|
|
Background
First, I decided to collect the data on all earthquakes and hurricanes that occurred since 1900. I searched all available databases on the internet and also wrote a few emails to Dr. Rick McKenzie (Berkley Digital Seismic Network) and Dr. Madeleine Zibres (U.S. Geological Survey) asking which database is more complete and is better to use. When doing the internet search, I found that the large database on volcano eruptions is also available. Of course, most volcanoes are now known and monitored, and even big eruptions do not cause as much destruction and problems as earthquakes, but still, these eruptions might be caused by similar geological processes, so I decided to include volcanoes in the analysis as well. The data on hurricanes was also quite easy to find, but looking at them, I learned that hurricanes are actually seasonal events that occur only in late summer - early fall. This means that hurricanes, although they may cause a lot of damage, do not actually reflect the weather anomalies for the entire year. So, I decided to see if the information on other weather-related events is available, and found a large database on droughts (they are less seasonal than hurricanes).
I have also heard about sunspots that indicate the increase or decrease in solar activity. Some connections have been recently discovered between our planet’s weather and the number of sunspots (which steadily grows and then declines in an 11 year cycle), so I decided to compare hurricanes and droughts with sunspots just to see these connections myself.
A description of earthquakes, volcanoes, hurricanes, droughts, and sunspots is given below.
An earthquake occurs due to the release of stress between two tectonic plates (the major ones include Eurasian, Australian, Philippine, Juan de Fuca, Cocos, Pacific, Antarctic, Nazca, North American, Caribbean, South American, Scotia, Arabian, African and Indian plates). Over the years these huge plates drift against each other, building up a huge amount of tension. Until one or the other snaps, either partly going below the first, or climbing onto it. When this happens, a huge amount of force is released (sometimes compared to the force of a few nuclear bombs), making both plates vibrate violently. And while these tremors may last only a few seconds, the effects of a large earthquake can level entire cities. Yet, only a few ways exist to measure earthquakes (and not to predict them, at that), and the most convenient is, perhaps, the most complicated. In order to locate where the earthquake occurs, the region must have numerous seismic stations that are constantly recording seismic activity. After the earthquake has begun, all the seismic stations compare the time at which the earthquake hit them, and with what force, so by a gradual process of estimation (depth, longitude, latitude) the origin is determined.
Figure 1. Earthquake epicenters (an epicenter is a point directly above the origin of an earthquake, also known as a hypocenter) since 1963. It is evident that the majority of earthquakes occur along the tectonic plate boundaries. Image from en.wikipedia.org/wiki/Earthquake
A volcano is different from a mountain in the sense that it does not form due to the folding or crumpling of the tectonic plates, but because of the buildup of various substances that it has previously expelled. The type of eruption can vary, based on the thickness and adhesiveness of the magma. If the magma is runny and thin, then it will most likely flow down the volcano’s sides without any violent eruptions, yet if the magma is thick, gases trapped inside will have a harder time escaping, but when they do, they will cause a fierce explosion, sending large chunks of magma into the air. So far there exists no reliable way to predict when an eruption will occur, and where. Yet some scientists suspect that volcanic activity is somehow tied with the solar activity.
Figure 2. The basic layout of a volcano. Image from www.yourdictionary.com/ahd/v/v0139800.html
Volcanic eruptions occur only in certain places because the Earth's outermost shell - the lithosphere - is broken into a series of slabs known as lithospheric or tectonic plates. These plates are rigid, but they float on the hotter, softer layer in the Earth's mantle. There are 16 major plates. As the plates move about, they spread apart, collide, or slide past each other. Volcanoes occur most frequently at plate boundaries.
Figure 3. The global map of volcanoes. Image from http://vulcan.wr.usgs.gov/Outreach/AboutVolcanoes
A hurricane (to be defined as so, it must have winds higher than 75 m/h), nature’s strongest storm usually originates over water, at the equator, minus/plus 500 kilometers North or South (otherwise, the Coriolis force is too weak). However, there are a few conditions that must be present: the water over which the hurricane is to form must be adequate temperature (26.5+ degrees Celsius); there must be a rapid decrease in temperature with height above ground so that a potentially unstable condition in the atmosphere is possible; the winds at all levels of the atmosphere (30,000+) must be blowing at the same direction and with the same speed; there should be a high pressure area in the atmosphere flowing outward, above the growing storm. If all of these conditions are met, it is possible that a hurricane will form.
Figure 4. The above diagram shows the paths of all the hurricanes since 1950. Image from NOAA Coastal Services Center website.
Droughts are periods of time in which a specific place is much drier than normal in terms of moisture, and these events are an integral part of our weather. They are often caused by a lack of rain, which is in turn caused by events such as El-Nino.
Figure 5. The number of drought/famine disasters from 1974 until 2003. Image from www.em-dat.net/disasters/maps.htm
Sunspots are mysterious and powerful phenomena. These dark “spots”, sometimes as large as 50,000 km squared, float on the Sun’s surface, first expanding and then slowly contracting. The number of sunspots on the surface is not always the same, but varies with an 11 year cycle, over which the number of sunspots steadily grows, and then subsides. There is evidence that the number of sunspots has a direct influence on the weather patterns of Earth.
Figure 6. The typical sunspot. Image from www.wzw.tum.de/dvs/idolon/idolonhtml/natfrac.html
A correlation analysis is a way to find out how much two graphs (two sample populations with N-elements) are alike. Correlation is measured on a scale of 1.0 to -1.0 where 1.0 means 100% correlation and -1.0 is 100% anticorrelation. The formula to calculate a correlation between two populations, X & Y, where R is the correlation coefficient, is shown below:
where the mean values of X and Y are represented by:
An example of a perfect correlation is:
X = [1, 2, 3, 4, 5]
Y = [3, 4, 5, 6, 7]
IDL has several internal routines to perform the correlation analysis. Therefore, I did not have to program these formulas myself but used two IDL functions: CORRELATE (plain correlation) and C_CORRELATE (correlation with the specified shift, or lag, between X and Y).
X = [1, 2, 3, 4, 5]
Y = [3, 4, 5, 6, 7]
IDL has several internal routines to perform the correlation analysis. Therefore, I did not have to program these formulas myself but used two IDL functions: CORRELATE (plain correlation) and C_CORRELATE (correlation with the specified shift, or lag, between X and Y).