Author
 |
LESCH, SCOTT - UC RIVERSIDE, CA |
|
Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 1/2/2004 Publication Date: 3/1/2005 Citation: Lesch, S.M. 2005. Sensor-directed spatial response surface sampling designs for characterizing spatial variation in soil properties. Computers and Electronics in Agriculture. 46:153-179. Interpretive Summary: In many applied precision farming applications, remotely sensed survey data are collected specifically because this data correlates well with some soil property of interest. The purpose of this paper is to present a unified statistical sampling and modeling strategy for predicting soil property information from such spatially referenced sensor data. Various spatial linear prediction models and linear geostatistical models are reviewed, and the assumptions needed to reduce these more complicated models to a spatially referenced, ordinary linear regression model are discussed. A deterministic sampling strategy for estimating an ordinary linear regression model in the spatial setting is then described. This strategy can in principal be used to select a minimal number of optimal sample site locations for estimating an ordinary (spatially referenced) regression model. Relevant residual diagnostic tests and prediction statistics are also reviewed, and then a detailed case study of a salinity survey using electromagnetic induction and four-electrode sensor data is presented. The results presented in this study suggest that this methodology should be applicable to many types of precision farming survey applications where soil property / sensor data prediction models need to be estimated using only a limited number of soil samples. Technical Abstract: In many applied precision farming applications, remotely sensed survey data are collected specifically because this data correlates well with some soil property of interest. Additionally, a general model for the functional relationship between the soil property and sensor data is often known a priori, but the exact parameter estimates associated with the model must still be determined via some type of site-specific sampling strategy. The main objective of this paper is to present a unified sampling and modeling strategy for predicting soil property information from such spatially referenced sensor data. Some common types of spatial linear prediction models and linear geostatistical models are reviewed, and the assumptions needed to reduce these more complicated models to a spatially referenced, ordinary linear regression model are discussed. A deterministic sampling strategy for estimating an ordinary linear regression model in the spatial setting is then described. This strategy can in principal be used to select a minimal number of optimal sample site locations that satisfy the residual independence assumptions in the ordinary model. Relevant residual diagnostic tests and prediction statistics are also reviewed, and then a detailed case study of a salinity survey using electromagnetic induction and four-electrode sensor data is presented. This case study is used to demonstrate the overall modeling and sampling methodology, and the effectiveness of the sampling strategy. |
