Landslides occur around the world, in various conditions of weather and terrain, costing billions of dollars, causing thousands of deaths and worsening environmental,
economic and political problems. A serious question associated with landslides is the disordered of human occupation - characterizing a geopolitical and economic issue. Actions to mitigate the problems arising from this phenomenon include environmental monitoring of the areas of greatest susceptibility for managing the higher risk population. The monitoring processes consist of recognizing soil dynamics and
climatic factors, responsible for causing destabilization and landslides. In the Brazilian context, there is a strong relationship between heavy rains and landslides, since the phenomenon of runoff water infiltrated the soil and produces erosion. Therefore, a predictive monitoring of the first soil layers is an important feature for detecting areas of imminent risk of slippage. Several approaches for monitoring were
presented internationally, they are based on different technologies, such as acoustic waves in the soil; detectors of relative humidity and rainfall selective analysis. All these approaches are anchored in the paradigm of remote sensing and smartsensor networks. Based on that, this paper proposes a model to estimate the imminence of of landslides risk. The proposed model is built on four pillars: (i) proposing a physics
simulation environment, from the creation of a laboratory scale test environments, (ii) recognizing the technology and refinement of pre-existing models, (iv) developing of
multiparametric smart sensor and (v) developing of a peer-to-peer network of sensors. This study aggregates to the scientific and technical contribution a framework for monitoring, useful for studying of landslides dynamics. Such
framework is crucial for prototyping monitoring networks to be used by civil defense.
