2006 Philadelphia Annual Meeting (22–25 October 2006)

Paper No. 8
Presentation Time: 3:30 PM


LONDONO, Ana Cristina, Geology, Univ of Cincinnati, ML0013, Cincinnati, OH 45221, ana_cristinal@hotmail.com

The Wari (600-1000 AD), Inca (1450-1532 AD) agricultural terraces in Moquegua, southern Peru offer a very good setting for testing the validity of models of landscape evolution of wash-dominated hillslopes and the application to these models for investigating the erosional history and morphologic dating of hillslopes on a millennial scale. The use of these constructed earthworks reduces the uncertainties inherent in previous studies in which natural features of unknown or uncertain initial morphology and age were used. The absence of vegetation these slopes greatly attenuates or eliminates the influence of creep processes.

Erosion of the Wari terraces is more pronounced than the Inca terraces. Well defined channels of 20-40 cm of width and about 15 cm of depth are abundantly distributed along the slope and are quasi parallel to each other. Gravel and sand accumulated at the terrace bases indicates that, at least in the current stage of erosion, the terraces are a closed system. In the Inca terraces, the erosion is concentrated in small areas, mainly near the edge of the walls, accumulating sediments as fans in the subsequent step. Even though the initial morphology of both Wari and Inca terraces, indicates slight lateral and downslope inclination of the terrace tread to promote water drainage during irrigation, the observed pattern of erosion appears not to be influenced by it. Generally erosion is evenly distributed along the length of the terrace tread. Thus, while wash processes act on the slope, the downhill gradient of the terraces seems to predominate over the lateral inclination of the step.

Initial work on hillslope erosion patterns in Inca terraces using WEPP, a process-based erosion model, indicated that the more sensitive sections of the Inca terraces to erosion are the edges of the steps where rounding occurs. This is supported by field evidence that the terrace risers where flow is concentrated have been destroyed and the soil has been removed.