The Modeling Lab
Learning objectives
- Assemble the whole forward problem in one workbench
- Build a model, choose a view, set the wavelet
- Move between the impedance model, section, and trace
- Export the exact array you see as a real file
Everything, in One Place
The final part gathers every tool of the course into a single workbench. You BUILD a model from a preset, a layered sequence, a wedge, a channel, or a fault block, and shape it with heterogeneity. You choose a VIEW: the acoustic impedance model itself, the seismic section the convolutional engine makes from it, or a single synthetic trace. You set the WAVELET frequency. And you EXPORT the exact array you are looking at, as a real NPY, CSV, or SEG-Y-lite file with its byte count shown, ready to load in a research code.
The Forward Problem in Your Hands
This is the whole of the course made operational: an earth model in, a synthetic out, and a file you can carry away. Every knob is one you now understand, the impedance contrasts, the wavelet bandwidth, the heterogeneity texture, the file format, so nothing here is a black box.
By design the Lab runs only the cheap, exact engines: the model builders, the convolutional section, and the exporters. That is a deliberate fit-for-purpose choice, it keeps the workbench instant. The heavier engines, the full wave equation and the elastic gathers, are the same physics at higher cost, too heavy to run interactively in a browser but perfectly suited to a dedicated runtime. The next section hands the very model you built here to a Python environment to run them for real, and points you onward to the industry-scale codes.