Soil properties are "dynamic" and evolve in space and time. This spatial and temporal variability depends upon factors such as the geological and geomorphological characteristics, climate, vegetation cover and the degree of human impact.
Often, for a better understanding of soil characteristics spatio-temporal evolution models (geostatistical and others) are used to predict soil properties. Models are a helpful tool to understand the soil spatio-temporal evolution and give a comprehensive view. However, depending on the model used the spatio-temporal predictions could be significantly different, using the same database. Therefore it is important to identify the best predictor and minimize the errors in order to select the best fitted model. Since models are widely used to estimate a large range of soil properties (e.g heavy metals, erosion) they have important impacts in soil management (e.g. remediation, fertilization, erosion mitigation) and related decision making. Therefore it is of fundamental importance to find accurate and validated models. This session is focused on the problematic of model validation and improvement in order to have better estimations. Communications about soil properties modelling and validation from all scales are very welcome.

The papers submitted to this session will be considered for a Focus in J Plant Nutrition and Soil Science (http://onlinelibrary.wiley.com/journal/10.1002/%28ISSN%291522-2624/)