Elements for Life

[

] 125

of a number of major earthquakes with a magnitude greater

than 5.0 and a focal depth of less than 50 km.

NASA analysed data from a number of satellites including:

NASA’s Moderate Resolution Imaging Spectroradiometer

(MODIS) on Terra and Aqua; NASA’s Atmospheric Infrared

Sounder (AIRS) on Aqua; NOAA’s Geostationary Operational

Environmental Satellite (GOES) and Polar Orbiting

Environmental Satellites (POES); the French Centre National

d’Etudes Spatiale’s Detection of Electro-Magnetic Emissions

Transmitted from Earthquake Regions (DEMETER); and ground

observations The rationale for using this ensemble of observa-

tions was sufficient spatial and temporal coverage of precursor

signals to enable correlations to past earthquake events.

Bhuj, India

On 26 January 2001, a magnitude 7.7 earthquake struck Bhuj

(Gujarat), India. The quake occurred during a period of clear

weather, which is ideal for TIR observations. NASA analysed

data fromTerra/MODIS, which observes the Gujarat region

twice a day. As shown in the image above, the main stress was

released along the Katrol Hill and Mainland faults.

The time

series in the upper right shows the average land surface temper-

ature (LST) anomaly (departure from average) as measured by

the nighttime overpass of Terra/MODIS from December 2000

to February 2001 as well as a running average that smoothes

out some of the more extreme values – the day of the quake

is shaded in grey. The time series shows that there is a

pronounced and statistically significant positive LST anomaly,

with temperatures running up to 4 degrees Celsius above the

average background values. This anomaly starts five to six days

prior to the quake in the area within 200 km of the epicentre.

Shown in the bottom portion of the illustration are a series of

images from Terra/MODIS taken between 18 January and 21

January that show the evolution of the LST anomaly over a

100 km

region a few days prior to the quake. The anomalously

high temperature values are observed over this period, and

values return to near normal on 22 January. This kind of

anomaly could be associated with stress-related changes of air

near the ground in response to the release of radon gas and

changing humidity levels. (Water vapour and radon would be

highly absorbed in the 10-11 μm range).

Colima, Mexico

On 21 January 2003, a magnitude 7.8 earthquake struck

Colima, Mexico. NASA used a variety of independent data

sources to analyse the atmospheric variations caused by the

quake. For this analysis, we had access to data from MODIS

on both Terra and Aqua – Aqua had not yet launched at the

time of the Gujarat eruption – and we also analysed ground-

based temperature data. The appearance and temporal

evolution of the atmospheric variations are synchronised in

time and demonstrate a similar spatial distribution in all of the

datasets used. Using nighttime emissions from polar orbiting

satellite data, we have analyzed the LST data over 90 days by

means of the 11-12 metre emissivity ratio covering an area of

100 km

around the epicentre.

The graph illustrates the variations of the nighttime LST for

Terra (curve A) and daytime LST for Aqua (curve B) for the

period from 1 December 2002 up to 1 March 2003 for the

square 100km

around the Colima epicentre. Between 15 and

17 January, a pronounced LST increase can be observed for the

area close to the epicentre of the event. To verify the signifi-

grated, interdisciplinary approach is advocated by US seis-

mologist Ari Ben-Menahem, who said: “Unless we launch a

concentrated interdisciplinary effort, we will always be

surprised by the next major earthquake.”

Current scientific research indicates that satellite thermal

imaging data has not only revealed stationary (long-lived) thermal

anomalies associated with large linear structures and fault systems

in the Earth’s crust

but also transient (short-lived) features prior

to major earthquakes.

These short-lived anomalies:

• Typically appear between four and fourteen days before

an earthquake

• Affect regions of several to tens of thousands of square

kilometres

• Display a positive deviation of 2-4 degrees Celsius or more

• Die out a few days after the event.

These anomalies are not simply the result of thermal variations

caused by a heat pulse rising from below the Earth’s crust, as

the speed at which the anomalies appear and disappear is much

more rapid than what is seen with a heat pulse.

Feasibility of space observations for earthquake studies

Observational and scientific evidence collected over the last

20 years confirms that EM phenomena often accompany or

precede earthquake events. Recent studies also confirm that

there is strong coupling between EM activity in the atmos-

pheric boundary layer and the ionosphere – both direct

coupling through EM phenomena and indirect coupling

through tectonically forced vertically propagating gravity waves

that are strongly related to enhanced tectonic activity. The

concept of lithosperic-atmospheric-ionospheric coupling

(LAIC)

links increased gas emanations in advance of seismic

activity to a chain of physical processes involving ionisation

of air molecules and plasma chemical reactions in the ionos-

phere. LAIC views the ionosphere as part of the global electric

circuit, and suggests that the ionosphere immediately reacts to

changes in electric properties near the ground. A number of

different spacecraft have detected these kinds of phenomena

from space over the past few decades.

It is unlikely that any single existing method for advance

earthquake detection – for example magnetic field, electric

field, thermal infrared, surface latent heat flux (SLHF), and

global positioning system (GPS)/total electron count (TEC) –

can provide sufficient information to detect potential earth-

quake phenomena from space on a global scale. Rather, the

envisioned solution would bring together a number of differ-

ent satellite Earth observations and ground measurements in

an integrated ‘sensor web’ to provide the information neces-

sary for advance warning of tectonic events.

Using TIR measurements to detect earthquakes

Thermal infrared (TIR) measurements are a possible method

of detecting earthquakes in advance from space. The increased

number of satellite measurements (including a number of

National Aeronautics and Space Administration (NASA) and

National Oceanic and Atmospheric Administration (NOAA)

satellites) of surface temperature at different infrared wave-

lengths in recent years has opened up this ‘thermal brand’ of

detection techniques. TIR surveys gave an indication of the

appearance – from days to weeks before the event – of anom-

alous space-time TIR transients that have been associated with

the location (epicentre and local tectonic structures) and time