Premium
Verification of porosity predictors for fluvial sand‐gravel deposits
Author(s) -
Frings Roy M.,
Schüttrumpf Holger,
Vollmer Stefan
Publication year - 2011
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/2010wr009690
Subject(s) - porosity , standard deviation , sediment , grain size , fluvial , geology , soil science , mineralogy , multivariate statistics , geomorphology , geotechnical engineering , statistics , mathematics , structural basin
Although porosity is a key property of sediment mixtures, little is known of the natural variations in porosity in fluvial systems. Porosity predictors can help to generate such information. The objective of this study was to determine the accuracy of porosity predictors for fluvial sand‐gravel mixtures. In order to do so, porosity measurements were done in the Rhine River, using a diving bell to allow undisturbed sampling under water. In addition, laboratory experiments were conducted, and porosity data from literature were reanalyzed. Measured porosity values range from 0.06 to 0.48. Our study shows that predictors based on the median grain size, the deviation from the Fuller curve, or the sediment standard deviation are unable to reproduce the observed variation in porosity. Predictors based on the entire grain size distribution perform better but are biased and have a large prediction error. This suggests that they do not account correctly for the grain size effect on porosity. On the other hand, parameters such as grain shape and the mechanism of deposition probably also have a distinct influence on porosity. Slightly more accurate predictions can be obtained with tailor‐made predictors for the river under interest. Multivariate regression analysis on the Rhine data set produced a predictor with two independent parameters: the sediment standard deviation and the number of grains smaller than 0.5 mm, representing grain mixing effects and adhesion effects, respectively. Despite the shortcomings, porosity predictors provide valuable insights in the spatial variation in porosity, which is demonstrated in a case study for the Rhine River.