Dust flux, Vostok ice core

Dust flux, Vostok ice core
Two dimensional phase space reconstruction of dust flux from the Vostok core over the period 186-4 ka using the time derivative method. Dust flux on the x-axis, rate of change is on the y-axis. From Gipp (2001).

Wednesday, April 25, 2012

Natural and Keynesian disasters

It seems that every week we are treated to a breathless dissertation of some natural disaster or other that is going to doom us. A few months ago I commented on one such impending crisis. Every so often I like to look at my stock portfolio, and it gets me thinking about disasters again.

It isn't a new phenomenon.

A common theme is a possible causative role for human activity. Indeed, there are reasonable physical mechanisms by which human activities may contribute to natural disasters. We have altered the composition of the atmosphere to measurable effect, and some believe that some component of recent climate variability is due to this action.

Human activity and earthquakes

Does human activity cause earthquakes?

We have known for a long time that injection of liquids into areas where there are tectonic stresses may reduce friction enough to trigger earthquakes--and direct relationships between (thus far, small) swarms of earthquakes and fluid injections at depth by the oil industry has been demonstrated in southwestern Ontario (Mereu et al., 1986), Colorado (Major and Simon, 1969), and other places--although no connection has ever been established between such action and large earthquakes.

The construction of large hydroelectric projects, which fills vast reservoirs with water, changes local mass balance and fluid pressures at depth, which has been suggested as a trigger for large earthquakes, such as the large Gujarat earthquake in January 2001. This mechanism has not been positively established as a trigger.

Given the amount of human activity, we might expect that there are more earthquakes now than in the past. It certainly seems that way reading the papers. Agencies like USGS (United States Geological Survey) and BGS (British Geological Survey) are valuable sources for data on geophysical disasters so that we may determine whether there is any correlation between catastrophes and human activities.

1900 - 1999

1900 13 1930 13    1960   22 1990     13
1901 14 1931 26    1961   18 1991 10
1902      8 1932 13    1962   15 1992N 23
1903 10 1933   14   1963   20 1993M 16
1904    16 1934   22   1964   15 1994     15
1905    26 1935   24   1965   22 1995E   25
1906    32 1936   21   1966   19      1996     22
1907    27 1937   22   1967   16      1997     20
1908    18 1938   26   1968   30      1998     16
1909    32 1939   21   1969   27      1999     23
1910    36 1940   23   1970 29
1911    24 1941   24    1971   23
1912    22 1942   27    1972   20
1913    23 1943* 41   1973   16
1914    22 1944   31   1974   21
1915    18 1945   27   1975   21
1916    25 1946   35   1976$ 25
1917    21 1947   26   1977   16
1918    21 1948   28   1978   18
1919    14 1949   36   1979   15
1920     8 1950   39   1980   18
1921    11 1951   21   1981   14
1922    14 1952   17   1982  10
1923    23 1953   22   1983   15
1924    18 1954   17   1984   8
1925    17 1955   19   1985   15
1926    19 1956   15   1986#  6
1927    20 1957   34   1987   11
1928    22 1958   10   1988    8
1929    19 1959   15   1989    7

Total 1900-1997 = 1960 events = 20 per year

* Most active year since 1900
# Least active year since 1900
$ Year with most people killed since 1900 (295,000 - 699,000;
 dominated by the Tangshan quake with casualty estimate from
 255,000 - 655,000)
N First full year of operation on NSN/digital recording system
M Year moment magnitude quotes were introduced
E Year energy magnitude quotes were introduced

Statistics were compiled from the Earthquake Data Base System of the
U.S. Geological Survey, National Earthquake Information Center, Golden CO

The table above used to appear on the USGS website years ago--but I can no longer find it. Part of it appears here.

Magnitude 7 quakes are pretty large, and it is reasonable to assume that we would be aware of them anywhere in the world going back to the beginning of the last century. We could not do the same sort of study for small earthquakes, as a those in remote settings would go undetected.

The maximum number of large earthquakes in the last hundred or so years occurred in 1943. There is no detectable correlation between large earthquakes on a global scale and human activity, although local correlations may exist which have not yet been established.

Human activity and catastrophe

However as we peruse the Munich Re statistics, we note that the economic costs of natural disasters do correlate with human activity. This correlation is due to the rapid spread of occupation of previously marginal lands, and the greater concentrations of wealth and developed property in geologically dangerous areas.

Comparison of distribution of insurable losses worldwide in 2011 and from 1980 to 2011. Figures from Munich Re annual report.

For instance, we note that there has been a remarkable increase in the fraction of insured losses due to natural catastrophes in Asia. From 1980 to 2011, Asia was responsible for 13% of insured losses worldwide. But in 2011 alone, the fraction of worldwide losses in Asia spiked to 44%. A lot of this was due to the damage done by the Tohoku earthquake and tsunami, but this was only responsible for 30% of Asia's insurable losses. This increase in importance is a reflection of the rapid growth of property values in Asia.

Data source here; graphic from Munich Re annual report.

We see a steady climb in the number of US Federal disaster declarations over the past sixty years. Are we to conclude that there have been more disasters? Or is it simply that more people have been impacted by them? If so, why?

Over the last 25 years it looks like US tornado activity is up, but worldwide hurricane activity is down. Global earthquakes, as we saw above, don't really have a trend. But there has been a definite increase in the number of declared disasters in the US. These have not been caused by an increase in the number of disasters. What we are seeing is the effect of the growth of suburbs and the expansion of expensive coastal properties in hurricane-prone areas of the United States.

In tornado-prone areas, sprawl like that depicted above greatly increases the chances that houses are impacted by tornadoes. Similarly, sprawl into forested areas increases the likelihood of forest fires consuming the neighbourhood; sprawl up the sides of hills increases the likely impacts of landslides, etc.

Keynesianism and the growth of suburbs

The symbiosis between the Keynesian expansion of the economy and the growth of suburbs in US cities has been ably discussed by Beauregard (2006). Sprawl was driven by the flow of money, the "American dream" of owning a home in the suburbs, and facilitated by the widespread ownership of cars. The suburbs were designed with cars in mind.

The growth of suburbs fulfilled two roles. Lots of houses were available for new buyers, which kept prices down; and city governments discovered that developer's fees and the new land taxes initially exceeded the maintenance cost of the new roads and infrastructure built to support them,. Unfortunately, as time passed and the infrastructure aged, soon maintenance costs exceeded tax revenues, necessitating another round of growth. Suburbs were able to maintain the required level of growth for a few decades, but we are reaching the point everywhere (it seems) where there cannot be enough new growth to maintain our crumbling infrastructure.

The mindset of the "ownership society" really drove demand for housing, and the best places to expand were in the southwest, so that cities like Phoenix and Las Vegas really grew. Low interest rates plus easy money led to a bubble in house prices and an explosion of sprawl.

The Austrian school of economics teaches us that easy money leads to malinvestment. Suburban growth certainly seems to qualify. Our urban sprawl malinvestment has left us with the interwoven problems of unlivable cities, financial crisis, and increased death and destruction from natural disasters.


Beauregard, R. A., 2006. When America became suburban. University of Minnesota Press, Minneapolis, 271 p.

Major, M. W. and R. B. Simon, 1968. A Seismic Study of the Denver (Derby) Earthquakes, 63 Q. Colo. School Mines 9.

Mereu, R. F., J. Brunet, K. Morissey, B. Price & A. Yapp, 1986. A study of Microearthquakes of the Gobles Oil Field Area of Southwestern Ontario, 76 Bull. Seismol. Soc. Am. 1215.

Nicholson, C. and R. L. Wesson, 1990.  Earthquake Hazard Associated with Deep Well Injection--A Report to the U.S. Environmental Protection Agency, U.S. Geol. Surv. Bull. #1951.

No comments:

Post a Comment