Weird Behaviour of Albedo Model at Low Resolutions

[BriekArcsec]

I've noticed that, when using low-resolution CERES data in the albedo module, I get quite unrealistic results for the calculated albedo fluxes.

As an example, this scenario simulates a spacecraft with 6 coarse sun sensors (one on each face), which has nearly zero angular velocity (in the ECI frame). The simulation covers roughly one complete orbit of the S/C (in LEO), and was executed thrice: once with low-res CERES data, once with medium-res data, and once with high-res data.

The resulting plots of albedo flux ratio at the sun sensors, are the following (in order of increasing resolution):

As can be seen on these images, lowering the albedo resolution results in an additional oscillation of the albedo fluxes (as compared with the 'true' values simulated using high-res data). The period of these oscillations seems to correspond roughly to the pixel size of the albedo data, with lower pixel size (e.g. CERES 10x10) resulting in oscillations with a longer period, and higher amplitude.

I'm not sure if you are aware of this issue, and it's inherent to the manner in which Basilisk computes albedo effects? Or is this phenomenon indicative of an underlying flaw in the albedo module?

[schaubh]

I haven't used the albedo module in a while. The module does document the numerical module, and the source code should show you how this is being computed.