Development of a parallel river transport algorithm and applications to climate studies
MetadataShow full item record
The global hydrologic cycle plays a central role in the interactive functioning of the Earth’s climate system. The theme of this study is land-oceanatmosphere interaction. Continental runoff has a notable effect on the global hydrologic cycle, both directly as freshwater forcing on the oceans and indirectly through effects on global patterns of precipitation and coincident feedbacks to continental runoff. To demonstrate this, a series of three projects involving both observations and modeling were completed. The first phase involved the development of a parallel river transport model to deliver runoff from the land surface to the oceans at the appropriate location and time. Within each watershed, the river routing algorithm used cell-to-cell routing by considering the mass balance of surface inflows and outflows. This river transport model was then incorporated into a climate system model. The second and third phases involved the use of a climate system model to investigate the effect of continental runoff. To test the sensitivity of the oceans to freshwater input from runoff, in the second phase, a number of 70-year simulations were conducted, using the ocean and ice components of a highresolution climate model with observed runoff and modeled atmospheric forcing. A half-degree observed runoff data set, consisting of both annual and monthly averages, was used for this forcing. Differences in sea surface temperature and salinity between simulations with and without the addition of runoff were found in the Arctic, Tropical Atlantic, and North Atlantic Oceans. The differences were especially pronounced in the North Atlantic. By affecting sea surface temperatures and salinity, the addition of freshwater from river runoff led to a reduction in North Atlantic Deep Water formation and a corresponding slowdown of heat transport. The third phase used a fully coupled land-ocean-atmosphere-ice model plus the river transport model from the first phase. Two 200-year simulations were conducted, with and without the river component. The simulation with rivers had reduced oceanic meridional heat transport. Reduced convective rainfall and runoff during January in the simulation with rivers indicated a feedback from the continental runoff flux into the oceans back to the land surface.