, etc.), were the dominant contributors to the century-scale trend of rising total riverine DOC export, followed by changes in atmospheric CO2, nitrogen deposition, and climate. Decadal and interannual variations of DOC export were largely attributed to year-to-year climatic variability and extreme flooding events, which have been exacerbated by human activity. LMPs show incremental contributions to DOC increase since the 1960s, indicating the importance of sustainable agricultural practices in coping with future environmental changes such as extreme flooding events. Compared to the observational-based estimate, the modeled DOC export was 20% higher, while DOC concentrations were slightly lower. Further refinements in model structure and input data sets should enable reductions in uncertainties in our prediction of century-long trends in DOC.