We subject three publically available models of layer 5 pyramidal neurons to three simple computational challenges. Practically, can we realistically model single neurons? Philosophically, should we realistically model single neurons? We use layer 5 neocortical pyramidal neurons as a test case to examine these issues. We then attempt to address two fundamental questions. We briefly review the state of the art of compartmental modeling highlighting recent progress and intrinsic flaws. Thus researchers have asked what are the computational abilities of single neurons and attempted to give answers using realistic models. This rapidly advancing field is driven by the discovery that some neurons don't merely sum their inputs and fire if the sum exceeds some threshold. Here we discuss the oxymoronic, realistic modeling of single neurons. Contrary to this basic premise, with the 21st century has come a surge in computational efforts to model biological processes in great detail. A model should not be a facsimile of reality it is an aid for understanding it. A successful model shows how a complex phenomenon arises from relatively simple principles while preserving major physical or biological rules and predicting novel experiments. Scientific models are abstractions that aim to explain natural phenomena. Is realistic neuronal modeling realistic? Strain should be compatible among the cross-sections: If at all, it should vary smoothly and systematically along a given fault zone. Also, from the deformed and restored cross-sections we can measure the strain incurred during deformation. Structures should change continuously from one section to another. Additional constraints are provided by comparison of adjacent cross-sections. If this is possible without producing gaps or overlaps, the interpretation is considered valid (but not unique) for a single cross-section. The structure sections are checked for consistency by restoring them to an undeformed state. We use balancing of serial, parallel cross-sections to constrain subsurface extrapolations. The fault zone geometries are never fully constrained by data and must be extrapolated to depth. We present work-in-progress from the Thuringian Basin in central Germany. Obtaining a more accurate representation of faults and fault zones is therefore challenging.
Also, faults tend to split up into several branches, forming fault zones. Boreholes located close to a fault can therefore cross it at depth, resulting in stratigraphic control points allocated to the wrong block.
Most natural faults are inclined and may change dips according to rock type or flatten into mechanically weak layers.
Besides being geologically and mechanically unreasonable, this also causes technical difficulties in the modelling workflow. as vertical downward projections of fault traces observed at the surface. Still so, many existing models treat faults in a simplistic fashion, e.g. This is one reason why a " realistic" representation of faults in 3D models is desirable. Faults affect the continuity of aquifers and can themselves act as fluid conduits or barriers. Such models typically comprise the bounding surfaces of stratigraphic layers and faults. Zehner, B.ģD computer models of geological architecture are evolving into a standard tool for visualization and analysis. Towards " realistic" fault zones in a 3D structure model of the Thuringian Basin, Germany
0 kommentar(er)
