Worldwide studies show 80%–90% of all sediments eroded from watersheds is trapped within river networks such as reservoirs, ponds, and wetlands. To represent the impact of impoundments on sediment routing in watershed modeling, Soil and Water Assessment Tool (SWAT) developers recommend to model reservoirs, ponds, and wetlands using impoundment tools (ITs). This study evaluates performance of SWAT ITs in the modeling of a small, agricultural watershed dominated by lakes and wetlands. The study demonstrates how to incorporate impoundments into the SWAT model, and discusses and evaluates involved parameters. The study then recommends an appropriate calibration sequence, i.e., landscape parameters calibration, followed by pond/wetlands calibration, then channel parameter calibrations, and lastly, reservoir parameter calibration. Results of this study demonstrate not following SWAT recommendation regarding modeling water land use as an impoundment depreciates SWAT performance, and may lead to misplaced calibration efforts and model over‐calibration. Further, the chosen method to model impoundments’ outflow significantly impacts sediment loads in the watershed, while streamflow simulation is not very sensitive. This study also allowed calculation of mass accumulation rates in modeled impoundments where the annual mass accumulation rate in wetlands (2.3 T/ha/yr) was 39% higher than mass accumulation rate in reservoirs (1.4 T/ha/yr).

The Amazon River is the largest fluvial source of fresh water and sediment to the global ocean and has the longest tidally influenced reach in the world. Two major rivers, the Tapajós and Xingu, enter the Amazon along its tidal reach. However, unlike most fluvial confluences, these are not one‐way conduits through which water and sediment flow downstream towards the sea. The drowned‐river valleys (rias) at the confluences of the Tapajós and Xingu with the Amazon River experience water‐level fluctuations associated not only with the seasonal rise and fall of the river network, but also with semidiurnal tides that propagate as far as 800 km up the Amazon River. Superimposed seasonal and tidal forcing, distinct sediment and temperature signatures of Amazon and tributary waters, and antecedent geomorphology combine to create mainstem–tributary confluences that act as sediment traps rather than sources of sediment. Hydrodynamic measurements are combined with data from sediment cores to determine the distribution of tributary‐derived and Amazon‐derived sediment within the ria basins, characterize the sediment‐transport mechanisms within the confluence areas and estimate rates of sediment accumulation within both rias. The Tapajós and Xingu ria basins trap the majority of the sediment carried by the tributaries themselves in addition to ca 20 Mt year−1 of sediment sourced from the Amazon River. These findings have implications for the interpretation of stratigraphy associated with incised‐valley systems, such as those that dominated the transfer of sediment to the oceans during lowstands in sea level.

Abstract. This study describes and implements an integrated, multimedia, process-based system-level approach to estimating nitrogen (N) fate and transport in large river basins. The modeling system includes the following components: (1) Community Multiscale Air Quality (CMAQ), (2) Weather Research and Forecasting Model (WRF), (3) Environmental Policy Integrated Climate (EPIC), and (4) Soil and Water Assessment Tool (SWAT). The previously developed Fertilizer Emission Scenario Tool for CMAQ (FEST-C), an advanced user interface, integrated EPIC with the WRF model and CMAQ. The FEST-C system, driven by process-based WRF weather simulations, includes atmospheric N additions to agricultural cropland and agricultural cropland contributions to ammonia emissions. This study focuses on integrating the watershed hydrology and water quality model with FEST-C system so that a full multimedia assessment on water quality in large river basins to address impacts of fertilization, meteorology, and atmospheric N deposition on water quality can be achieved. Objectives of this paper are to describe how to expand the previous effort by integrating the SWAT model with the FEST-C (CMAQ/WRF/EPIC) modeling system, as well as to demonstrate application of the Integrated Modeling System (IMS) to the Mississippi River basin (MRB) to simulate streamflow and dissolved N loadings to the Gulf of Mexico (GOM). IMS simulation results generally agree with US Geological Survey (USGS) observations/estimations; the annual simulated streamflow is 218.9 mm and USGS observation is 211.1 mm and the annual simulated dissolved N is 2.1 kg ha−1 and the USGS estimation is 2.8 kg ha−1. Integrating SWAT with the CMAQ/WRF/EPIC modeling system allows for its use within large river basins without losing EPIC's more detailed biogeochemistry processes, which will strengthen the assessment of impacts of future climate scenarios, regulatory and voluntary programs for N oxide air emissions, and land use and land management on N transport and transformation in large river basins.