Continental scale models of water balance and fluvial transport: An application to South America


  • Charles J. Vörösmarty,

  • Berrien Moore III,

  • Annette L. Grace,

  • M. Patricia Gildea,

  • Jerry M. Melillo,

  • Bruce J. Peterson,

  • Edward B. Rastetter,

  • Paul A. Steudler


A coupled water balance and water transport model (WBM / WTM) was constructed as part of a larger study of global biogeochemistry. The WBM / WTM provides critical hydrologic information to models of terrestrial primary production, organic matter decay, riverine nutrient flux and trace gas exchanges with the troposphere. Specifically, it creates high-resolution data sets for monthly soil moisture, evapotranspiration, runoff, river discharge and floodplain inundation. As a first step toward eventual global coverage, the WBM / WTM was applied to South America, represented by more than 5700 1/2° (latitude / longitude) grid cells. The WBM transforms spatially complex data on long-term climate, vegetation, soils and topography into predictions of soil moisture (SM), evapotranspiration (ET) and runoff (RO). For South America, field capacity in soils ranged from 27 to 582 mm of water, and computed values for mean annual SM, ET and RO were 284 mm, 1059 mm/yr and 619 mm/yr, respectively. There were large differences regionally and over the year. The transport model uses WBM-derived runoff, information on fluvial topology, linear transfer through river channels and a simple representation of floodplain inundation to generate monthly discharge estimates for any cell within a simulated catchment. The WTM successfully determined the timing and magnitude of discharge at selected locations within the Amazon / Tocantins basin. It also demonstrated the importance of floodplain inundation in defining flow regime on the mainstem Amazon. Estimated mean annual discharge was 207,000 m3/s for the Amazon River and 17,000 m3/s for the Tocantins. In these basins, 45% of the incident precipitation emerges as river flow; 55% is lost to ET. The model described in this paper will be expanded to include the dynamics of carbon, major nutrients and sediments. It will serve as a semimechanistic tool to quantify the transport of materials from the landscape to the world's oceans. Such a capability becomes increasingly important as we seek to understand the impacts of climate and land use change on major river systems of the globe.