Skip to main content
ARS Home » Research » Publications at this Location » Publication #168057
Title: TERRAIN-BASED SCALING AND SIMULATION OF CROPPING SYSTEMS

Author
item Green, Timothy
item Erskine, Robert - Rob
item Ascough Ii, James
item Ahuja, Lajpat
item SALAS, JOSE - COLORADO STATE UNIVERSITY
item MARTINEZ, ANA - COLORADO STATE UNIVERSITY

Submitted to: Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 11/23/2003
Publication Date: 11/24/2003
Citation: Green, T.R., Erskine, R.H., Ascough Ii, J.C., Ahuja, L.R., Salas, J.D., Martinez, A. 2003. Terrain-based scaling and simulation of cropping systems. Meeting Proceedings. Integrative Modelling of Sustanable Land Use, ETH Zurich, Nov 24-25, 2003.

Interpretive Summary: Agriculture comprises much of the world's land area and is critical to environmental, economic, and social sustainability. We propose that differential spatial management of agricultural lands, considering space-time interactions, can improve both net production and environmental concerns. To investigate this topic, we have focused on cropping systems, particularly those in the Great Plains of the USA. Our approach is to first understand the space-time scaling behavior of crop yield in relation to soil water and topography. Fractal geometry is used to characterize the spatial organization of crop yield, near-surface (top 300 mm) soil moisture, and topographic attributes computed from high-resolution (5-m and 10-m spacing) elevation data. Next, a spatially correlated artifical neural network is used to predict crop yield from topographic attributes, showing the advantage of the neural network method over multiple linear regression. Potential yield zones are subsequently classified to delineate land areas within fields for spatial measurements and numerical simulations. Building upon currently available soil hydrology and agricultural systems models, we are working on a terrain-based model of interacting land areas, where surface runoff and lateral subsurface flow and chemicals are routed within a large field. As a prototype, the one-dimensional Root Zone Water Quality Model has been implemented spatially in a GIS framework, allowing for runoff-runon between delineated land areas. The ongoing development and implementation of such scientific advances in scaling and simulation will aid enhanced space-time management of agricultural systems.

Technical Abstract: Agriculture comprises much of the world's land area and is critical to environmental, economic, and social sustainability. We propose that differential spatial management of agricultural lands, considering space-time interactions, can improve both net production and environmental concerns. To investigate this topic, we have focused on cropping systems, particularly those in the Great Plains of the USA. Our approach is to first understand the space-time scaling behavior of crop yield in relation to soil water and topography. Fractal geometry is used to characterize the spatial organization of crop yield, near-surface (top 300 mm) soil moisture, and topographic attributes computed from high-resolution (5-m and 10-m spacing) elevation data. Next, a spatially correlated artifical neural network is used to predict crop yield from topographic attributes, showing the advantage of the neural network method over multiple linear regression. Potential yield zones are subsequently classified to delineate land areas within fields for spatial measurements and numerical simulations. Building upon currently available soil hydrology and agricultural systems models, we are working on a terrain-based model of interacting land areas, where surface runoff and lateral subsurface flow and chemicals are routed within a large field. As a prototype, the one-dimensional Root Zone Water Quality Model has been implemented spatially in a GIS framework, allowing for runoff-runon between delineated land areas. The ongoing development and implementation of such scientific advances in scaling and simulation will aid enhanced space-time management of agricultural systems.