User:Kitography

Hi! I'm Kitography, I like to fact-check random facts. Citations are a good thing for the world.

Partially Confirmed: 3,084 died in 1993 South Asian monsoon flood

The Mystery

As of writing this, this appears on the List of deadliest floods without linking to an article; it got there copied from the page on Flood. The entry was added to the page Flood as part of a much larger edit on 22nd of Nov, 2008 by Poetaris. This entry in particular was not cited.

I've asked on Poetaris' talk page where that information came from. In doing so, I found out that the user has been blocked by user YellowMonkey, who seems to have left Wikipedia. This message from YellowMonkey claims that this user was a Wikipedia:Sock farming. I'm therefore not expecting a response.

An early entry in the Flood edit which introduced this fact did have a citation: "http://www.nbc10.com/news/4030540/detail.html+Worst+Natural+Disasters+In+History". Unfortunately that particular link no longer leads to the article, and was not archived by the Wayback Machine. However, a similar link was referenced in an old Reddit post. That page, by NBC10, was archived by the Wayback Machine. That similar article does not list any 1993 incident.

At the time this information was added, citations already existing on the page were twelve references, twelve external links, and two options for further reading. Two in particular looked very relevant.

• Further reading: "O'Connor, Jim E. and John E. Costa. (2004). The World's Largest Floods, Past and Present: Their Causes and Magnitudes [Circular 1254]. Washington, D.C.: U.S. Department of the Interior, U.S. Geological Survey." looked promising, but the article only has two tables listing floods (on page 3 and 10), neither of which includes any flood that occurred in 1993.
• Reference "Southasianfloods.org" looked relevant. Unfortunately, it currently links to a page titled "South Asian Foods" with a footer stating it was Copyright 2021. Checking a much earlier version of the site, this did used to link to a page with an list of floods. That webpage reads: "The archives are sourced from the 'NASA-supported Dartmouth Flood Observatory' provided at www.dartmouth.edu/artsci/geog/floods/ " - that is already listed as a source.

Following up on the source of southasianfloods.org, I found the list of floods compiled by Dartmouth they referenced still exists, and says 3,083 people died in the event with identification tag "1993-037" (the monsoon flood). That is one person off from the death toll claimed in the list of floods, which implies this isn't "the" reference used (or perhaps a typo?), but it looks like a reputable source.

Correcting the Minor Record (One Death Off, But Also Adding Citations)

To update this reference, I looked at the live website hosting the Dartmouth Flood Observatory, which is active, but does not seem to have the detailed notes (or any notes) about floods from before the year 2000, and seems to be looking a very restricted number of deaths (just India, and just a week of the flood period has 232 deaths, I guess) on line 778 of it's hard-to-read text-table. Failing to find a source with the full information available from the archived link, I've updated the reference to be to that link, and the number of deaths to match.

In doing this, I found an article which seems to reference the same flood, titled 1993 India floods. From the citation I found from the Dartmouth Flood Observatory, it lists that of the 3083 deaths, 1050 were from India, less than the 1800 from Nepal. That makes the tile India floods seem inappropriate, so I've asked on the talk page to rename it (or whether I should make a bigger page).

This claim was referenced by 1993#Date unknown but the dates are very much known, so I've moved it to the official start according to the reference for death count used, and deleted that section, but added other sources that could support the "over four thousand" claim originally listed.

The list of floods page formerly improperly implied that this flood only hit India, so I've updated it to refer to this as South Asia instead.

This didn't just happen in Pakistan

When searching for the remarkably precise "3,084" deaths originally listed using a search engine, I came across a number of websites, tweets and even what looked like news articles that cited this number (unfortunately I haven't been able to get wayback machine to take a copy of the search results for a duckduckgo search, so these are the live ones). This included the page Climate of Pakistan which has implied since a September 2010 edit that the 3,084 deaths all occurred in Pakistan. The source which shows the 3,083 deaths separates out 15 of them as having occurred in Pakistan, so this is a huge difference. This change was made without adding a citation so I've contacted the user on their talk page.

I've done my due diligence looking for the citation from the above claim in case it was improperly cited. For instance, at the time this Monsoon was added to Climate of Pakistan, there were only thirteen references on the page, so I went through all of them and found nothing relevant:

• (1) "http://www.pakmet.com.pk/latest%20news/Latest%20News.html" this is infact a latest news link, the earliest wayback capture is February 2010, where it does include weather advisories, but nothing about anything so far back as 1993.
• (2) "http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=1559&tstamp= " the only reference to Pakistan on this states it's hottest ever recorded temperature occured.
• (3) "http://punjabgovt.nic.in/punjabataglance/SomeFacts.htm" No longer online, this is an Indian website with information about Punjab. There is nothing directly about Pakistan (nor about the 1993 Monsoons) here.
• (4) "http://www.pakmet.com.pk/latest%20news/Latest%20News.html" - this is identical to (1)
• (5) "http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=1559&tstamp=" - this is identical to (2)
• (6) "http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=1559&tstamp=" - this is identical to (2)
• (7) "http://www.pakmet.com.pk/latest%20news/Latest%20News.html" - this is identical to (1)
• (8) "http://www.essl.org/ECSS/2007/abs/02-Case-study/sheikh-1-sec02.oral.pdf" - looks to be about a thunderstorm in 2001
• (9) "http://www.ncdc.noaa.gov/oa/climate/extremes/2001/july/extremes0701.html" - the reference to Pakistan here is about torrential floods between July 21 - 31, 2001 "killed some 200 people", nothing to do with 1993
• (10) "http://www.dawn.com/2002/07/27/local1.htm" - unfortunately this website is down, and wayback has not archived this particular entry on the site.
• (11) "http://www.pakmet.com.pk/cdpc/Climate/Karachi_Climate_Data.txt" - this is heat and rainfall data for the Karachi Airport, 1993 is not noted as the year with the most rainfall in any month
• (12) "http://www.pakmet.com.pk/journal/july2003floods-sindh.htm" - this has a lot of information on July 2003 floods in Pakistan, but 1993 is not mentioned
• (13) "http://www.dawn.com/wps/wcm/connect/dawn-content-library/dawn/news/pakistan/03-pakistan-flood-crisis-bigger-than-tsunami-haiti-un-ss-05" - this looks like an article about Pakistan floods in 2010, 1993 is not mentioned.

I've also searched links from the Climate of Pakistan to other pages. For instance, there was a link to the page Geography_of_Pakistan#Climate which does not mention this 1993 incident. However, there was also a link to List of extreme weather records in Pakistan which does mention this, including the specific "3,084" deaths added in an edit with no citation that was added by the same user, Nabil rais. That edit does link to the Monsoon page. Looking at that page, nothing mentions flooding in 1993 at all. It does link to the Monsoon of South Asia page which has been split off since March, 2011 and also doesn't mention anything about 1993 or Pakistan beyond the initial link as part of South Asia.

While I await a response from Nabil rais, I have adjusted the wording, and added a citation on both of the pages this implication that the deaths were in Pakistan appeared: Climate of Pakistan, List of extreme weather records in Pakistan. I've also added an entry to the list of floods in Pakistan where it wasn't mentioned at all, because fifteen people did die, which is notable.

Beyond Wikipedia

I'm hesitant to link to the websites in play here due to how linking is used to measure the impact of websites, and I do not want to promote them further. Unfortunately, I haven't figure out how to store a DuckDuckGo Search results page on Wayback Machine, so I'm linking, below to the Wayback of the sites themselves:

• One article on "Not PC" ""Are Pakistan Floods Really the Worst in History Because of Climate Change"". 2022-09-04. uses the same line precisely as the article Climate of Pakistan did: "In 1993, flooding during Monsoon rains killed 3,084 people." as part of a claim that the 2022 floods were not the "worst in history" for Pakistan. I've attempted to leave a comment linking to this write-up.
• Another article from the "Daily Sceptic""Are Pakistans Floods Really the Worst In History Because of Climate Change". 2022-09-02. looks like an earlier, longer version of that article, where the author writes "The death toll from the current floods is estimated at just over 1,000 people. The two-minutes research I did found four Pakistan floods – 1950, 1992, 1993 and 2010 – which killed more than that." To leave a comment on this article requires paying money to sign in.
• This claim is also echoed over Twitter. I don't have a Twitter.
• Several links come up to papers which do not quote this statistic.
• There are copies of the wikipedia article "Climate of Pakistan" for some reason.