Climate-Related Deaths and Insecurity

By Andy May – Re-Blogged From

In this post we will discuss the assertion that there will be more climate-related deaths due to man-made global warming. This is the fifth post in a series of seven.

There will be more heat-related deaths

The IPCC AR5 report does not have much to say regarding climate-related mortality, they do mention that heat-related deaths will increase in several places, the following is from page 49 of the WG2 technical summary:

“At present the worldwide burden of human ill-health from climate change is relatively small compared with effects of other stressors and is not well quantified. However, there has been increased heat-related mortality and decreased cold-related mortality in some regions as a result of warming (medium confidence).”

In 2014, a National Health Statistics Report “Deaths Attributed to Heat, Cold, and Other Weather Events in the United States, 2006-2010” was published, you can download it here. The report used death certificate data collected by the CDC. Over the five years of the study, 10,649 people died from weather-related causes, 31% from excessive heat, 63% from excessive cold and 6% from other weather-related causes like floods or lightning.

In a very interesting paper that compares different mortality databases (Dixon, et al., 2005) report that the CDC data from 1979 to 1999 found that there were 3,829 heat-related deaths due to weather conditions. Over the same period there were 15,707 deaths due to hypothermia or excessive natural cold, excluding anthropogenic cold deaths. Examples of anthropogenic heat and cold-related deaths are boiler room accidents, kitchen accidents and factory accidents involving excessive heat or cold. In examining several studies on heat and cold-related mortality, Dixon, et al. report:

“Interestingly, depending on the database used and the compiling U.S. agency, completely different results can be obtained. Several studies show that heat-related deaths outnumber cold-related deaths, while other studies conclude the exact opposite. We are not suggesting that any particular study is consistently inferior to another, but, rather, that it is absolutely critical to identify the exact data source, as well as the benefits and limitations of the database, used in these studies.”

In addition to the large difference between heat-related deaths and cold-related deaths, deaths in general are lower in the summer than in the winter. Figure 1 from Dixon et al., 2005, plots the gross mortality for 12 cities in the United States. Optimal mortality (meaning the fewest deaths) is always in the summer. Yet, some studies that show heat-related deaths are more numerous than cold-related deaths, “detrend” the gross mortality data by taking out the pattern in figure 1 and working with the residuals (Kalkstein, 1991). This takes out much of the effect of winter on mortality and is a questionable methodology, in my opinion, for our purpose.

Figure 1: The percent of all deaths by month for 12 U.S. cities, source Dixon et al., 2005

Detrending the gross mortality data is useful for examining the effect of a weather event on mortality. The residuals from the trend could be used to investigate the deaths due to a heat wave or snow storm, for example. But, we are not interested in weather events for this post, we are interested in the effect of long term climate changes on mortality. Thus, the trend is more useful to us than the residuals.

Another source of weather-related mortality statistics is the NCDC “Storm Data” dataset. Dixon, et al., 2005 provides a very thorough examination of this database and concludes:

“Temperature extreme deaths listed in NCDC’s Storm Data are skewed heavily toward heat-related deaths.”

The database relies on media reports of storm deaths and misses a lot of individual deaths. Because it is based on media reports, it is also skewed toward media reported storm events and misses individual weather-related deaths that occur in the absence of a storm. These occur more often in cold winter weather, thus creating a warm bias. Further, media reports, during an event, often contain errors and these are not corrected later in this database. For whatever reason, the NCDC storm database seems to overestimate heat-related deaths and underestimate winter-related deaths. See Dixon et al., 2005 for a complete discussion.

The CDC data has problems as well, but is far more comprehensive and should be more reliable, in my opinion. Classification of the cause of death is mostly done by professionals and it is quality controlled, but misclassifications do occur. However, it seems likely that weather-related cold deaths outnumber weather-related heat deaths, in the U.S. by a factor of two or more, as shown in the CDC data. Dixon, et al., 2005 do not agree and conclude:

“Of the datasets identified in this study, the one that appears to be least influenced by the above limitations [misclassifications of cause of death] is gross mortality. However, the gross mortality data must be detrended in order to remove a persistent winter-dominant death maximum. Another major limitation of gross mortality is in obtaining regional daily mortality as opposed to only daily mortality for metropolitan areas.”

Removing “a persistent winter-dominant death maximum” prior to the analysis of weather-related deaths introduces a bias and suffers from the fallacy of removing the forest to study a tree. It’s OK if the tree is all you want to study, not so useful if you want to study the forest.

If the Earth warms there will certainly be more heat-related deaths. But, warmer temperatures will reduce the number of cold-related deaths. Since, currently there are more deaths due to cold, we expect the avoided cold-related deaths to outnumber the increase in heat-related deaths. For the foreseeable future, global warming will save lives.

“The first complete survey for the world was published in 2006, and what it shows us very clearly is that climate change will not cause massive disruptions or huge death tolls. Actually, the direct impact of climate change in 2050 will mean fewer dead, and not by a small amount. In total, about 1.4 million people will be saved each year, due to more than 1.7 million fewer deaths from cardiovascular diseases and 365,000 more deaths from respiratory disorders. This holds true for the United States and Europe (each with about 175,000 saved), as for the rest of the industrialized world. But even China and India will see more than 720,000 saved each year, with deaths avoided outweighing extra deaths nine to one.” Lomborg, Bjorn. Cool It (Kindle Locations 681-686).

The 2006 paper that Lomborg refers to is Bosello, Rosen and Tol, 2006; “Economy-Wide Estimates of the Implications of Climate Change: Human Health“. The paper computes the climate-related health effects of a 1.16°C global average temperature change in 2050. Figure one shows where this temperature falls on the IPCC AR5 temperature prediction scenarios. It is roughly the temperature change associated with the RCP 4.5 scenario.

Figure 1, source IPCC AR5 The Physical Science Basis, page 21

“For almost every location in the world, there is an “optimal” temperature at which deaths are the lowest. On either side of this temperature—both when it gets colder and warmer—death rates increase. However, what the optimal temperature is, is a different issue. If you live in Helsinki, your optimal temperature is about 59°F, whereas in Athens you do best at 75°F. The important point to notice is that the best temperature is typically very similar to the average summer temperature. Thus, the actual temperature will only rarely go above the optimal temperature, but very often it will be below. In Helsinki, the optimal temperature is typically exceeded only 18 days per year, whereas it is below that temperature a full 312 days. Research shows that although 298 extra people die each year from it being too hot in Helsinki, some 1,655 people die from it being too cold.”

“It may not be so surprising that cold kills in Finland, but the same holds true in Athens. Even though absolute temperatures of course are much higher in Athens than in Helsinki, temperatures still run higher than the optimum one only 63 days per year, whereas 251 days are below it. Again, the death toll from excess heat in Athens is 1,376 people each year, whereas the death toll from excess cold is 7,852.”

“This trail of statistics leads us to two conclusions. First, we are very adaptable creatures. We live well both at 59°F and 75°F. We can adapt to both cold and heat.” Lomborg, Bjorn. Cool It (Kindle Locations 344-355).

The IPCC predicts that the global temperature in 2100 will be about 2°C higher than today. But nobody lives at the global average. Temperatures are rising more in Siberia and Canada than anywhere else and Antarctica is getting colder. Land temperatures rise faster than ocean temperatures and temperatures at night are rising faster than daytime temperatures. The reality of climate change isn’t necessarily bad, it doesn’t mean a fierce heat wave, the temperature changes we are considering are small and will be welcomed by many.

Lomborg estimates, in Cool It, that for the temperature increase of about 0.4°C observed from the 1970s to 2005, we get about 620,000 avoided cold deaths and 130,000 extra heat deaths. Thus, global warming (1970-2005) saved almost five times more people than it killed.

According to the Fourth National Climate Assessment extreme cold events and cold waves are less frequent. They also say, rather disingenuously, that heat waves “have become more frequent in the United States since the 1960s.” This is misleading because the 1960s were unusually cool, see figure 2 and because the most severe heat waves, in the U.S. occurred in the 1930s according to the EPA (figure 3).

Figure 2 (data: HADCRUT 4)

Figure 3 (source EPA, 2016)


The IPCC WGII AR5 Technical Summary claims that climate change will cause great uncertainty and increase population insecurity. Most of the claims in the section are easily dismissed as silly, like the threat of war will be increased, or cultural values will be eroded. But, some are worth discussing, for example:

“Climate change over the 21st century is projected to increase displacement of people (medium evidence, high agreement). Displacement risk increases when populations that lack the resources for planned migration experience higher exposure to extreme weather events, in both rural and urban areas, particularly in developing countries with low income. Expanding opportunities for mobility can reduce vulnerability for such populations. Changes in migration patterns can be responses to both extreme weather events and longer-term climate variability and change, and migration can also be an effective adaptation strategy. There is low confidence in quantitative projections of changes in mobility, due to its complex, multi-causal nature.”

Warming, unlike war or repressive governments, is a gradual thing and air conditioning is currently not very expensive as long as electricity is available and cheap. If an area becomes too warm, people may want to move to cooler places, but we are not talking about rapid warming. Warming is expected to be at a rate of 0.1 to 0.2°C per decade. This is also a time of rapid increases in standard of living. As people become more affluent, they will have many more options to adapt to the changes in-place or by moving to a cooler locale. “Wealthier is Healthier” as noted by Pritchett and Summers in 1993.

The IPCC also claims that climate change will slow down economic growth and make poverty reduction more difficult, increase hunger, etc. But, we have examined these issues elsewhere in this series of posts and shown that there is no evidence, other than unvalidated models, to support these projections.


We have shown that, according to the CDC death certificate database, more lives will be saved due to lower cold weather mortality than caused by higher temperatures in the U.S. In addition, the World Health Organization has found that studying climate-related mortality is very complex and the results of their study are uncertain, they conclude:

“Climate change is expected to affect the distribution of deaths from the direct physiological effects of exposure to high or low temperatures (i.e. reduced mortality in winter, especially in high latitude countries, but increases in summer mortality, especially in low latitudes). However, the overall global effect on mortality is likely to be more or less neutral. The effect on the total burden of disease has not been estimated, as it is unclear to what extent deaths in heat extremes are simply advancing deaths that would have occurred soon in any case.”

The U.S. and many other developed countries in the temperate latitudes have good statistics on the cause of death and in these areas, it seems highly likely that cold weather-related deaths outnumber hot weather-related deaths and it follows then, in these areas, that warming will reduce the weather-related deaths overall. As shown in Dixon, et al. many more deaths occur in the winter than in the summer in any case, regardless of the cause. However, this is in temperate areas and 50% of the surface of the Earth lies between 30°N and 30°S. The data for this latitude band is incomplete and not as accurate. So, it seems likely that we should accept the WHO conclusion that we simply do not know what the global effect of global warming will be over the entire world, but it will likely be “more or less neutral.”

Climate changes, whether natural or man-made, are very likely to make some populations insecure. Droughts, flooding, severe weather, and other weather-related problems will increase in some areas and other areas will benefit. Due to these changes, people will move, as they always have in such circumstances. The cold period from about 100 AD to 400AD, near the end of the Roman Empire, may have initiated the so-called “Migration Period.” This resulted in the Visigoths and other migrating people conquering large parts of the Western Roman Empire. The timeline here marks the cold period on the Central Greenland Temperature record.

Each city on the Earth has an ideal temperature, as Lomborg has pointed out. This ideal temperature is usually close to the average local temperature in the summer. Above this temperature and below this temperature deaths increase. The range of ideal temperatures is large, much larger than the average warming expected in the various climate model projections for the next several hundred years. Thus, as man has spread over the planet he has already adapted to more extreme temperatures than we are likely see anywhere in the next few hundred years. In Helsinki, where the ideal temperature is 59°F, people do well. They also do well in Athens where the ideal temperature is 75°F and in Philadelphia where they do well at 80°F. It does not seem that climate change is a health risk.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s