the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Quantifying and attributing the role of anthropogenic climate change in industrial-era retreat of Pine Island Glacier
Abstract. The West Antarctic Ice Sheet (WAIS) has undergone rapid change over the satellite era, characterized by significant thinning, grounding line retreat, and mass loss. Over one-third of the ice loss from this region is from Pine Island Glacier (PIG). However, robust causal links between anthropogenic climate change and PIG ice loss have yet to be established. Here, we quantify the role of anthropogenic climate change in observed retreat of PIG over the 20th century and how this may evolve up to 2200. To do so, we use an ensemble Kalman inversion data assimilation method embedded into the calibrate-emulate-sample (CES) uncertainty quantification framework. This procedure yields observationally constrained probability distributions of both model and climate forcing parameters. Our analysis suggest that it is unlikely that the extent of 20th century PIG retreat would have taken place without anthropogenically driven trends in ice-sheet forcing and that anthropogenic forcing exacerbated the extent of PIG retreat over the 20th century, by approximately 18%. We also find significant retreat even with no anthropogenic trends in forcing, potentially highlighting the role of ice-sheet memory associated with long, slow retreat over the Holocene in controlling present retreat of the WAIS. We further find that significant anthropogenic signals in climate forcing only emerge in the middle of the 22nd century. An important caveat to this work is our choice of initial state, which is larger than expected in practice and may render the anthropogenic forcing contribution in our simulations to be smaller than it is in practice.
- Preprint
(3463 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2315', John Erich Christian, 02 Sep 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2315/egusphere-2025-2315-RC1-supplement.pdf
-
RC2: 'Comment on egusphere-2025-2315', Andy Aschwanden, 13 Sep 2025
Review of "Quantifying and attributing the role of anthropogenic climate change in industrial-era retreat of Pine Island Glacier" by Bradley et al
This study applies a uncertainty quantification framework called "calibrate-emulate-sample" to the multi-centennial evolution of Pine Island Glacier.
The motivation for the "calibrate-emulate-sample" method is its iterative nature, which results in more members that agree with observations, compared to the standard Latin Hypercube Sampling plus Importance Sampling. The experimental design is very thoughtful, the manuscript is very well written and the authors do an excellent job highlighting potential caveats when necessary. The methods are clearly explained, and I was able to understand the gist of the "calibrate-emulate-sample" framework (I am not an expert in this field and thus not qualified to assess the details of the method).Main comments:
- Focusing only on Figure 7a, one is tempted to conclude that all four ensembles explain the two observations of grounding line position equally well. This makes me wonder how robust your attribution to anthropogenic climate change really is. Are there any additional observations that could be used to further constrain the ensembles? (e.g. Sentinel images that provide the front position, observed velocities). How well does your ice sheet model reproduce reality besides retreat, e.g., velocities, dhdt? Â In addition, mapping grounding line position onto a center-line is a relatively weak metric. Would you get a different result by using the floating/grounded mask and a Jaquard Score?
- You construct three emulators, one for each target (GL 1930, 2015, Volume 2015). Would your findings change if you used one emulator that predicts all three targets?
Kind regards,Andy Aschwanden
University of Alaska FairbanksMinor comments:
Melt prefactor vs Sliding prefactor. The sampler seems most opinionate about these two parameters, are they strongly anti-correlated?
L 135: "which premultiplies the . $A_$..." (remove ".")L 219 and 220: There is no Figure 2d, I assume you mean 2c.
L 343 ...by Cleary et al (2021)...
L 393 (and elsewhere) "hasn't" -> "has not"
Figure 3: "...as a function of (c-e) model and (f-h) climate parameters..." I think those are switched, (c-e) are climate and (f-h) are model parameters.
Â
Citation: https://doi.org/10.5194/egusphere-2025-2315-RC2
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
468 | 99 | 13 | 580 | 8 | 12 |
- HTML: 468
- PDF: 99
- XML: 13
- Total: 580
- BibTeX: 8
- EndNote: 12
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1