5 ms·

This article is a little bit misleading, unless I am misinterpreting his pictures. But that would be horrible for an article trying to explain pictures. Here i

by jsight 7y ago

This article is a little bit misleading, unless I am misinterpreting his pictures. But that would be horrible for an article trying to explain pictures. Here is the graphical archive for all 5 day forecasts of Irma: https://www.nhc.noaa.gov/archive/2017/IRMA_graphics.php?product=5day_cone_with_line https://www.nhc.noaa.gov/archive/2017/IRMA_graphics.php?prod... Now note the image that the article depicts for Hurricane Irma with this text: "These misunderstandings have real-world consequences. As Hurricane Irma approached Florida’s east coast in 2017, people near the cone’s center began preparing and even evacuating." The image and the graphical archive do not correlate. Where did the author get this image from? The actual NHC cone covers the entire width of Florida (apart from maybe the smallest western tip of the keys) for days before the storm arrives. Its actually pretty accurate. IMO, this article and its poor visualizations is part of the problem.

Allower 7y ago
No doubt its a map they made themselves displayed in their own coverage >_<
Jabbles 7y ago
That is very concerning. Here's an article written at the time: https://www.washingtonpost.com/graphics/2017/national/hurricane-irma/?%3Ftid%3D=sm_pg&noredirect=on https://www.washingtonpost.com/graphics/2017/national/hurric... "Forecasting models predicted the storm’s path with relative accuracy."
- albertocairo 7y ago
  Forecasts are getting better and better, but they still have large uncertainty. The piece explains the cone as the NHC itself explains it. Scroll down here: https://www.nhc.noaa.gov/refresh/graphics_at5+shtml/204603.shtml?cone#contents https://www.nhc.noaa.gov/refresh/graphics_at5+shtml/204603.s...
  - albertocairo 7y ago
    These are the most relevant paragraphs: NHC tropical cyclone forecast tracks can be in error. This forecast uncertainty is conveyed by the track forecast "cone", the solid white and stippled white areas in the graphic. The solid white area depicts the track forecast uncertainty for days 1-3 of the forecast, while the stippled area depicts the uncertainty on days 4-5. Historical data indicate that the entire 5-day path of the center of the tropical cyclone will remain within the cone about 60-70% of the time. To form the cone, a set of imaginary circles are placed along the forecast track at the 12, 24, 36, 48, 72, 96, and 120 h positions, where the size of each circle is set so that it encloses 67% of the previous five years official forecast errors. The cone is then formed by smoothly connecting the area swept out by the set of circles. It is also important to realize that a tropical cyclone is not a point. Their effects can span many hundreds of miles from the center. The area experiencing hurricane force (one-minute average wind speeds of at least 74 mph) and tropical storm force (one-minute average wind speeds of 39-73 mph) winds can extend well beyond the white areas shown enclosing the most likely track area of the center. The distribution of hurricane and tropical storm force winds in this tropical cyclone can be seen in the Wind History graphic linked above. Considering the combined forecast uncertainties in track, intensity, and size, the chances that any particular location will experience winds of 34 kt (tropical storm force), 50 kt, or 64 kt (hurricane force) from this tropical cyclone are presented in tabular form for selected locations and forecast positions. This information is also presented in graphical form for the 34 kt, 50 kt, and 64 kt thresholds.
spectramax 7y ago
The general problem that the article raises is still valid even though the pictures don’t match. The article is pointing out that the increasing circle sizes represent variation of centerline paths, not the actual size of the hurricane (which is a common misconception). Just because the plots don’t match doesn’t make the article irrelevant.
- jsight 7y ago
  I get your point, but there is another angle here. People have a strong tendency not to trust forecasters, despite relative accuracy. This article implies a lack of accuracy that wasn't actually the case. It would be incredibly easy to come away from this article with the wrong lessons.
  - nightski 7y ago
    Where do you see the article implying the forecast was inaccurate? To me the author is only saying that how people interpret it is inaccurate which is the entire point of the article.
    - albertocairo 7y ago
      Yes, exactly. The cone is as accurate as it can be, by which I mean: accurate but with a huge degree of uncertainty. The challenge is that some readers tend to read it as deterministic, and get it wrong by: 1. Thinking "If I'm inside of the cone I'm in danger, if I'm outside I'm good" (wrong! 1/3 of the time the storm will be outside of the cone!) 2. If I'm inside of the cone AND under its center line/dots of I'll worry a lot, but if I'm inside and in between the center of the cone and its boundary, I can relax (I'd be VERY careful with this, as it isn't true either).
      - spectramax 7y ago
        And your article drives this point home really well. I honestly don't understand the criticism - the hurricane example you picked from 2017 is just to make a point - which people in this thread are unnecessarily complaining about.
albertocairo 7y ago
Hi Jsight, Alberto Cairo here. I wrote the piece, but didn't design the maps, so I'll ask. In any case, even if there's an inconsistency between the maps in the piece—which are intended to be illustrative, I guess—and the NHC ones, it doesn't undermine, I think, the overall message, which is that, regardless of whether you are inside of the cone or outside and close to it, it's safer to prepare just in case. People should pay less attention to the center dots or line of the cone (many do that) and focus more on facts such as that (a) the cone is essentially tons of possible positions of the center of the storm in the next few days, (b) the storm may end up going outside of the cone 1/3 of the time (so I'd say it doesn't matter much that, say, Naples is outside of the cone, but very close to it on the map.) I've been in Miami since 2012 and lived through Irma, so I remember what I wrote in the piece well. When the central line of the cone was running over Miami, people over here got crazy, and people on the West coast of Florida were watchful, but a bit less concerned (there's a link to an LATimes article in the piece that describes that), again likely because some folks tend to focus too much on the center line and dots inside the cone. There's tons of evidence for this kind of misinterpretation (one academic paper is linked from the piece). The piece was reviewed by scientists at UM's School of Marine and Atmospheric Sciences; I won't say it's 100% correct, but I'm relatively confident the description, albeit simplified (I didn't explain what that "uncertainty" is for, instance!), is sound.
- albertocairo 7y ago
  By the way, thank you for being so thorough and attentive.
- hnburnsy 7y ago
  Truly nice of you to comment but did you not review the graphics and the source of them to make sure they matched your narrative? Seems important considering it is a piece about graphics and interpretation. Again appreciate you taking the time to join the discussion.
  - albertocairo 7y ago
    Apparently it was an oversight (moving the cone slightly to the right or left on the map) and will be corrected.
anonu 7y ago
The article is trying to explain what a confidence interval is. I think it does a good job and connecting it with real world consequence of where statistical representations and interpretation lead to people being complacent about risks.
OrgNet 7y ago
Here is the current hurricane: https://www.nhc.noaa.gov/archive/2019/DORIAN_graphics.php?product=5day_cone_with_line https://www.nhc.noaa.gov/archive/2019/DORIAN_graphics.php?pr...

by jsight 7y ago

Allower 7y ago

No doubt its a map they made themselves displayed in their own coverage >_<

Jabbles 7y ago

That is very concerning. Here's an article written at the time: https://www.washingtonpost.com/graphics/2017/national/hurricane-irma/?%3Ftid%3D=sm_pg&noredirect=on https://www.washingtonpost.com/graphics/2017/national/hurric... "Forecasting models predicted the storm’s path with relative accuracy."

albertocairo 7y ago
Forecasts are getting better and better, but they still have large uncertainty. The piece explains the cone as the NHC itself explains it. Scroll down here: https://www.nhc.noaa.gov/refresh/graphics_at5+shtml/204603.shtml?cone#contents https://www.nhc.noaa.gov/refresh/graphics_at5+shtml/204603.s...
- albertocairo 7y ago
  These are the most relevant paragraphs: NHC tropical cyclone forecast tracks can be in error. This forecast uncertainty is conveyed by the track forecast "cone", the solid white and stippled white areas in the graphic. The solid white area depicts the track forecast uncertainty for days 1-3 of the forecast, while the stippled area depicts the uncertainty on days 4-5. Historical data indicate that the entire 5-day path of the center of the tropical cyclone will remain within the cone about 60-70% of the time. To form the cone, a set of imaginary circles are placed along the forecast track at the 12, 24, 36, 48, 72, 96, and 120 h positions, where the size of each circle is set so that it encloses 67% of the previous five years official forecast errors. The cone is then formed by smoothly connecting the area swept out by the set of circles. It is also important to realize that a tropical cyclone is not a point. Their effects can span many hundreds of miles from the center. The area experiencing hurricane force (one-minute average wind speeds of at least 74 mph) and tropical storm force (one-minute average wind speeds of 39-73 mph) winds can extend well beyond the white areas shown enclosing the most likely track area of the center. The distribution of hurricane and tropical storm force winds in this tropical cyclone can be seen in the Wind History graphic linked above. Considering the combined forecast uncertainties in track, intensity, and size, the chances that any particular location will experience winds of 34 kt (tropical storm force), 50 kt, or 64 kt (hurricane force) from this tropical cyclone are presented in tabular form for selected locations and forecast positions. This information is also presented in graphical form for the 34 kt, 50 kt, and 64 kt thresholds.

spectramax 7y ago

The general problem that the article raises is still valid even though the pictures don’t match. The article is pointing out that the increasing circle sizes represent variation of centerline paths, not the actual size of the hurricane (which is a common misconception). Just because the plots don’t match doesn’t make the article irrelevant.

jsight 7y ago
I get your point, but there is another angle here. People have a strong tendency not to trust forecasters, despite relative accuracy. This article implies a lack of accuracy that wasn't actually the case. It would be incredibly easy to come away from this article with the wrong lessons.
- nightski 7y ago
  Where do you see the article implying the forecast was inaccurate? To me the author is only saying that how people interpret it is inaccurate which is the entire point of the article.
  - albertocairo 7y ago
    Yes, exactly. The cone is as accurate as it can be, by which I mean: accurate but with a huge degree of uncertainty. The challenge is that some readers tend to read it as deterministic, and get it wrong by: 1. Thinking "If I'm inside of the cone I'm in danger, if I'm outside I'm good" (wrong! 1/3 of the time the storm will be outside of the cone!) 2. If I'm inside of the cone AND under its center line/dots of I'll worry a lot, but if I'm inside and in between the center of the cone and its boundary, I can relax (I'd be VERY careful with this, as it isn't true either).
    - spectramax 7y ago
      And your article drives this point home really well. I honestly don't understand the criticism - the hurricane example you picked from 2017 is just to make a point - which people in this thread are unnecessarily complaining about.

albertocairo 7y ago

Hi Jsight, Alberto Cairo here. I wrote the piece, but didn't design the maps, so I'll ask. In any case, even if there's an inconsistency between the maps in the piece—which are intended to be illustrative, I guess—and the NHC ones, it doesn't undermine, I think, the overall message, which is that, regardless of whether you are inside of the cone or outside and close to it, it's safer to prepare just in case. People should pay less attention to the center dots or line of the cone (many do that) and focus more on facts such as that (a) the cone is essentially tons of possible positions of the center of the storm in the next few days, (b) the storm may end up going outside of the cone 1/3 of the time (so I'd say it doesn't matter much that, say, Naples is outside of the cone, but very close to it on the map.) I've been in Miami since 2012 and lived through Irma, so I remember what I wrote in the piece well. When the central line of the cone was running over Miami, people over here got crazy, and people on the West coast of Florida were watchful, but a bit less concerned (there's a link to an LATimes article in the piece that describes that), again likely because some folks tend to focus too much on the center line and dots inside the cone. There's tons of evidence for this kind of misinterpretation (one academic paper is linked from the piece). The piece was reviewed by scientists at UM's School of Marine and Atmospheric Sciences; I won't say it's 100% correct, but I'm relatively confident the description, albeit simplified (I didn't explain what that "uncertainty" is for, instance!), is sound.

albertocairo 7y ago
By the way, thank you for being so thorough and attentive.
hnburnsy 7y ago
Truly nice of you to comment but did you not review the graphics and the source of them to make sure they matched your narrative? Seems important considering it is a piece about graphics and interpretation. Again appreciate you taking the time to join the discussion.
- albertocairo 7y ago
  Apparently it was an oversight (moving the cone slightly to the right or left on the map) and will be corrected.

anonu 7y ago

The article is trying to explain what a confidence interval is. I think it does a good job and connecting it with real world consequence of where statistical representations and interpretation lead to people being complacent about risks.

OrgNet 7y ago

Here is the current hurricane: https://www.nhc.noaa.gov/archive/2019/DORIAN_graphics.php?product=5day_cone_with_line https://www.nhc.noaa.gov/archive/2019/DORIAN_graphics.php?pr...