A fast surrogate model for 3D Earth glacial isostatic adjustment using Tensorflow (v2.8.0) artificial neural networks
Bibliographic record
Abstract
Abstract. Models of glacial isostatic adjustment (GIA) play a central role in the interpretation of various geologic and geodetic data to understand and simulate past and future changes in ice sheets and sea level, as well as to infer rheological properties of the deep Earth. During the past few decades, a major advance has been the development of models that include 3D Earth structure, as opposed to 1D spherically symmetric (SS) structure. However, a major limitation in employing 3D GIA models is their high computational expense. As such, we have developed a method using artificial neural networks (ANNs) and the Tensorflow library to predict the influence of 3D Earth models with the goal of more affordably exploring the parameter space of these models, specifically the radial (1D) viscosity profile to which the lateral variations are added. Our goal is to test whether the use of an ANN to produce a fast surrogate model can accurately predict the difference in GIA model outputs (i.e., relative sea level (RSL) and uplift rates) for the 3D case relative to the SS case. If so, the surrogate model can be used with a computationally efficient SS (Earth) GIA model to generate output that replicates that from a 3D (Earth) GIA model. Evaluation of the surrogate model performance for deglacial RSL indicates that it is able to provide useful estimates of this field throughout the parameter space when trained on only ≈15 % (≈50) of the parameter vectors considered (330 in total). We applied the surrogate model in a model–data comparison exercise using RSL data distributed along the North American coasts from the Canadian Arctic to the US Gulf Coast. We found that the surrogate model is able to successfully reproduce the model–data misfit values such that the region of minimum misfit either generally overlaps the 3D GIA model results or is within two increments of the radial viscosity model parameter space (defined here as lithosphere thickness, upper-mantle viscosity, and lower-mantle viscosity). The surrogate model can, therefore, be used to accurately explore this aspect of the 3D Earth model parameter space. In summary, this work demonstrates the utility of machine learning in 3D Earth GIA modelling, and so future work to expand on this initial proof-of-concept analysis is warranted.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".