Neural Networks Package for fitting Pedotransfer Functions
The
development of models simulating soil processes has increased rapidly in recent
years. These models have been developed to improve the understanding of important
soil processes and also to act as tools for evaluating agricultural and environmental
problems. However, models usually require a large number of parameters to
describe the transport coefficient, content of substances in the soil, or
other physical and chemical properties. Thus, collecting soil property data
has become an urgent need to feed the very hungry, almost insatiable, environmental
(simulation) management models.
The term pedotransfer function (PTF) was coined by Bouma (1989) as translating
data we have into what we need. Pedotransfer functions allow basic information
from soil surveys to be translated into other more laborious and expensively
determined soil properties.
Neural networks attempt to build a mathematical model that supposedly works in an analogous way to human brain. The advantage of a neural net is that it can predict one or more outputs from a flexible network of functions of input variables. We do not need to know the functional form of the response surface. To describe a system, there is no assumed structure of the model, instead the networks are adjusted or 'trained' so that a particular input leads to a specific target output.
Neuropack is a software that can be used to generate or fit pedotransfer functions using neural networks. The trained networks subsequently can be used to validate and predict new soil samples.
Neuropack has the following unique features:
