Methods and systems for plant performance analysis

A method for assessing wide-area performance of two or more crop varieties using a linear mixed model that incorporates geostatistical components and includes parameters for fixed effects, random effects and covariances, the method comprising:

• constructing a wide-area database that includes spatial coordinates of testing locations of two or more crop varieties, geographic regions for assessing performance of two or more crop varieties, each geographic region containing a plurality of geographic locations and performance trait values of two or more crop varieties at their respective testing locations;
• constructing a linear mixed model for the data in the wide-area database that includes modeling the large-scale performance of each of the two or more crop varieties by fixed effect trend surfaces, modeling spatial autocovariance among the random error components of each of the two or more crop varieties and modeling spatial crosscovariances of random errors among the two or more crop varieties in a mutivariate coregionalization model;
• estimating the parameters for the fixed effects, random effects and covariances by fitting the linear mixed model with data in the wide-area database;
• and
• estimating long-term expected performance of the two or more crop varieties for any of the one or more geographic regions using the parameter estimates.

A method for assessing wide-area performance of two or more crop varieties using a linear mixed model that incorporates geostatistical components and includes parameters for fixed effects, random effects and covariances, the method comprising:

• constructing a wide-area database that includes spatial coordinates of testing locations of two or more crop varieties, performance trait values of two or more crop varieties from one or more time periods at their respective testing locations and spatial coordinates of other geographic locations for assessing performance of two or more crop varieties;
• constructing a linear mixed model for the data in the wide-area database that includes modeling the large-scale performance of each of the two or more crop varieties by fixed effect trend surfaces, modeling spatial autocovariance among the random error components of each of the two or more crop varieties and modeling spatial crosscovariances of random errors among the two or more crop varieties in a mutivariate coregionalization model;
• estimating the parameters for the fixed effects, random effects and covariances by fitting the linear mixed model with data in the wide-area database;
• and
• predicting the performance of two or more crop varieties for any of one or more geographic locations and for any of one or more time periods using the parameter estimates.

A method for assessing wide-area performance of two or more crop varieties using a linear mixed model that incorporates geostatistical components and includes parameters for fixed effects, random effects and covariances, the method comprising:

• constructing a wide-area database that includes spatial coordinates of testing locations of two or more crop varieties, geographic regions for assessing performance of two or more crop varieties, each geographic region containing a plurality of geographic locations, and performance trait values of two or more crop varieties from one or more time periods at their respective testing locations;
• constructing a linear mixed model for the data in the wide-area database that includes modeling the large-scale performance of each of the two or more crop varieties by fixed effect trend surfaces, modeling spatial autocovariance among the random error components of each of the two or more crop varieties and modeling spatial crosscovariances of random errors among the two or more crop varieties in a mutivariate coregionalization model;
• estimating the parameters for the fixed effects, random effects and covariances by fitting the linear mixed model with data in the wide-area database;
• and
• predicting the performance of two or more crop varieties for any of one or more geographic regions and for any of one or more time periods using the parameter estimates.

A method for assessing wide-area performance of two or more crop varieties using a linear mixed model that incorporates geostatistical components and includes parameters for fixed effects, random effects and covariances, the method comprising:

• constructing a wide-area database that includes spatial coordinates of testing locations of two or more crop varieties, geographic regions for assessing performance of the two or more crop varieties, each geographic region containing a plurality of geographic locations, spatial coordinates of other geographic locations for assessing performance of two or more crop varieties and performance trait values of two or more crop varieties at their respective testing locations;
• constructing a linear mixed model for the data in the wide-area database that includes modeling the large-scale performance of each of the two or more crop varieties by fixed effect trend surfaces, modeling spatial autocovariance among the random error components of each of the two or more crop varieties and modeling spatial crosscovariances of random errors among the two or more crop varieties in a mutivariate coregionalization model;
• estimating the parameters for the fixed effects, random effects and covariances by fitting the linear mixed model with data in the wide-area database;
• estimating long-term expected performance of any of the one or more crop varieties for any of one or more geographic locations using the parameter estimates;
• estimating long-term expected performance of two or more crop varieties for any of one or more geographic regions using the parameter estimates;
• predicting the performance of two or more crop varieties for any of the one or more geographic locations and for any of one or more time periods using the parameter estimates;
• and
• predicting the performance of two or more crop varieties for any of the one or more geographic regions and for any of the one or more time periods using the parameter estimates.