Surface water models: Run-off and Sediment Yield 

Two surface water models, Soil and Water Assessment Tool (SWAT) and Kinematic Runoff and Erosion Model (KINEROS), were used to simulate relative runoff and sediment yield in each of the watersheds in the Loreto region.  These two models are incorporated into AGWA or Automated Geospatial Watershed Assessment Tool.  AGWA is a GIS (Geographic Information System) based hydrologic modeling tool that is designed for performing relative assessments (change analysis) resulting from land cover/use change (Burns et al. 2005).  Land cover data was obtained from Conabio- Comision nacional para el conocimiento Y uso de la biodiversidad.  Storm data was obtained from Servicio Meteorogico Nacional. Comision Nacional Del Agua.

SWAT is a continuous-time model developed at the USDA-ARS to predict the impact of land management practices on water, and sediment yields in large (basin scale) complex watersheds with varying soils, land use and management conditions over long periods of time (> 1 year).  SWAT will provide large scale assessment of the status of surface water hydrology in the watershed and is useful for ungaged watersheds and watersheds with limited data such as ours.  We used a 10-m Digital Elevation Model (DEM) and AGWA to delineate the 15 small watersheds in the study area (Figure 15).  Land cover and FAO soils data (Food and Agriculture Organization of the United Nations-digital soil map of the world) of the region, and 13 years of daily precipitation data (1978-1990) were used to simulate runoff and sediment yield in each of the watersheds.  Since there are no data available to calibrate the models, the runoff and sediment yield values are not absolute values, and are best suited for comparing the effect of land use change on in the watersheds.

To delineate the flood zones and determine sediment yield from large single storm events KINEROS was used.  KINEROS is a physically-based model designed to simulate runoff and erosion from single storm events in small watersheds less than about 100 km².  DEM (25m), FAO soils, land cover data and rainfall time series for single storm events (10, 25, and 50 year return period) were used to simulate runoff and sediment yield for each event in each basin.  The estimated runoff values at the main channel intersections were used to delineate the flood plain boundaries for different storm events.  These maps were then used as part of the landscape assessment and planning process.  It should be noted that the accuracy of surface water models is dependent upon the accuracy and scale of surface elevation information.  For the Loreto region only a 25 meter DEM was available.  This scale worked well for the steeper portions of the watershed but gave only approximated results in the flatter areas.  Unfortunately, these are the areas of concern.  This lack of resolution continues to be problematic in rural areas with only sparse data.  Sediment yield values at the watershed outlet were classified to three ordinal classes and status of each watershed in terms of sediment values for 50 year storm event presented (Figure 16).  The sediment yield classes were then used by the marine ecologists to assess impacts on marine habitats in the project.