Deep neural networks: a flexible framework for soil modelling

<p>Since the first applications of machine learning (ML) methods in the 80s, ML adoption in soil science has increased considerably. In parallel, the size of the soil datasets has also increased. However, current soil modelling is mostly based on “traditional” ML approaches, not taking full advantage of large datasets or the multiple opportunities provided by more advanced modelling methods. Here I present the latest examples in the use of ML for soil predictive modelling, specifically the use of deep learning models in the context of soil spatial modelling and soil spectroscopy. Additionally, I will show less traditional ML applications that allow the use of field data into numerical workflows, and some advanced training techniques that showcase the flexibility of neural networks and open new, exciting opportunities to solve soil modelling challenges.</p>

Study of Input Parameters of Layered Half-Space used for Soil Modelling

This article deals with a problem of soil modeling. That is a necessary part of foundation modeling.The soil modeling can directly influent the resulting deformation of the foundations and therefore deformationof the whole building. The accurate procedure that can simulate the foundation-soil interaction is still unknown.A greater understanding of the issue through experiments and numerical modeling is important for a goodapplication and proper use in practice. To improve this understanding an experimental test of a concrete slabwas performed on special testing equipment. The experiment is complemented by numerical modeling.Numerical models are created in the Ansys software. The article features three cube models and a half-spheremodel. Deformations on all featured models are discussed and compared to the experimental values andsimplified hand calculation. The subsoil model is based on the Boussinesq half-space theory. The soil section isassumed homogeneous and layered. The size of models is assumed as a changing parameter from 10 m to 30 m.A recommended model size was evaluated from the parametric study as a match of deformation from modelsand the experiment. This computed model size is compared with an affected depth described by the standardČSN 731001. Computation was performed using supercomputer Anselm in the National SupercomputingCenter IT4Innovations in the VSB-Technical University of Ostrava.

Do we need to new set of soil principles to guide the cross-sector engagement of soil education

<p>The importance of and role that soil plays in ensuring the future sustainability of human and planetary health is well established and the emerging soil security concept clearly identifies that education is a crucial component to ensure the securing soil to be fit for this purpose. Traditionally education in soil science has been limited to developing expertise in the discipline. To be truly effective we need to go beyond the boundaries of the discipline, and even its natural home of academia, and begin to explore the types of learning that can be developed to engage the whole community and raise our collective connectivity with soil.</p><p>Previously theoretical frameworks based around the dimensions of to ‘know’, ‘know of’ and ‘be aware’ of soil have been accepted by the education community as well as experiential learning practices framed by the teaching-research-Industry-learning (TRIL) models. There is now the emerging question of the need for a set of newly proposed set of principles, in the same way as a set of elementary assumptions have been developed for disciplines in biology and geology, which will impact the design of learning and its engagement within the disciplinary and broader community. Starting with the Pedon this elementary level will ensure awareness of soil. Coupled with outward focused responsibility of providing salient knowledge together with the social intelligence will use the second principle of Processes to provide resolution to soil related problems. Traditionally, this knowledge is often used to tackle well know threats, but more recently the advances in digital soil mapping and decreasing soil modelling have enabled greater interdisciplinary opportunities to solve soil knowledge based around the principles of variation and ultimately forecasting soil change.</p><p>This paper will align the set of principles against the current soil science education practices and how these can be used to engage with the broader community outside of academia.  </p>

Rain Use Efficiency in Southern Romania after Using the Device for Soil Modelling in Interrupted Furrows for Weeding Crops -DMBC-5

The global impact of climate change, with the potential to affect agriculture through changes in temperature, rainwater distribution and amount, leads to the need to develop integrated technologies that increase rain use efficiency and support soil and environmental quality, ensuring higher agricultural outputs with lower costs. Soil modeling in interrupted furrows is a very efficient rainwater harvesting system, but has not yet been scientifically assessed for the conditions in Southern Romania as the other soil working systems have been. For this purpose, a soil modeling device consisting of a frame with triangle for coupling to the tractor, wheels for adjusting and limiting the working depth, soil loosening knives, ridge ploughs and furrow compartmenting equipment was compared with the conventional technology for sunflower culture in two localities in Southern Romania: Crânguri in Giurgiu County and Mărculeşti in Calarasi County. The device for soil modeling in interrupted furrows for weeding crops provides a viable option that has positive effects on soil properties and ensures increased crop yields compared to conventional works as well as the opportunity to increase the efficiency of water recovery in agriculture by harvesting rainwater.