|
Descripción
|
|
|---|---|
| Soil particle size distribution (PSD) can be viewed as a mathematical measure supported in the interval of grain sizes. PSD not only describes soil, but it also is a useful predictor for many other soil physical properties, as bulk density or Ksat. In the case of soil, available information about this distribution is usually very scarce, and given in form of triplets, i.e., the mass of three different size intervals, e.g. the USDA clay, silt and sand percentages. The size of these intervals has changed over time for different reasons; nowadays triplet limits seem to be arbitrary. Without any more assumptions, this limited knowledge would not suffice to recover the whole distribution. However, the Information Entropy (IE) approach models the PSD as a selfsimilar measure. In this setting, the PSD is the solution of a differential equation that uses the coarse triplet measure- ments as initial condition. In this sense, the information contained in the input triplet, i.e., the available information of the PSD, entirely determines the final distribution. This poses a problem: which size intervals should we measure? Indeed, different input triplets -different initial conditions- will lead to different dis- tributions. So one can speak about how much ?information? a triplet stores based on its potential to correctly reconstruct the actual PSD. Results indicate that the information contained in certain soil triplets is sufficient to rebuild the PSD: for each soil sample tested there is always at least a triplet that contains enough information to simulate the whole distribution. Furthermore, this conceptual approach provides us with a useful param- eter, which is also triplet-dependant: the ?coarse-entropy?. In the spirit of the pedofransfer functions, that use the PSD to estimate difficult to measure soil physical properties, we investigate which triplets store greater capabil- ity of predicting some of these properties -Ksat and packing- by means of the IE parametrization. Overall, the textural triplet selection appears to be application specific. | |
|
Internacional
|
Si |
|
ISSN o ISBN
|
978-84-948550-7-8 |
|
Entidad relacionada
|
|
|
Nacionalidad Entidad
|
Sin nacionalidad |
|
Lugar del congreso
|
Ávila, España |