the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Predictability of Arctic Sea Ice Drift in Coupled Climate Models
- 1Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany
- anow at: MARUM – Center for Marine Environmental Science & Institute of Environmental Physics, University of Bremen, Bremen, Germany
- 1Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany
- anow at: MARUM – Center for Marine Environmental Science & Institute of Environmental Physics, University of Bremen, Bremen, Germany
Abstract. Skillful sea ice drift forecasts are crucial for scientific mission planning and marine safety. Wind is the dominant driver of ice motion variability, but more slowly varying components of the climate system, in particular ice thickness and ocean currents, bear the potential to render ice drift more predictable than the wind. In this study, we provide the first assessment of Arctic sea ice drift predictability in four coupled general circulation models (GCMs), using a suite of "perfect-model" ensemble simulations. We find the position vector from Lagrangian trajectories of virtual buoys to remain predictable for at least 90 (45) days lead time for initializations in January (July), reaching about 80 % of the position uncertainty of a climatological reference forecast. In contrast, the uncertainty of Eulerian drift vector predictions reaches the level of the climatological uncertainty within less than four weeks. Spatial patterns of uncertainty, varying with season and across models, develop in all investigated GCMs. For two models providing near-surface wind data (AWI-CM1 and HadGEM1.2), we find spatial patterns and large fractions of the variance to be explained by wind vector uncertainty. The latter implies that sea ice drift is only marginally more predictable than wind. Nevertheless, particularly one of the four models (GFDL-CM3) shows a significant correlation of up to −0.85 between initial ice thickness and target position uncertainty in large parts of the Arctic. Our results provide a first assessment of the inherent predictability of ice motion in coupled climate models, they can be used to put current real-world forecast skill into perspective, and highlight model diversity of sea ice drift predictability.
Simon F. Reifenberg and Helge F. Goessling
Status: closed
-
RC1: 'Comment on tc-2022-41', Anonymous Referee #1, 08 Apr 2022
The manuscript “Predictability of Arctic Sea Ice Drift in Coupled Climate Models” discusses potential predictability of sea ice motion in four climate models from Eulerian and Lagrangian perspectives. The authors identify the potential predictability horizon and identify the wind variability as the main source of uncertainty. The role of initial ice thickness was found to be small. The manuscript is very well written, with good presentation of methodology and results, with deeply though discussions. It provides an important contribution to understanding of sea ice predictability in general. My minor comments only concern few overly complicated explanations in the text that should be simplified for less experience readers.
Line 30:
Although ‘errors’ and ‘uncertainty’ are well established terms, it is better to provide here a concise and clear definition of these terms (as well as ‘accuracy’ and ‘skill’) as understood by the authors for avoiding ambiguity in the rest of the manuscript.
Line 69:
What is “climatological uncertainty”? The following explanation “the uncertainty of an ensemble forecast constructed from independent years simulated by the same model with constant mean climate and variance” seems very short and hard to understand.
Does it mean that a model is initialized at some point of time, then it is run for several years (and external forcing is the same every year), then an ensemble is constructed from individual years, then the uncertainty of a predictand in this ensemble is computed and used as a reference? Is there a reference to justify building the “climatological uncertainty” this way? For how many years should the model run? What if the model doesn’t stabilize around a constant climate and the “climate uncertainty” continues to grow with the number of years?
If my understanding is wrong, a better explanation, possibly with a scheme, is worth adding here. On such a scheme the error, uncertainty, accuracy and predictability can be visually shown for easier understanding by readers not well familiar with the topic.
Line 208 – 211: It is difficult to understand how the measure of uncertainty is computed.
“variance ellipse”, “semi-major axis” In which space? Dimensionality of this space?
Can an equation be added here?
Line 229 – 231: This sentence is also difficult to digest. How a plane can be tangential to a point (barycenter)? Please add an equation.
Line 352: What does it mean “normalized uncertainty reaches the climatological uncertainty”? Wasn’t the normalization done to the climatological uncertainty? (eq. 4)? Shouldn’t it read “uncertainty reaches the climatological uncertainty, i.e. normalized uncertainty reaches 1”?
Line 410: A reference to Fig. 10 should be added.
- AC1: 'Reply on RC1', Simon F. Reifenberg, 03 Jun 2022
-
RC2: 'Comment on tc-2022-41', Anonymous Referee #2, 06 May 2022
Summary:
This paper explores the limits of predictability of sea ice drift in four “perfect-model” simulations, and finds that the uncertainty in the winds is the primary limit to predictability. The thickness of the sea ice in one of the four models shows a negative correlation with position uncertainty. This is an interesting paper that should be accepted after mostly minor suggestions.
Minor Comments:
1) The ice speeds discussed in section 3, and shown in Figure 2 of 10 cm/s in the models seem really fast compared to observations which seem to be less than 5 cm/s. For example, https://nsidc.org/cryosphere/seaice/processes/circulation.html, shows that the typical ice speed less than 5 cm/s. And looking at some other recent papers such as Kwok, et al. 2013 (https://doi.org/10.1002/jgrc.20191) some similar numbers. Please discuss possible implications that the faster model speeds may have on the conclusions of this paper.
2) Line 1: I think it is worth restating Nansen’s rule of thumb explicitly here.
3) The paper tends to be too wordy. This and other comments below are aimed at tightening up the text. For example, on line 32, the authors write “Here, we therefore differentiate…”. I would tighten this up to simply state “We differentiate…”. I would comb through the paper and reduce the use of these transition words.
4) Lines 72 -93 seemed out of place. I would maybe move it up above line 58? I don’t feel strongly about this.
5) Line 73, site a few “recent studies”.
6) Line 81 stating “(two-dimensional)” is not necessary since this should be implied by the discussion of area.
7) Line 154: I suggest stating the “1st July” as “1 July” or “July 1st” or “the 1st of July”., then restate “1st January” in the same way.
8) Line 202: delete “which are both two-dimensional quantities”.
9) Lines 205- 210: Too wordy. I think the authors can delete most of lines 206-207, and just go with lines 208-209.
10) Lines 237-238. Too wordy. I would delete the first sentence starting at line 237, and simply say “Maps of average ice thickness for the months of March and September are presented in Day et al. (2016).”
11) Line 246: delete “previously introduced”.
12) Line 274: delete “In the following”, and start the sentence as “We now consider the differences in the trajectories…”.
13) Combine Figures 4 and 5.
14) Line 293, delete “In the following, “
15) Line 294 and 295: change “a (normalized) uncertainty” to “an uncertainty”.
16) Figure 6: Capitalize “Uncertainty” under colorbar.
17) Line 306: delete “also”.
18) Line 309: change “with January and July initializations” to “for January and July”.
19) Line 334: delete “an additional point of view –“.
20) Line 341: delete this sentence.
21) Line 344: change “…position here. This enables…” to “…which enables…”.
22) Line 346: change “of normalized” to “for”
23) Line 436: delete “also”.
24) Line 455: change “affect” to “cause”.
25) Line 485-486: delete sentence starting with “Our study…”.
26) Line 487: change “within few days” to “within a few days”.
- AC2: 'Reply on RC2', Simon F. Reifenberg, 03 Jun 2022
Status: closed
-
RC1: 'Comment on tc-2022-41', Anonymous Referee #1, 08 Apr 2022
The manuscript “Predictability of Arctic Sea Ice Drift in Coupled Climate Models” discusses potential predictability of sea ice motion in four climate models from Eulerian and Lagrangian perspectives. The authors identify the potential predictability horizon and identify the wind variability as the main source of uncertainty. The role of initial ice thickness was found to be small. The manuscript is very well written, with good presentation of methodology and results, with deeply though discussions. It provides an important contribution to understanding of sea ice predictability in general. My minor comments only concern few overly complicated explanations in the text that should be simplified for less experience readers.
Line 30:
Although ‘errors’ and ‘uncertainty’ are well established terms, it is better to provide here a concise and clear definition of these terms (as well as ‘accuracy’ and ‘skill’) as understood by the authors for avoiding ambiguity in the rest of the manuscript.
Line 69:
What is “climatological uncertainty”? The following explanation “the uncertainty of an ensemble forecast constructed from independent years simulated by the same model with constant mean climate and variance” seems very short and hard to understand.
Does it mean that a model is initialized at some point of time, then it is run for several years (and external forcing is the same every year), then an ensemble is constructed from individual years, then the uncertainty of a predictand in this ensemble is computed and used as a reference? Is there a reference to justify building the “climatological uncertainty” this way? For how many years should the model run? What if the model doesn’t stabilize around a constant climate and the “climate uncertainty” continues to grow with the number of years?
If my understanding is wrong, a better explanation, possibly with a scheme, is worth adding here. On such a scheme the error, uncertainty, accuracy and predictability can be visually shown for easier understanding by readers not well familiar with the topic.
Line 208 – 211: It is difficult to understand how the measure of uncertainty is computed.
“variance ellipse”, “semi-major axis” In which space? Dimensionality of this space?
Can an equation be added here?
Line 229 – 231: This sentence is also difficult to digest. How a plane can be tangential to a point (barycenter)? Please add an equation.
Line 352: What does it mean “normalized uncertainty reaches the climatological uncertainty”? Wasn’t the normalization done to the climatological uncertainty? (eq. 4)? Shouldn’t it read “uncertainty reaches the climatological uncertainty, i.e. normalized uncertainty reaches 1”?
Line 410: A reference to Fig. 10 should be added.
- AC1: 'Reply on RC1', Simon F. Reifenberg, 03 Jun 2022
-
RC2: 'Comment on tc-2022-41', Anonymous Referee #2, 06 May 2022
Summary:
This paper explores the limits of predictability of sea ice drift in four “perfect-model” simulations, and finds that the uncertainty in the winds is the primary limit to predictability. The thickness of the sea ice in one of the four models shows a negative correlation with position uncertainty. This is an interesting paper that should be accepted after mostly minor suggestions.
Minor Comments:
1) The ice speeds discussed in section 3, and shown in Figure 2 of 10 cm/s in the models seem really fast compared to observations which seem to be less than 5 cm/s. For example, https://nsidc.org/cryosphere/seaice/processes/circulation.html, shows that the typical ice speed less than 5 cm/s. And looking at some other recent papers such as Kwok, et al. 2013 (https://doi.org/10.1002/jgrc.20191) some similar numbers. Please discuss possible implications that the faster model speeds may have on the conclusions of this paper.
2) Line 1: I think it is worth restating Nansen’s rule of thumb explicitly here.
3) The paper tends to be too wordy. This and other comments below are aimed at tightening up the text. For example, on line 32, the authors write “Here, we therefore differentiate…”. I would tighten this up to simply state “We differentiate…”. I would comb through the paper and reduce the use of these transition words.
4) Lines 72 -93 seemed out of place. I would maybe move it up above line 58? I don’t feel strongly about this.
5) Line 73, site a few “recent studies”.
6) Line 81 stating “(two-dimensional)” is not necessary since this should be implied by the discussion of area.
7) Line 154: I suggest stating the “1st July” as “1 July” or “July 1st” or “the 1st of July”., then restate “1st January” in the same way.
8) Line 202: delete “which are both two-dimensional quantities”.
9) Lines 205- 210: Too wordy. I think the authors can delete most of lines 206-207, and just go with lines 208-209.
10) Lines 237-238. Too wordy. I would delete the first sentence starting at line 237, and simply say “Maps of average ice thickness for the months of March and September are presented in Day et al. (2016).”
11) Line 246: delete “previously introduced”.
12) Line 274: delete “In the following”, and start the sentence as “We now consider the differences in the trajectories…”.
13) Combine Figures 4 and 5.
14) Line 293, delete “In the following, “
15) Line 294 and 295: change “a (normalized) uncertainty” to “an uncertainty”.
16) Figure 6: Capitalize “Uncertainty” under colorbar.
17) Line 306: delete “also”.
18) Line 309: change “with January and July initializations” to “for January and July”.
19) Line 334: delete “an additional point of view –“.
20) Line 341: delete this sentence.
21) Line 344: change “…position here. This enables…” to “…which enables…”.
22) Line 346: change “of normalized” to “for”
23) Line 436: delete “also”.
24) Line 455: change “affect” to “cause”.
25) Line 485-486: delete sentence starting with “Our study…”.
26) Line 487: change “within few days” to “within a few days”.
- AC2: 'Reply on RC2', Simon F. Reifenberg, 03 Jun 2022
Simon F. Reifenberg and Helge F. Goessling
Simon F. Reifenberg and Helge F. Goessling
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 370 | 117 | 17 | 504 | 9 | 7 |
- HTML: 370
- PDF: 117
- XML: 17
- Total: 504
- BibTeX: 9
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1