Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

The Best Distributions for Maximum and Minimum Daily Temperatures of Columbus, Ohio

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  2007 ASAE Annual Meeting  072298.(doi:10.13031/2013.23656)
Authors:   Tekin ÖZTEKIN, Sedat KARAMAN, Larry C BROWN
Keywords:   Keywords: Maximum and minimum daily temperature, normality, FRANMOD model, Columbus

In this study, the normality and best fitting distributions for the daily maximum and minimum temperatures measured at the Columbus, Valley-Crossing during the periods of 1916-1949 and 1980-2006 were searched. To represent four seasons, the temperatures of months of January, April, July, and October were considered. The Kolmogorov-Smirnov normality test (P = 0.05) for the normality analysis, and FRANMOD model for determining the best fitting distributions among 36 continuous distributions (2-P (Parameter) standard normal, 2-P log-normal, 3-P log-normal, 2-P gamma, 3-P Pearson type 3, 3-P log-Pearson type 3, 3-P U.S. Water Resources Council method, 2-P Pearson type 5, 3-P generalized gamma, 1-P exponential, 2-P exponential, 2-P Gumbel (maxima), 2-P Gumbel (minima), 2-P Frechet (maxima), 3-P Frechet (minima), 3-P Frechet (maxima), 2-P Weibull (minima), 3-P Weibull (minima), 3-P Weibull (maxima), 3-P generalized extreme value, 4-P two-component extreme value, 2-P log-Gumbel, 3-P log-Gumbel, 2-P logistic, 3-P generalized logistic, 2-P log-logistic, 2-P standard beta, 4-P generalized beta, 2-P Pareto, 2-P generalized Pareto, 3-P generalized Pareto, 5-P Wakeby, 3-P beta-kappa, 3-P beta-P, 2-P kappa, and, 4-P kappa) to the observed temperatures were used. Based on the results, we determined that the minimum and maximum daily temperatures are not normally distributed. Considering the minimum and maximum temperatures observed in January and July, respectively, the distributions of 3-P beta-kappa and 3-P generalized logistic produced better statistics. In place of the 2-P standard normal, employing these distributions in the CLIGEN (WEPP model weather generator) to generate daily maximum and minimum temperatures for a randomly selected 10 years of simulation improved efficiency of the model by means of better representing observed values.

(Download PDF)    (Export to EndNotes)