Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Genetic algorithm for parameter and scale selection to predict soil moisture patterns

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2008 Providence, Rhode Island, June 29 – July 2, 2008  083733.(doi:10.13031/2013.24894)
Authors:   Lingyuan Yang, Amy L Kaleita
Keywords:   Evolutionary computation, scaling, soil water content

Soil moisture is a critical component of hydrological processes, and its spatio-temporal distribution depends on many geographical factors (such as elevation, slope, and aspect, etc.). Each of the factors is likely influential over a different scale and to a different degree. Near-surface soil moisture data were collected across a working 10-ha field southwest of Ames, IA in growing seasons of 2004 to 2007. A genetic algorithm is developed to compare geographical factors to the moisture patterns over a range of scales. The genetic algorithm will develop a model in which each factor is computed over a different scale for use in prediction of reference variable. Optimized scales for each parameter are arrived at through successive generations, including crossover and mutation of this evolutionary algorithm. Using this approach, not only are the primary influential relationships uncovered, but the most appropriate scale for comparison to moisture pattern is identified. The results of this analysis can be used to predict the spatio-temporal patterns of soil moisture across a region a priori.

(Download PDF)    (Export to EndNotes)