4 ms·

> “Heat dome over North America, showing high temperatures predicted across the continent.” Such is the caption to a map showing no temperatures, or even poten

> “Heat dome over North America, showing high temperatures predicted across the continent.” Such is the caption to a map showing no temperatures, or even potential heat but rather a predicted delta from regional norms. This sort of behavior was exactly why I unsubscribed from the guardian after years of support post Snowden. I have no doubt this summer has been warmer on average. Here in Denver the smoke cover from Oregon’s fires must be contributing a measurable impact itself. That said our forecast for the coming week is highs between 85-90, which feels in line for most July’s since I’ve lived here.

edrxty 5y ago
It's not even a norms delta, it's a probability of any deviation from norms. The creators (https://www.cpc.ncep.noaa.gov/ https://www.cpc.ncep.noaa.gov/) even put out a survey a couple years ago trying to figure out if anyone actually understood the data. The maps have pretty much always looked like that in the summer over the past five years.
- sschueller 5y ago
  "The 6-10 day Outlook gives the confidence that a forecaster has, given as a probability, that the observed temperature, averaged over upcoming days 6, 7, 8, 9 and 10 will be in the range of one of three possible categories - below (B), normal (N), or above (A). For any calendar 5-day period, these categories can be defined by separating the 30 years of the climatology period, 1981-2010 (30 years), into the coldest 10 years, the middle 10 years, and the warmest 10 years. Because each of these categories occurs 1/3 of the time (10 times) during 1981-2010, for any particular calendar 5-day period, the probability of any category being selected at random from the 1981-2010 set of 30 observations is one in three (1/3), or 33.33%. This is also called the climatological probability. The sum of the climatological probabilities of the three categories at any point on the map is 100%. Graphical Information The colored shading on the map indicates the degree of confidence the forecaster has in the category indicated, where "B" and blue colors indicate "below-normal" and "A" and orange-red colors indicate "above normal". The darker the shading, the greater is the level of confidence. The numbers labeling the contours separating different shades gives the probability that the indicated category (A, B, or N) will occur. The probabilities of all three categories are implied on the map, and sum to 100%. The forecast probabilities given on the map generally fall far short of complete confidence (100%) in any single category. When the probability of the above (A) or below (B) category is greater than 33.33% by some amount, the probability of the opposite category declines by that amount, while the probability of the middle category remains at 33.33%. In the event that the "N" category is greater than 33.33%, the probabilities of both the "A" and "B" categories is each reduced by 1/2 the amount that the "N" category exceeds 33.33%. When the probability of "A", or "B" reaches 63.33% or higher, the odds of the opposite category reach a minimum allowed value of 3.33%, while the odds of the middle category are allowed to drop below 33.33%. The dashed contours on the map give the average of the temperature over the set of 30 observations for the calendar 5-day period during 1981-2010, in degrees Farenheit." Uff, not sure why the guardian picked that map when they could have found something more clear for the average reader, like me...
  - edrxty 5y ago
    Yeah it gets cited a ton because the colors look intuitive but really the map isn't particularly useful. Would honestly be better if the hadn't used "hot/cold" colors because as it stands it's basically the metrological equivalent to a stroop test...
mwill 5y ago
I mean, the second half of the article is almost entirely quotes from "Michael Wehner, a a senior scientist at the Lawrence Berkeley National Laboratory" Including the section “You expect hotter heatwaves with climate change but the estimates may have been overly conservative,” Wehner said. “With the Pacific north-west heatwave you’d conclude the event would be almost impossible without climate change but in a straightforward statistical analysis from before this summer you’d also include it would be impossible with climate change, too. That is problematic, because the event happened.” That does seem more compelling and newsworthy than "feels in line for most July's since I've lived here." to me? The map might not be the greatest graphic but the article still has meat right?
- lmm 5y ago
  Without statistics there is no meat; the Guardian is good at narrative but not so good at objectively reporting on trends. I see only one concrete point in the article; that the heat in certain areas broke records by 5.5 degrees is a relevant factual datum, but to understand what it actually means you have to ask things like: how many standard deviations is that? How long have these records been kept in this many places, and how many record-breakings are "normal" (I'd guess that in any given year the hot and cold temperature records would be broken somewhere).
  - wesleywt 5y ago
    You get these in the scientific articles, you cannot expect a newspaper article to include these. It should only include the interpretation of experts of these statistics. Not the statistics itself. That data is up to you to look into yourself...if you are an expert in the field.
    - arcturus17 5y ago
      The questions posed by your parent comment are rather high-level inquiries based on statistics and basic scientific reasoning. They do not require domain expertise to answer, just a decent command of statistics and a curious mind, and I agree with the parent that they are fundamental to knowing the scientific truth. > you cannot expect a newspaper article to include these. Why not? If you're driving a narrative, using hyperbolic language such as "punishing" or coining terms such as "heat dome", the burden-of-proof standards should be equally high on your part. I'd expect any scientific reporter in a newspaper like The Guardian to have basic scientific inquiry skills, and if the hypothesis is "this is caused by humans" to try hard to refute it, and not to leave questions like that unanswered. Is that too high of a standard?

edrxty 5y ago

It's not even a norms delta, it's a probability of any deviation from norms. The creators (https://www.cpc.ncep.noaa.gov/ https://www.cpc.ncep.noaa.gov/) even put out a survey a couple years ago trying to figure out if anyone actually understood the data. The maps have pretty much always looked like that in the summer over the past five years.

sschueller 5y ago
"The 6-10 day Outlook gives the confidence that a forecaster has, given as a probability, that the observed temperature, averaged over upcoming days 6, 7, 8, 9 and 10 will be in the range of one of three possible categories - below (B), normal (N), or above (A). For any calendar 5-day period, these categories can be defined by separating the 30 years of the climatology period, 1981-2010 (30 years), into the coldest 10 years, the middle 10 years, and the warmest 10 years. Because each of these categories occurs 1/3 of the time (10 times) during 1981-2010, for any particular calendar 5-day period, the probability of any category being selected at random from the 1981-2010 set of 30 observations is one in three (1/3), or 33.33%. This is also called the climatological probability. The sum of the climatological probabilities of the three categories at any point on the map is 100%. Graphical Information The colored shading on the map indicates the degree of confidence the forecaster has in the category indicated, where "B" and blue colors indicate "below-normal" and "A" and orange-red colors indicate "above normal". The darker the shading, the greater is the level of confidence. The numbers labeling the contours separating different shades gives the probability that the indicated category (A, B, or N) will occur. The probabilities of all three categories are implied on the map, and sum to 100%. The forecast probabilities given on the map generally fall far short of complete confidence (100%) in any single category. When the probability of the above (A) or below (B) category is greater than 33.33% by some amount, the probability of the opposite category declines by that amount, while the probability of the middle category remains at 33.33%. In the event that the "N" category is greater than 33.33%, the probabilities of both the "A" and "B" categories is each reduced by 1/2 the amount that the "N" category exceeds 33.33%. When the probability of "A", or "B" reaches 63.33% or higher, the odds of the opposite category reach a minimum allowed value of 3.33%, while the odds of the middle category are allowed to drop below 33.33%. The dashed contours on the map give the average of the temperature over the set of 30 observations for the calendar 5-day period during 1981-2010, in degrees Farenheit." Uff, not sure why the guardian picked that map when they could have found something more clear for the average reader, like me...
- edrxty 5y ago
  Yeah it gets cited a ton because the colors look intuitive but really the map isn't particularly useful. Would honestly be better if the hadn't used "hot/cold" colors because as it stands it's basically the metrological equivalent to a stroop test...

mwill 5y ago

I mean, the second half of the article is almost entirely quotes from "Michael Wehner, a a senior scientist at the Lawrence Berkeley National Laboratory" Including the section “You expect hotter heatwaves with climate change but the estimates may have been overly conservative,” Wehner said. “With the Pacific north-west heatwave you’d conclude the event would be almost impossible without climate change but in a straightforward statistical analysis from before this summer you’d also include it would be impossible with climate change, too. That is problematic, because the event happened.” That does seem more compelling and newsworthy than "feels in line for most July's since I've lived here." to me? The map might not be the greatest graphic but the article still has meat right?

lmm 5y ago
Without statistics there is no meat; the Guardian is good at narrative but not so good at objectively reporting on trends. I see only one concrete point in the article; that the heat in certain areas broke records by 5.5 degrees is a relevant factual datum, but to understand what it actually means you have to ask things like: how many standard deviations is that? How long have these records been kept in this many places, and how many record-breakings are "normal" (I'd guess that in any given year the hot and cold temperature records would be broken somewhere).
- wesleywt 5y ago
  You get these in the scientific articles, you cannot expect a newspaper article to include these. It should only include the interpretation of experts of these statistics. Not the statistics itself. That data is up to you to look into yourself...if you are an expert in the field.
  - arcturus17 5y ago
    The questions posed by your parent comment are rather high-level inquiries based on statistics and basic scientific reasoning. They do not require domain expertise to answer, just a decent command of statistics and a curious mind, and I agree with the parent that they are fundamental to knowing the scientific truth. > you cannot expect a newspaper article to include these. Why not? If you're driving a narrative, using hyperbolic language such as "punishing" or coining terms such as "heat dome", the burden-of-proof standards should be equally high on your part. I'd expect any scientific reporter in a newspaper like The Guardian to have basic scientific inquiry skills, and if the hypothesis is "this is caused by humans" to try hard to refute it, and not to leave questions like that unanswered. Is that too high of a standard?