CL_mod_atmRefract

Description

• Effect of refraction on elevation due to Earth atmosphere (for visible light) as seen from Earth surface.

  The function computes the apparent elevation from the true elevation ("true" meaning: "that would be observed without atmosphere"). Because of refraction, a body actually below the horizon can still be visible.

  Two models are available: Benett or Saemudsson.

  Effect of temperature and pressure can (roughly) be taken into account.

• Notes:

  - The refraction models come from Meeus (see Bibliography below).

  - The refraction corrections are slightly adjusted so that there are exactly 0 for an elevation of 90 deg and unchanged for a true elevation of 0 deg.

  - Saemundson's formula directly gives the correction to be applied to the true elevation. But Bennett's formula is to be applied to the apparent elevation. The formula is inversed so that the correction can be computed from the true elevation. The effect of temperature and pressure is applied to both models, allthough Meeus mentions it for Saemundsson's model only.

  - The refraction corrections are computable for true elevations above -1 deg. The correction is undefined (%nan) below this value.

  The 2 models are consistent to within less than 7 arcseconds between apparent elevations 0 and 90 degrees. Benett model is supposed to be slightly more accurate (accuracy assumed better than 4 arcseconds between 0 and 90 deg (apparent elevation).

Parameters

true_elev:

True elevation [rad]. (1xN or 1x1)

temp:

(optional) Atmospheric temperature [K]. Default is [] which means 283.15 K (10 °C). (1xN or 1x1)

pres:

(optional) Atmospheric pressure [Pa]. Default is [] which means 1.01e5 Pa (1010 mbar). (1xN or 1x1)

mod:

(string, optional) Model used: "ben" = Benett, "sae" = Saemudsson. Default is "ben". (1x1)

app_elev:

Apparent elevation [rad]. (1xN)

Authors

• CNES - DCT/SB

Bibliography

• 1) Astronomical algorithms, Jean Meeus, 2nd Edition, p 106

Examples

elev = (0 : 90) * (%pi/180); r = CL_mod_atmRefract(elev) - elev; scf(); plot(elev * (180/%pi), r * (180/%pi), "b");