Skip to main content
ARS Home » Research » Publications at this Location » Publication #190910
Title: RELATING CROP YIELD TO TOPOGRAPHIC ATTRIBUTES USING SPATIAL ANALYSIS NEURAL NETWORKS AND REGRESSION

Author
item Green, Timothy
item SALAS, JOSE - COLORADO STATE UNIVERSITY
item MARTINEZ, ANA - COLORADO STATE UNIVERSITY
item Erskine, Robert - Rob

Submitted to: Geoderma
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/18/2006
Publication Date: 2/21/2007
Citation: Green, T.R., Salas, J., Martinez, A., Erskine, R.H. 2007. Relating crop yield to topographic attributes using spatial analysis neural networks and regression . Geoderma.

Interpretive Summary: Land-surface topographic attributes can be useful for estimating stable spatial patterns of crop yield. We present spatial analyses of grain yield for three fields of dryland winter wheat in northeastern Colorado using topographic attributes as input or explanatory variables. Topographic attributes including relief, slope, aspect, curvature, specific contributing area, and wetness index are computed from a 10-m digital elevation model. A Spatial Analysis Neural Network (SANN) algorithm is used for joint spatial interpolation and yield prediction from topographic attributes. SANN prediction errors are compared with the results of multiple linear regression (MLR) relationships. SANN out-performed MLR in multivariate estimation, but not for the univariate cases. The greatest advantage of SANN was seen using four or more topographic attributes, whereas MLR showed diminishing returns with more than three explanatory variables. Prediction/interpolation errors within a given field were reduced substantially by using the spatial coordinates (latitude and longitude) in tandem with topographic attributes. Thus, we have demonstrated the combined utility of topographic attributes with SANN for estimating spatial patterns of dryland crop yield.

Technical Abstract: Land-surface topographic attributes can be useful for estimating stable spatial patterns of crop yield. We present spatial analyses of grain yield for three fields of dryland winter wheat in northeastern Colorado using topographic attributes as input or explanatory variables. Topographic attributes including relief, slope, aspect, curvature, specific contributing area, and wetness index are computed from a 10-m digital elevation model. A Spatial Analysis Neural Network (SANN) algorithm is used for joint spatial interpolation and yield prediction from the topographic attributes. SANN prediction errors are compared with the results of multiple linear regression (MLR) relationships. SANN and MLR are assessed in terms of bias and relative root mean squared error (rRMSE) using validation data. SANN out-performed MLR in multivariate estimation, but not for the univariate cases. The greatest advantage of SANN was seen using four or more topographic attributes, whereas MLR showed diminishing returns with more than three explanatory variables. Prediction/interpolation errors within a given field were reduced substantially by using the spatial coordinates (latitude and longitude) in tandem with topographic attributes. The rRMSE value reached a minimum of 0.44 (model efficiency, E = 0.80) for interpolation with SANN on the West field. Using only topographic attributes as input, the minimum rRMSE values were 0.59 (E = 0.65) for SANN with 5 variables and 0.72 (E = 0.48) for MLR with 4 or 5 explanatory variables. Thus, we have demonstrated the combined utility of topographic attributes with SANN for estimating spatial patterns of dryland crop yield.