A Direct Measurement of Atmospheric Dispersion in N-band Spectra: Implications for Mid-IR Systems on ELTs

Adaptive optics will almost completely remove the effects of atmospheric turbulence at 10 microns on the Extremely Large Telescope (ELT) generation of telescopes. In this paper, we observationally confirm that the next most important limitation to image quality is atmospheric dispersion, rather than telescope diffraction. By using the 6.5 meter MMT with its unique mid-IR adaptive optics system, we measure atmospheric dispersion in the N-band with the newly commissioned spectroscopic mode on MIRAC4-BLINC. Our results indicate that atmospheric dispersion is generally linear in the N-band, although there is some residual curvature. We compare our measurements to theory, and make predictions for ELT Strehls and image FHWM with and without an atmospheric dispersion corrector (ADC). We find that for many mid-IR applications, an ADC will be necessary on ELTs.

💡 Research Summary

**
This paper presents the first direct, ground‑based measurement of atmospheric dispersion in the mid‑infrared N‑band (8.26–11.27 µm) using the 6.5‑m MMT telescope equipped with a dedicated mid‑IR adaptive optics (AO) system and the newly commissioned spectroscopic mode of MIRAC4‑BLINC. By observing bright visible standards (α Her and γ Aql) at several airmasses (1.3, 1.53, 1.82, 2.53) and using a 0.73″ slit with a low‑resolution (R≈100) grism, the authors extracted the wavelength‑dependent centroid shift of the point‑spread function (PSF), which directly yields the atmospheric dispersion curve after subtracting the intrinsic grism curvature measured at 1.05 airmass (assumed negligible).

The measured dispersion is essentially linear across the N‑band, with slopes ranging from ~10 mas µm⁻¹ at low airmass to ~34 mas µm⁻¹ at the highest airmass, confirming that the blue side of the band is refracted more strongly than the red side. A modest non‑linear curvature is also detected, especially shortward of ~9.5 µm, consistent with the S‑shaped refractivity caused by molecular resonances of CO₂ and H₂O.

The authors compare their results with the theoretical refractivity models of Mathar (2004, 2007), which compute the index of refraction from HITRAN line data. After applying the same 1.05 airmass subtraction, the models reproduce the observed linear trends very well for three of the four airmass sets; the 1.53 airmass case shows a systematic offset, possibly due to an unmonitored moisture layer or a minor instrumental tilt. Overall, the agreement validates the Mathar models for predicting mid‑IR atmospheric dispersion, though real‑time meteorological data may be required for precise corrections.

To assess the impact on future Extremely Large Telescopes (ELTs) of 24 m, 30 m, and 42 m apertures, the authors convolve the measured dispersion curves with diffraction‑limited PSFs (including a 20 % central obscuration) and generate synthetic images for three observing scenarios: (1) broadband N‑band imaging (8–14 µm), (2) narrow‑band imaging (10 % filter centered at 10.5 µm), and (3) slit spectroscopy. Without any atmospheric dispersion corrector (ADC), Strehl ratios drop dramatically (to <30 % at 1.5 airmass for the 42 m case) and full‑width‑half‑maximum (FWHM) values increase by tens of milliarcseconds, producing severe elongation along the altitude axis. Introducing a simple linear ADC that removes the bulk linear trend restores Strehl ratios to >80 % and brings FWHM close to the diffraction limit for airmasses up to ~2.5. The residual non‑linear curvature after linear correction causes only minor degradation, even for the largest apertures, indicating that a higher‑order “non‑linear” ADC is not strictly required for most mid‑IR applications.

The study further shows that narrow‑band imaging still suffers measurable degradation without an ADC, especially for the 42 m telescope, while the linear ADC fully mitigates the effect. For slit spectroscopy, chromatic slit losses become significant without correction, reducing throughput and spectral fidelity.

In summary, the paper demonstrates experimentally that atmospheric dispersion, not diffraction, will be the dominant limitation to mid‑IR image quality on ELTs once AO removes turbulence. A linear atmospheric dispersion corrector is essential for broadband imaging and beneficial for narrow‑band and spectroscopic modes. The measured dispersion agrees well with existing theoretical models, providing a solid foundation for ADC design. Future work should focus on refining real‑time atmospheric monitoring and exploring modest higher‑order corrections for observations at very high airmass.