Science Facts - HITRAN Data Overview

HITRAN - high-resolution transmission molecular absorption database

The HITRAN database is "a compilation of spectroscopic parameters that a variety of computer codes use to predict and simulate the transmission and emission of light in the atmosphere". Originally started by the Air Force Cambridge Research Laboratories (AFCRL) in the late 1960's, it is now being developed at the Atomic and Molecular Physics Division, Harvard-Smithsonian Center for Astrophysics.

The data is available for free, but getting enough information to actually use the data is extremely difficult.

One of the first problems - wavenumber

There are 2 different data formats - 100 byte (used before 2004) and a 160 byte format (used starting in 2004).

It is never clear if it is the natural log or the base-10 log that should be used.

Data Format

The HITRAN documents are very clear about the data formats.

 21  442.005539 3.705D-27 3.917E-05.0674.0802 2685.58910.780.000000 10  8             P 39 455 2 2-1

This is an analysis of the first data line. It is 100 characters long.

12 3 456789*12345 6789*12345 6789*12345 6789* 12345 6789*12345 6789 *1234567 89* 123 456789*12 3456789*1 234 56789*
 2 1   442.005539  3.705D-27  3.917E-05 .0674 .0802  2685.5891 0.78 0.000000  10   8               P 39  455  2 2-1
 2 1      12           10        10       5     5        10      4      8     3   3      9         9      3     6

Isotopologues

IR spectra exist because certain molecules absorb energy because of a resonance between the frequency of the radiation and the natural vibrational frequencies of different parts of the molecules. Think of a weight on a spring. The frequency of oscillation (ω) is related to the square root of the spring constant (k) divided by the mass (m).

ω = sqrt(k/m)

In molecules, the spring constants are related to the bond strengths which, in turn, are related to the number of protons in the nucleus while the mass is related to the sum of the number of protons and the number of neutrons in the nucleus.

Each specific type of element has the same number of protons, but the number of neutrons is not fixed. When the number of neutrons is too high or too low, the nucleus is unstable and will spontaneously change. When the change produces a change in nuclear charge, then that isotope is referred to as radioactive. For instance, all carbon atoms have 6 protons. The normal atoms (>98%) have 6 neutrons. There are 3 common isotopes - carbon-12, carbon-13, and carbon-14 (which is radioactive). As a result, the IR spectra of two molecules with different isotopes are not identical.

The HITRAN database contains CO2 spectra for 8 different isotopologues shown below. 626 is interpreted as 16-O 12-C 16-O.

OCO	Isotopologue Fractional Abundance
OCO	1973	1986	2004
626	0.98414	0.9842	0.98420
636	0.01105	0.0110	0.01106
628	0.00402	0.0039	0.0039471
627	0.000730	0.0008	0.000734
638	0.0000452	0.000044	0.00004434
637	0.00000820	0.000009	0.00000825
828	0.00000412	0.0000040	0.0000039573
728	-	0.000002	0.00000147

I have shown multiple years because I think it is interesting. I have no idea how they can claim 5 significant figures on something that changes every few years.

Carbon-14 (used for radiocarbon dating) is not listed because it occurs (as 646) at about 1 part per trillion.

Carbon-13 is interesting (636). This is the molecule that is used to prove that burning fossil fuels produces the additional CO2 in the atmosphere. Obviously (from these numbers), the change in the 13C/12C ratio can not be seen in the atmosphere. Apparently, the ratio is measured in trees and corals. This is because we have samples that go back several hundred years, well before we were able to measure the ratio. At least one paper provides evidence that 13C/12C changes measured in the atmosphere are due to the southern El Nino and are NOT related to burning fossil fuels.

Also, note that even though Carbon-13 occurs in about 1.1% of the atmospheric CO2, my program ignores this. The purpose of the program was to help understand the effects of CO2, and I think that including all the isotopologues actually confuses the graphs.

Author: Robert Clemenzi