The transition frequencies are from astronomical observations by
(1) C. Cabezas, E. Roueff, B. Tercero, M. Agúndez, N. Marcelino, P. de Vicente, and J. Cernicharo, 1982, J. Mol. Spectrosc. 92, 497.
The line at 47696.5 MHz appears to be slightly blended and displayed somewhat larger residuals in the fits; its uncertainty was increased slightly. Estimates of two sextic distortion parameters do not affect the fit much, but may improve the calculations somewhat at higher J or K_a.
The calculation is sufficiently accurate for observations in cold sources, possibly not in luke-warm souces. The calculation was truncated at 200 GHz and at K_a = 5. Transitions with K_a > 2 or above 100 GHz should be viewed with caution.
Quadrupole splitting due to the ¹⁴N nucleus may be resolvable for low J. A separate hyperfine calculation is provided for K_a ≤ 1 and J up to 5. The spin-multiplicity was considered in the calculation of the partition function.
The dipole moment of the main isotopic species is from
(2) M. Bester, K. M. T. Yamada, G. Winnewisser, W. Joentgen, H.-J. Altenbach, and E. Vogel, 1984, Astron. Astrophys. 137, L20.