Cumulative distribution function (CDF) plots of observed spring wheat yields for Melfort, Swift Current and Lacombe. The power-law (A) and log-normal (B) distributions are used as theoretical distributions. The line of best fit is also represented. Kolmogorov-Smirnov (KS) statistic and p-values are computed according to the methodology proposed by Clauset et al. (2009). 2500 synthetic data sets were generated for the computation. Good fits are characterized by small KS and p-value < 0.1.