*2.6. Statistical Design and Analysis*

Year, irrigation, row spacing, and tillage were treated as fixed effects. Analysis of variance of main effects utilized Proc MIXED and GLM of the Statistical Analysis System, Version 9.4 (SAS Institute, Cary, NC, USA). Linear regression and analysis of covariance utilized Proc REG and Proc GLM with contrast statements to determine differences among slopes and intercepts of the regression equations [26]. Significance was determined at *p* < 0.05 for all means separation tests and used the P diff statement with Tukey's adjustment for all main effects.

Interactions for all main effects were initially tested. When no interactions were detected, only main effect results are presented. Years were analyzed and are presented separately for analysis of variance as a consequence of management differences between the seasons: planting in 2016 was into native perennial grass, into wheat stubble in 2017; different seed used in each year. Years were combined for regression analysis after observing similar trends for both years.
