Freshwater distribution in the Arctic Ocean: Simulation with a highresolution model and model-data comparison

Robert Newton, Peter Schlosser, Douglas G. Martinson, Wieslaw Maslowski

Research output: Contribution to journalArticlepeer-review

36 Scopus citations

Abstract

A high-resolution numerical simulation of the Arctic Ocean is analyzed in order to study the fate of river runoff and freshwater fluxes in the Arctic Ocean. The model is driven by realistic winds and thermodynamic forcing from the European Centre for Medium-range Weather Forecasting (ECMWF) Reanalysis data set. Dye tracers have been added to visualize the pathways followed by low-salinity water from the major Arctic rivers and Bering Strait Inflow. The model is spun up using repeated forcing with the 1979 annual cycle for 20 years; then the 1979 through 1998 atmospheric forcing is applied. Under the influence of the 1979 through early 1980s winds, a large plume of river runoff exits the broad Eurasian shelf seas along the Lomonosov Ridge. Starting in about 1985, the locus of shelf-to-basin transport shifts eastward to the Alpha-Mendeleyev ridge complex. This shift in the model output is related to changes in the sea-surface height (SSH) fields, which we attribute primarily to shifts in surface wind stresses. Model resolution, runoff inputs, and relaxation terms in the Lena River delta region are analyzed in detail to expose issues with model performance at boundaries with freshwater inflow. Suggestions are made for improving future simulations of river runoff in basin-scale models.

Original languageEnglish (US)
Article numberC05024
JournalJournal of Geophysical Research: Oceans
Volume113
Issue number5
DOIs
StatePublished - May 8 2008
Externally publishedYes

ASJC Scopus subject areas

  • Geochemistry and Petrology
  • Geophysics
  • Oceanography
  • Space and Planetary Science
  • Earth and Planetary Sciences (miscellaneous)

Fingerprint

Dive into the research topics of 'Freshwater distribution in the Arctic Ocean: Simulation with a highresolution model and model-data comparison'. Together they form a unique fingerprint.

Cite this