University of Tabriz Water and Soil Science 2008-5133 23 2 2013 07 23 Estimation of Soil Penetration Resistance Using Fractal Parameters of Particle and Aggregate Size Distributions Estimation of Soil Penetration Resistance Using Fractal Parameters of Particle and Aggregate Size Distributions 13 27 95 FA Mohammad Jorreh Hossein Bayat Ali Akbar Safari Sanjani Naser Davatghar Journal Article 2012 01 23 Penetration resistance (PR) is a highly variabe soil property and there is no accurate and representative PR data in most databases. On the other hand, direct measurement of the PR is costly and expensive. The objective of this study was prediction of PR from easily available soil properties using pedotransfer functions (PTFs). Fifty three soil samples were taken from dominant soil series of Hamadan province. Particle, macro and micro aggregate size distributions, water content and bulk density were measured in the laboratory. PR and soil structural qualitative parameters were measured in the field. In the first regression model, bulk density and water content as conventional predictors of PR were introduced to the model. Then, soil structural qualitative parameters (aggregates shape, type and size), primary particle, micro and macro aggregate size distribution fractal parameters were introduced to the models to predict PR. Finally the improvement of PR  prediction using various developed PTFs were evaluated by the statistical indices as R², R²_adj, root mean square error, Akaike  information criterion and relative improvement. The results showed that using fractal parameters as predictors along with the conventional predictors significantly improved the accuracy and reliability of the models. Using soil structural qualitative parameters along with fractal parameters of particle size distribution may significantly improve the PR predictions. Penetration resistance (PR) is a highly variabe soil property and there is no accurate and representative PR data in most databases. On the other hand, direct measurement of the PR is costly and expensive. The objective of this study was prediction of PR from easily available soil properties using pedotransfer functions (PTFs). Fifty three soil samples were taken from dominant soil series of Hamadan province. Particle, macro and micro aggregate size distributions, water content and bulk density were measured in the laboratory. PR and soil structural qualitative parameters were measured in the field. In the first regression model, bulk density and water content as conventional predictors of PR were introduced to the model. Then, soil structural qualitative parameters (aggregates shape, type and size), primary particle, micro and macro aggregate size distribution fractal parameters were introduced to the models to predict PR. Finally the improvement of PR  prediction using various developed PTFs were evaluated by the statistical indices as R², R²_adj, root mean square error, Akaike  information criterion and relative improvement. The results showed that using fractal parameters as predictors along with the conventional predictors significantly improved the accuracy and reliability of the models. Using soil structural qualitative parameters along with fractal parameters of particle size distribution may significantly improve the PR predictions. https://water-soil.tabrizu.ac.ir/article_95_9b862f6af0bd6c6f6b682f5669601806.pdf