24.07.2014 Views

Rotational Raman scattering in the Earth's atmosphere ... - SRON

Rotational Raman scattering in the Earth's atmosphere ... - SRON

Rotational Raman scattering in the Earth's atmosphere ... - SRON

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

72 Chapter 3<br />

where λ is <strong>the</strong> wavelength of <strong>the</strong> <strong>in</strong>com<strong>in</strong>g light and λ ′ represents <strong>the</strong> correspond<strong>in</strong>g wavelength of<br />

<strong>the</strong> scattered light. f(T,J) is <strong>the</strong> fractional population <strong>in</strong> <strong>the</strong> <strong>in</strong>itial state and can be approximated by<br />

[Jo<strong>in</strong>er et al., 1995]<br />

f(T,J) = 1 [<br />

g(J) [2J+1] exp<br />

N f<br />

− E(J)<br />

kT<br />

]<br />

, (3.74)<br />

where g is <strong>the</strong> nuclear sp<strong>in</strong> statistical weight factor, k is <strong>the</strong> Boltzmann’s constant, T is <strong>the</strong> temperature,<br />

and E(J) = hcBJ(J +1) is <strong>the</strong> rotational energy (h is Planck’s constant, c is <strong>the</strong> speed of<br />

light, and B is <strong>the</strong> rotational constant). The coefficient N f can be determ<strong>in</strong>ed from <strong>the</strong> normalization<br />

condition<br />

∑<br />

f(T,J) = 1. (3.75)<br />

J<br />

b(J → J ′ ) represents <strong>the</strong> Placzek-Teller coefficient for <strong>the</strong> transition J → J ′ . For simple l<strong>in</strong>ear<br />

molecules <strong>the</strong>se coefficients are given by<br />

b(J →J+2) = 3(J+1)(J+2)<br />

2(2J+1)(2J+3) , (3.76)<br />

for Stokes l<strong>in</strong>es and<br />

b(J →J −2) =<br />

3J(J −1)<br />

2(2J −1)(2J+3) , (3.77)<br />

for anti-Stokes l<strong>in</strong>es. The change <strong>in</strong> wavelength ∆λ=λ ′ −λ due to rotational <strong>Raman</strong> <strong>scatter<strong>in</strong>g</strong> can be<br />

calculated easily from <strong>the</strong> energy difference between <strong>the</strong> <strong>in</strong>itial and f<strong>in</strong>al state, ∆E = E(J ′ )−E(J).<br />

The <strong>scatter<strong>in</strong>g</strong> phase matrix P, def<strong>in</strong>ed with respect to <strong>the</strong> plane of <strong>scatter<strong>in</strong>g</strong>, has <strong>the</strong> same structure<br />

Table 3.1: Parameters A, B, C, and normalization constant N of <strong>scatter<strong>in</strong>g</strong> phase function (3.78) for<br />

Rayleigh, Cabannes and rotational <strong>Raman</strong> <strong>scatter<strong>in</strong>g</strong>.<br />

A B C N<br />

Rayleigh 3(45+ε) 30(9−ε) 36ε (180+40ε)<br />

Cabannes 3(180+ε) 30(36−ε) 36ε 40(18+ε)<br />

rot. <strong>Raman</strong> 3 −30 36 40<br />

for Rayleigh, Cabannes and rotational <strong>Raman</strong> <strong>scatter<strong>in</strong>g</strong>, viz.<br />

⎛<br />

⎞<br />

A(1+cos 2 Θ) + C −A(1−cos 2 Θ) 0 0<br />

P(Θ) =<br />

1 −A(1−cos 2 Θ) A(1+cos 2 Θ) 0 0<br />

⎜<br />

⎟<br />

N ⎝ 0 0 2A cos Θ 0 ⎠<br />

0 0 0 B cos Θ<br />

(3.78)

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!