Talk:2009 Dusky Sound earthquake/GA1

GA Review
The edit link for this section can be used to add comments to the review.''

Nominator: 04:51, 5 June 2024 (UTC)

Reviewer: Marshelec (talk · contribs) 00:27, 28 June 2024 (UTC)

I will undertake a review of this article over the next 7 days or so. Looking forward to learning more about this topic.Marshelec (talk) 00:27, 28 June 2024 (UTC)

Well-written
The article generally achieves a high standard. Feedback about the content is split between items that are "required" - ie must be addressed to meet the GA standard, and items that are "discretionary". These latter items are suggestions in good faith to make the article as good as it can be, in the time available for the review. They are not essential to meeting the GA standard, although I hope they will be considered.

Infobox

 * Required


 * The Tsunami run-up figure of 2.3m is supported by the abstract of this source: . However, it does not match the text in the Tsunami section where 230cm is quoted. This discrepancy needs to be resolved.

Lead

 * Required


 * The first sentence needs to be reworked to comply with MOS:LEAD, in particular MOS:LEADSENTENCE. Here is one possible reworking:
 * The 2009 Dusky Sound earthquake was a 7.8 earthquake that struck a remote region of Fiordland in the South Island of New Zealand, on 15 July 2009 at 21:22 NZST (09:22 UTC).


 * The source no 2 says that the epicentre was 160km north-west of Invercargill, not 150km. If sources differ this possibly may be OK, but I would expect the GeoNet source to be reliable. This needs to be resolved.


 * Discretionary


 * There are three successive sentences in the lead that commence with It or Its. Minor reworking would avoid this.
 * It was the country's largest earthquake in terms of magnitude since the 1931 Hawke's Bay earthquake, and only caused minor damage. The word "and" should be replaced with "but". The contrast between the large magnitude of the earthquake, and the low level of damage is one of the most interesting aspects of this event. Using the word "but" will give the reader an immediate indication that there is a contrast that is an important part of this account.
 * "epicentre" should be wikilinked
 * "tsunami" should be wikilinked

Earthquake
 * Discretionary


 * The 2009 earthquake followed considerable seismic activity in northern Fiordland since 1988—six earthquakes above magnitude 6.0 struck the region before 2009,[10] including a magnitude 7.2 earthquake in 2003.[11] This sentence could be relocated to be the last sentence in the first paragraph of this section. This would create a better flow - although it would also be necessary to insert "2009 Dusky Sound earthquake" in the first sentence of the second paragraph.
 * "subduction" should be wikilinked
 * At this location, the Indo-Australian Plate subducts obliquely beneath the latter tectonic plate .. Suggest replacing "latter" with "Pacific Plate" for ease of reading.
 * The 2009 earthquake followed considerable seismic activity in .. Suggest replacing "followed" with "occurred after a sequence of seismic events in .... "
 * The rupture initiated at a depth of 30 km (19 mi) and propagated southwards and towards the surface before it ceased just 15 km (9.3 mi) beneath This sentence is awkward in the second half. Suggest a reworking to: "The rupture initiated at a depth of 30 km (19 mi) and propagated southwards and upwards before it ceased just 15 km (9.3 mi) beneath the surface."
 * The maximum displacement was estimated at 8.9 m (29 ft) about 23 km (14 mi) beneath the surface which occurred 16 seconds following the earthquake initiation. The word "which" in this sentence makes it awkward. Suggest it could be reworked to: "The maximum displacement was estimated at 8.9 m (29 ft) about 23 km (14 mi) beneath the surface, occurring around 16 seconds after the earthquake began".
 * As the rupture mainly propagated away from the mainland, the measured shaking intensity was weaker despite the earthquake's magnitude and shallow depth. Suggest this could be reworked to: "As the rupture mainly propagated away from the mainland, the measured shaking intensity was less than would normally be expected given the earthquake's magnitude and shallow depth"
 * Limited strong ground motion was recorded due to a lack of seismic stations near the earthquake's epicentre which was in a remote region Suggest this could be reworked to: "The earthquake epicentre was in a remote region where few seismic stations were installed, and as a result, few strong motion recordings are available".
 * The ground motion characteristic resulted in fewer landslides Suggest reworking to "..resulted in relatively few landslides"

Effects
 * Discretionary


 * At least 241 landslides were recorded, covering an area of about 5,600 km2.. Suggest reworking to .. over an area of about 5,600 km2. (To avoid any possible interpretation that the actual surface area of landslides was 5,600 km2).
 * From 21:30, KiwiRail services were suspended south of Oamaru and in Buller Gorge to check for damage which resulted in some delays. They were operational again by midnight because no damage was found. Suggest reworking these two sentences to: "From 21:30, KiwiRail suspended rail services south of Oamaru and in the Buller Gorge to check for damage. Services were resumed by midnight because no damage was found."
 * The low damage was attributed to the earthquake's remote location, slow moment release rate, low frequency shaking,[8][15] and New Zealand's building standards Suggest reworking this to: "The low level of damage, despite the large magnitude of the earthquake, was attributed to the earthquake's remote location, slow moment release rate, low frequency shaking,[8][15] and New Zealand's building standards.

Tsunami
 * Required


 * As noted above, a discrepancy in the tsunami run-up figure of 2.3m in the Infobox and the 230cm quoted in this section must be resolved.


 * Discretionary


 * ... the tsunami may have been higher in some areas that were not surveyed due to the remoteness of the affected area Suggest replacing the last "area" with "region" to avoid two uses of "area" in one sentence.

Verifiable with no original research

 * Content is well supported with inline citations of reliable sources
 * No original research is identified
 * Earwig copyvio tool does not indicate anything of concern

Broad in its coverage

 * Provides good coverage of the subject without straying off topic or providing excessive detail

Neutral

 * No issues found

Stable

 * No issues

Illustrated

 * Images are relevant to the subject, and have suitable captions

Initial round of feedback
Initial round of review feedback completed.Marshelec (talk) 07:41, 28 June 2024 (UTC)


 * I've incorporated every suggested change. I've also that the 1931 earthquake is the deadliest in New Zealand history to help explain why the the damage in 2009 was so surprising, but I'm unsure if this counts as some form of editorialising. ―Panamitsu (talk) 08:41, 28 June 2024 (UTC)

Second round of feedback

 * Required


 * On further reading, I see that I missed an important point in the first round of review comments - my apologies. In the lead, and in the Infobox, the depth is stated as 12 km, and this aligns with the first source. However, in the Earthquake section, it states that the earthquake began at 30km beneath the surface and propagated up, finishing 15km beneath the surface. I am no expert, but these do not seem to line up. I do not know the convention for defining the depth. It may be somewhat complex, given the description of how the earthquake propagated. This needs some investigation - to determine how best to state the depth in the lead and infobox, and to ensure alignment between the lead and the body.
 * The second sentence in the lead is currently: Its epicentre was located near Dusky Sound in Fiordland National Park, at a depth of 12 km (7.5 mi), which is 160 km (99 mi) north-west of Invercargill. Once the depth issue is sorted, this sentence needs reworking to put the clauses in a better order. I suggest: " Its epicentre was at a depth of 12 km (7.5 mi) and located near Dusky Sound in Fiordland National Park, around 160 km (99 mi) north-west of Invercargill."
 * Discretionary


 * Despite being the country's largest earthquake in terms of magnitude since 1931 Hawke's Bay earthquake, which remains as the deadliest in New Zealand history, the 2009 earthquake had no casualties and caused only minor damage.. This sentence is now a bit long and convoluted, particularly for the lead. I suggest leaving out the description of the Hawke's Bay earthquake here, and simplifying the sentence to: "It was the country's largest magnitude earthquake since the 1931 Hawke's Bay earthquake, but caused only minor damage and there were no casualties."
 * In the Earthquake section, I recommend a similar simplification. It was New Zealand's biggest earthquake magnitude since the 1931 Hawke's Bay earthquake, which remains as the deadliest earthquake in New Zealand history, and its magnitude was identical to the 2016 Kaikōura earthquake. could be simplified to "It was New Zealand's biggest earthquake magnitude since the 1931 Hawke's Bay earthquake, and equal in magnitude to the 2016 Kaikōura earthquake."

. I have added strikethoughs for feedback that has been resolved, and completed a second round of review. A couple of the discretionary items from the first round have not yet been addressed. The most significant remaining issue is the apparent inconsistency in the depth of the earthquake that I failed to pick up in the first round - sorry about that. Almost there now. Marshelec (talk) 00:53, 29 June 2024 (UTC)

Depth of the earthquake
I think I've done every suggestion now except for the depth. I'm also not that sure about how depth works and am confused with what to do here. In the preliminary report source it says,, but also says,. , since you know a lot more than me and I think you're the one who added the 30km-15km depth thing, could you please have a look at this? Thanks. ―Panamitsu (talk) 01:38, 29 June 2024 (UTC)
 * Unfortunately I can not figure out what is going on. I am unable to find more depths from another source. ―Panamitsu (talk) 08:38, 1 July 2024 (UTC)
 * What's confusing me is that a few of the sources like Geonet say that the rupture started at a depth of 30 kilometres, but do not mention the depth when it stopped. I have no idea why this is happening, and I wonder if it's because the 30km-15km number comes from a preliminary report which does not mention the 12km depth. I don't want to remove this number because it's the only one I can find, but it being in a preliminary report makes me a bit suspicious. ―Panamitsu (talk) 08:53, 1 July 2024 (UTC)
 * The 12 km depth seems to be first mentioned in the initial GeoNet report, published soon after the quake.. The University of Canterbury report (2010) includes this in the abstract initiating at about 30 km depth and rupturing upward and southwestward to about 15 km depth... On page 26 of this report, there is an important discussion about the analysis of the main shock, and this describes a "solution" that indicates initiation at 38 km depth. A different source, GeoScience Frontiers, describes: The updip and downdip edges of the rupture are estimated as 10 and 25.7 km, respectively . Rather frustratingly, in the Introduction, it says: This event appears to be one of the better recorded shallow (ca. 12 km) subduction thrust earthquakes. However, overall, the University of Canterbury source appears authoritative to me, and in my opinion is to be preferred over the immediate report from GeoNet. I suggest seeking another opinion, to try and close this out. I note that User:ChaseKiwi works on a lot of articles about physical geography, and volcanoes in particular. Perhaps this user might be willing to offer an opinion about how best to describe the depth of the earthquake in the article and Infobox. Are you happy to contact this user and ask for help ? Marshelec (talk) 21:55, 3 July 2024 (UTC)
 * Yeah are you able to give us a hand haha? The problem is that we were given an earthquake focal depth of 12 kilometres by GeoNet, but reports are giving us different depths for when the rupture started and ended, with the currently used source having depths that are always deeper than the focal depth. ―Panamitsu (talk) 22:40, 3 July 2024 (UTC)
 * Well I will look but do not get your hopes too high, although my experience in tectonic earthquakes like this one is that the later academic work is usually the winner. However I have just been puzzling over the the significantly different conclusions of 1991, 1999 and 2003 analysis of an 1987 Icelandic earthquake where I suspect you might not want to be in the same room as the two different schools of thought on volcano-tectonic interactions on the far side of the planet. ChaseKiwi (talk) 23:44, 3 July 2024 (UTC)
 * Yes, go with Fry et. al.s 2010 paper but mention the issues with initial interpretation later in the text. The initial data GeoNet had was biased by known issues that are in the 2010 paper. Can I suggest that although its not strictly necessary by the rules the two references in the lead could easily be moved to the text itself repeating what is said in the lead.   The phase "It had a depth of 12 km (7.5 mi), and an epicentre near Dusky Sound in Fiordland National Park," should read something like " Its rupture initiated at a depth about 38 km and it propagated upwards with an epicentre that was located at a depth of 12 km near Dusky Sound in Fiordland National Park,"
 * More detailed comments: Mahesh et al 2011 add nothing on depth issue. USGS at M 7.8 - 97 km WSW of Te Anau, New Zealand and USGS technical initial solutions gives lots of details of the initial epicentre modelling and you will note non preferred ISCGEM gives depth of 22.5 km ± 3.9.
 * In the text itself for a GA article I might have written something on the depth issues as explained by Fry et. al. like
 * "Early epicentre modelling by GNS and USGS agreed that the depth was about 12km and location .... large earthquakes like this one while commonly described as having a single point epicentre are best understood as a slip over a short period of time in a larger fault plane in three dimensions with the epicentre information being refined as more information and analysis becomes possible."
 * With regard to the minimum depth all that can be said is that no one found any obvious ground ruptures on land in a pretty rugged environment but as at least 70cm ground displacement and for all you know ground rupture occurred off the coast and was factor in the tsumani it may have reached surface. Perhaps best not said which is why not mentioned in academic papers. ChaseKiwi (talk) 08:00, 4 July 2024 (UTC)
 * Unfortunately I am completely lost even after your good help I have no idea what's going on and how to fix this problem. I'm not sure what's happening from here, if someone who understands the topic wants to update it, or if the GA review will be failed. ―Panamitsu (talk) 07:30, 5 July 2024 (UTC)
 * I am qualified but I don't work in the field; I think a range is reasonable and acceptable, and we have a citation for the given range from the Fry et al. 2010 paper. A single epicentre depth number for an event of this magnitude is misleading anyway. — Jon (talk) 13:04, 5 July 2024 (UTC)
 * Apologies - I waded in here and changed the Infobox depth parameter to the range given in the lead (30-15 km). If nothing else, I think we can safely put a range in the Infobox field, but feel free to change it as required (I added a precision tweak to keep the miles conversion as round integers). Perhaps it ought to be 38-8 km (UC report, p. 26)? — Jon (talk) 03:03, 5 July 2024 (UTC)
 * While would be usual to put the depth of epicentre in infobox which is a sort of average that becomes less and less relevant the bigger an earthquake so your change to a range may be reasonable. There will have been debate on this issue with other big earthquakes and a consensus may have been obtained in individual cases. That epicenter depths and locations are not a spot is well known but can be poorly understood by lay readers. I do not fully understand the maths of the various models myself but all epicentre calculations result from mathematical modelling. I am for moment not updating the article, especially lead myself as some of my general suggestions need citing, I was asked to provide a second opinion, and there are many other contributors to Wikipedia that have more understanding of seismology than me. ChaseKiwi (talk) 06:19, 5 July 2024 (UTC)
 * And I should have used term hypocentre rather than epicentre. What I mean about not being an expert on seismology rather than evidence based review of some fields of technical literature. ChaseKiwi (talk) 07:35, 5 July 2024 (UTC)
 * Do you think it'd be a better idea to state both the depth given by GeoNet and the depths from the preliminary report? I don't want to discount either source, especially given the quality of GeoScience. I'll write up a proper line tomorrow (I got out of bed to write this comment...) but my idea is that it could go something like this: . I don't want to scare quote, but to just indirectly show the reader that there is a conflict of information and they can decide what to do with the information themself. ―Panamitsu (talk) 13:13, 7 July 2024 (UTC)
 * It reads that you may not be aware that USGS and Geonet are usually independent high quality sources for the hypocentre and use different seismograph reports and models. The current (as opposed to historic Geonet entry (Internet Archive not accessible at moment) for depth is at M7.8 Dusky Sound Wed, Jul 15 2009:Technical and has hypocentre at 12±0 km while USGS has not defined the uncertainly. That Geonet has epicenter at Latitude -45.77 (± 0.0) and Longitude 166.59 (± 0.0) while USGS location is 45.762°S 166.562°E tells you that both models gave same hypocentre depth but certainly different longitudes which gives even more confidence that the hypocentre was 12km deep. Wording could read something like "Its rupture initiated at a depth about 38 km (24 mi) and it propagated upwards with an hypocentre that was located at a depth of 12 km (7.5 mi) near Dusky Sound in Fiordland National Park" as the term epicentre I used previously only applies to the surface projection of the hypocentre. ChaseKiwi (talk) 16:07, 7 July 2024 (UTC)
 * Works for me, esp. if we want to use a single number in the infobox — Jon (talk) 20:38, 7 July 2024 (UTC)
 * So you're suggesting that we omit the depth at the end of the rupture (if that's a correct way to describe it)? The problem Marshelec brought up is that the hypocentre depth is outside the 30-15km range. ―Panamitsu (talk) 06:58, 8 July 2024 (UTC)
 * Fry et al. (2010) says "Our best-fitting slip distribution [...] is characterized by a large rupture area of about 36 by 44 km, with a maximum slip of 8.9 m at 23 km depth on the main fault plane occurring 16 seconds after the rupture began. The rupture starts downdip at about 38 km depth and propagates upward reaching a depth of 8 km" (p. 26). The 30-15 km range in the abstract is the RMT which includes aftershock data. So, 12 km is inside their initial rupture range of 38-8 km. We could just as easily put the initially reported 12 km hypocentre depth in the infobox, or the 38-8 km rupture depth given in Fry 2010, who also say "focal depths in the Fiordland region are often poorly constrained using standard location methods" (p. 31) i.e. it's a complex geology that can frustrate model fitting. That's why I suggested using the range in the infobox rather than a single number. I don't think we should over-think it: for my money, I'd be happy with either, and I wouldn't fail a GA over it, but that's up to Marshelec! :) — Jon (talk) 23:27, 8 July 2024 (UTC)
 * Apologies for sounding like a broken record; I have no understanding of seismology. So after we mention the problems with Fiordland, what do we do with the 30-15 km range if we use the 38-8 km one, especially given that several sources say that it started at 30 kilometres? ―Panamitsu (talk) 08:00, 9 July 2024 (UTC)

@Panamitsu ok, sorry. Here's my suggestion: everywhere in the article where we say 12 km, describe it as the "initially reported" depth, and cite the archived GeoNet report. Leave the mentioned range as 30-15 km, citing Fry 2010, and resolve ambiguity with "subsequent analysis showed that..." or similar. Change the infobox depth back to 12 km. I'm happy to have a go if you'd rather. Jon (talk) 22:19, 9 July 2024 (UTC)
 * I think I have incorporated these suggestions, but please have a look as I may have misinterpreted something you said. Thanks for this. ―Panamitsu (talk) 00:36, 10 July 2024 (UTC)
 * Looks good to me! — Jon (talk) 01:38, 10 July 2024 (UTC)

Finishing the review
Great work on this article. It is an interesting account, and I have learned a lot during the review process. , and many thanks to both of you for assisting in helping with the understanding of how the different sources have analysed and reported on the event. Your assistance was really valuable and is most appreciated.Marshelec (talk) 22:02, 11 July 2024 (UTC)