That’s climatologist/macroeconomist/epidemiologist Steven Kopits snarks that I’m worrying about having to wear shorts in Madison on Sunday, when it hit 70 degrees. Just to clarify, global climate change has real implications.
That’s climatologist/macroeconomist/epidemiologist Steven Kopits snarks that I’m worrying about having to wear shorts in Madison on Sunday, when it hit 70 degrees. Just to clarify, global climate change has real implications.
Just a thought: adjust for population.
https://www.nhc.noaa.gov/pastdec.shtml No increase in frequency or intensity.
https://www.ncei.noaa.gov/access/monitoring/monthly-report/tornadoes/202313 Increase in detected tornadoes.
https://www.nature.com/articles/s41598-021-81143-5
Modern logging of tornado reports in the U.S. has begun in the 1950s, and the reports grew even more in the mid-1990s, mainly due to the increase in the detection of EF0–EF1 category tornadoes (weak tornadoes) after the installation of the NEXRAD Doppler radar system3,4,5,6,7,8,9,10. Other factors, such as better documentation, more media coverage, rise in the population, and storm chasing also contributed to this increased detection rate. Recent studies show that this secular trend (i.e., the long-term trend that is not due to seasonality) in the annual frequency of tornadoes has also been varying spatially. See, for example, Gensini and Brooks11 who showed that the temporal trends since 1979 have a significant spatial variability, and Moore12, who found that the number of tornadoes per year increased in the southeast U.S. and has decreased in the Great Plains and the mid-west regions. Studies by Farney and Dixon13 and Agee et al.6 show similar results.
Modern technology has improved the ability to detect and report localized severe storms accounting for the apparent trend of increased severe weather events.
Costs associated with severe weather can in significant part be attributed to movement of population into areas prone to severe events; e.g., Florida and Texas. Costs can also be associated with the increased valuation of property as average homes have become significantly larger since the 1950s.
Sorry for the incomplete duplicate. My internet has been acting up today and sometimes I’m not sure what has been sent.
Let’s see. A Cat 5 hurricane that hits a city with 5 million people is no more deadly than a Cat 4 hurricane that hits a city with 4 million people? I do declare Brucie – this is the dumbest thought ever.
The claim you must be making, in order for your claim to mean anything, is that detection and record-keeping have improved over any relevant period of increase in expensive weather events, so that we should discount the rise. So in the period from 2010 till now doesn’t count?
This is more excuse-making. This is the standard effort to dismiss evidence which is inconvenient to your position.
It wasn’t even hard to see what’s wrong with this “don’t forget population” distraction. You’re saying any old thing to talk down the damage from climate change.
technology does have an influence on detecting the events. but it really does not have an impact on the cost of the damage. and the yearly cost is rising. and the idea of higher property cost, etc has some validity. but normalizing to cpi eliminates the inflation component. so we are left with the fact that each event costs more money. but you really cannot use the argument that it is more costly simply because more people are living in those areas. populations are going to increase, that is what you want with your pro-life policies. you made that choice bruce. the fact of the matter is, each event is more costly and we are seeing more events because of global climate change, which is significantly impacted by man made actions.
what bruce should be saying is that we have negatively impacted global climate to the extent that it is not longer safe and economical to live in certain locations. therefore, to control escalating natural disaster costs, it is time to move significant population centers to locations less impacted by global climate change. that would be the honest statement coming from bruce.
Bruce Hall Costs associated with severe weather can in significant part be attributed to movement of population into areas prone to severe events; e.g., Florida and Texas.
If people moved to those prone areas, then presumably they did so because they believed moving there would improve their welfare. If they now have to move somewhere else, then their welfare loss is at least equal to the difference between first best and second best. That’s not a trivial cost and it’s likely to get much worse when the adverse effects of climate change really begin to bite towards the end of the century. You won’t be around to face those consequences, but your grandchildren will be the ones who suffer significant welfare loss.
Just a thought: adjust for population and property values.
https://www.nhc.noaa.gov/pastdec.shtml
No increase in frequency or intensity.
https://www.ncei.noaa.gov/access/monitoring/monthly-report/tornadoes/202313
Increase in detected tornadoes.
https://www.nature.com/articles/s41598-021-81143-5
Modern logging of tornado reports in the U.S. has begun in the 1950s, and the reports grew even more in the mid-1990s, mainly due to the increase in the detection of EF0–EF1 category tornadoes (weak tornadoes) after the installation of the NEXRAD Doppler radar system3,4,5,6,7,8,9,10. Other factors, such as better documentation, more media coverage, rise in the population, and storm chasing also contributed to this increased detection rate. Recent studies show that this secular trend (i.e., the long-term trend that is not due to seasonality) in the annual frequency of tornadoes has also been varying spatially. See, for example, Gensini and Brooks11 who showed that the temporal trends since 1979 have a significant spatial variability, and Moore12, who found that the number of tornadoes per year increased in the southeast U.S. and has decreased in the Great Plains and the mid-west regions. Studies by Farney and Dixon13 and Agee et al.6 show similar results.
Modern technology has improved the ability to detect and report localized severe storms accounting for the apparent trend of increased severe weather events.
Costs associated with severe weather can, in significant part, be attributed to movement of population into areas prone to severe events; e.g., Florida and Texas. Costs can also be associated with the increased valuation of property as average homes have become significantly larger since the 1950s.
https://www.census.gov/content/dam/Census/programs-surveys/ahs/working-papers/Housing-by-Year-Built.pdf
The devil’s in the details.
I’m curious how hard Bruce thinks it was to detect increased temperatures and hurricanes 75 years ago?? (admittedly a somewhat random number to pick) Does he think the measurements and ability to see hurricane damage on the coast was that much more difficult in 1950 than now??
It still fascinates me the number of morons visiting this blog all so they can make vapid arguments against a PhD, and then congratulate themselves after losing dozens of written arguments. Life is strange, as is a nation of illiterate people with the resources to not be illiterate,
Moses, thanks again for attempting to read what I wrote. Radar technology has improved over 75 years. It is especially relevant to localized storms (see the link I provided), but has also been used for detecting hurricanes that don’t make landfall. (If a tree falls in the forest and no one sees it happen, does it make a sound?)
With regard to hurricane intensity, I also provided you the link from the government with data so I presume you are not having an issue with that.
Aside from those minor data gathering issues, my point was that using costs without regard to population changes and increased value (beyond inflation) of property needs to be considered. It is not clear that is the case here given the much higher number of hurricanes of higher intensity mid-20th century.
I’m surprised that you see the data from the government and Nature (peer reviewed) as moronic. But I welcome your actual thoughts that refute them rather than mimicking pgl’s commentary style.
you hurricane frequency is for data related to continental landfall of hurricanes. that is a subset of hurricanes that occur. it may not be the best representation for hurricane formation in the Atlantic basin, in general. it ignores landfall in other places outside of the United States. many hurricanes do not make landfall. there is more ocean than land out there.
Amid all the horse race/election as usual reporting we get as the GOP nominates an adjudicated rapist and business fraudster Trump (https://www.politico.com/news/2024/02/16/trump-fraud-case-verdict-350-million-00141990) – reporters seldom ask what the GOP plans to do about climate change – but they have stated that they will repeal incentives for domestic clean energy production, electric vehicle manufacturing and electrification and pull the U.S. out of the landmark 2015 Paris climate accord ensuring our grandchildren face a near-hellish existence. https://www.npr.org/2023/11/08/1210965698/republican-candidates-climate-change
Although the way Trump is slurring words lately – difficult to get any idea of his policy on climate change – https://www.msn.com/en-us/health/other/trump-appears-to-be-showing-gross-signs-of-dementia-expert-points-to-new-evidence/ar-BB1iOWQW or any kind of policy beyond keeping himself out of jail and extracting revenge on his political enemies https://www.washingtonpost.com/politics/2023/11/05/trump-revenge-second-term/
Also – in regards to those that still insist on saying climate change is no big deal – here is what I would tell farmers in the upper MIdwest – ten years ago when I had to communicate about climate change – prepare for much earlier planting dates, plant wind resistant corn hybrids and prepare for more extreme weather events – for example – earliest tornado in Wisconsin history https://www.desmoinesregister.com/story/news/2024/02/09/how-rare-are-winter-tornadoes-in-wisconsin-what-causes-them/72534816007/
(Although because of U.S. taxpayer subsidized crop insurance – I’ve often thought that U.S. farmers get incentivized to plant on marginal lands and will be made whole if their crop gets destroyed by extreme events -ah – sweet socialism for farmers. BTW – Inflation Reduction Act does have incentives for conservation programs gee whiz – Once again – the Biden administration trying to help people out https://nationalaglawcenter.org/ag-the-ira-conservation-provisions-in-the-inflation-reduction-act/ )
Trump in many ways modeling Bibi Netanyahu. Main goal to keep himself out of jail – at huge costs the country. Although Trump is considerably more challenged by dementia so his policies could be even more unpredictable and dangerous. He has been known to make decisions based on wanting to show he is in charge and having no idea of the harm and confusion they cause.
Agreed. The general election will be won by winning swing voters in swing states. Rape, fraud, lies, serial adultery, bankruptcy, criminal convictions, two impeachment cases, disrespect to large parts of the population – swing voters need to be reminded, often, that yesterday’s inflation is a poor argument against criminality and self-dealing.
Trump opposes the views of most of the voting public on abortion, the environment, voting rights and a host of other issues. Biden’s age, not much different from Trump’s, is a poor reason to vote fora guy who opposes everything most of us believe.
I’d like to say I trust Biden’s election team to do the job, but they seem mostly to be on the defensive so far. Trump is appalling as a person and as a political leader. People need to be reminded.
You mean ‘cost/benefit’ implications, don’t you? Is a 70 deg March day in Wisconsin a cost or a benefit?
Clearly, you’ve misunderstood what a cost/benefit assessments is. The 70 degree day is neither a cost nor a benefit. Its effects are. Does the loss of plant and animal species due to climate change, the disruption of agriculture, higher energy bills at some times in some places and the rise in property loss and insurance costs outweigh a lower energy bill at some times and some place, and the ability to wear shorts in Madison in March?
Hope that clears up your confusion about how cost/benefit analysis is done.
You are making some good points. They are, however, weakened by the following:
1. The series is already adjusted for CPI, a large component of which is housing and the rest of which is correlated with housing prices.
2. U.S. population has increased almost exactly 50% since 1980. The average annual deflated disaster cost has increased far, far more than that.
3. Hurricanes hitting the U.S. as any definitive indicator of extreme events? Cherry-picking your data a little?
4. I read the Nature paper you linked to, and… I was a little surprised it got published at all. As a professional statistician of long standing – Distinguished Data Scientist at a large company with a multiple-award-winning data science organization, one of the top respondents on the stats.stackexchange site – I think I am qualified to comment on it to some extent, which I won’t here. Suffice it to say that had they used the price of lemons instead of the population density as an explanatory variable I am sure they would have found the price of lemons to be a significant driver of tornado counts.
5. “Costs associated with severe weather can, in significant part, be attributed to movement of population…” True!
But, as you say, the devil’s in the details.
Ithaqua,
Thanks for your reply. A couple of thoughts:
1. CPI for housing is a survey of rental costs then applied in general which may directionally account for some of the actual costs of purchasing/replacing homes, but not without some issues. Secondly, rental properties tend not to reflect the changes in the trend for homes being constructed over the past few decades.
https://www.brookings.edu/articles/how-does-the-consumer-price-index-account-for-the-cost-of-housing/
2. Overall population growth is not as important as where the growth has occurred. Some states have shown much slower growth than others over seven decades. Florida and Texas, where hurricanes and tornadoes are common, have seen much faster growth and into vulnerable areas. That may be why some areas are becoming uninsurable.
“Costs associated with severe weather can, in significant part, be attributed to movement of population…” True! I glad we can agree on the obvious.
3. Since the data chart for Menzie’s post was specific to the US, it seemed appropriate to examine data specific to the US.
4. I won’t attempt to challenge your statistical creds since I know nothing about you. You may disagree with the peer review process at Nature and that’s fine. But then, it’s only your opinion versus their process. Finding random correlations (your lemons) is a fun game, but analysis has to have some logic to it.
Argggh… meant to be a reply to Bruce Hall…
1. I agree that CPI is not a perfect indicator of housing values, at best an adequate one, but it does remove some of the effect.
2. “Overall population growth is not as important as where the growth has occurred.” However, state-level growth is mainly irrelevant. The reason is that growth in urban and suburban areas has been high, but in the much larger rural areas, it has been negative. Of course, the great bulk of tornados in, for example, Texas don’t occur in the DFW, Houston, San Antonio, and Austin areas, but out in the middle of nowhere, where there are fewer people to see them today than in 1980. See https://usafacts.org/data/topics/people-society/population-and-demographics/our-changing-population/state/texas/?endDate=2022-01-01&startDate=1980-01-01 for a nice map; for comparison with (spotted) tornado locations, https://data.statesman.com/tornado-archive/. Population growth in Harris County (Houston) alone over 1980 – 2023 amounted to 15% of the total population growth in Texas. I am quite sure it accounted for zero increase in the number of tornados spotted, given that there were already several million people living there. Without doing more work than I want, I can’t speak to the value of damaged property.
3. Given the two maps above, I think you can see that the paper really was extremely poor. All this data would enable one to compare tornado spotting rates with population and sensor quality and density over small areas over time, instead of over multi-state areas over time, while simultaneously estimating changing spatiotemporal differences in true tornado density. Picking 1991 as the arbitrary US-wide start date for Doppler radar when they could have *just asked the government* (https://vlab.noaa.gov/web/nws-heritage/-/introducing-nexrad as an interesting historical document) isn’t a good sign either. County tax assessors have the total assessed values of all property – commercial too – going back to the 1950s in most cases – and this would serve as a useful proxy for the property value available to be damaged in the county, addressing to some extent what we both agree is an important missing variable.
I’ll stop now, because it’s an interesting project, and I am now thinking about doing a much better analysis myself… which would be a significant commitment in time, but maybe I’ll do it anyway.