Simulation runs needed

[matroxel]

1: Baseline

Static ('unknown' error in):

2: Focus (z4)

3 & 4: Stigmatism/coma (z5-z8)

5: Piston/tile in each SCA

6: Change across focal plane of z4-z6

Dynamic (low freq):

7: z4(t), or z7(t), based on some power spectrum: p(nu) flat, p(nu) propto 1/(1+nu^2)

Dynamic (high freq):

8 & 9: Smearing (isotropic and anisotropic)

10: Random anisotropic smearing in 10-20% of images

[matroxel]

@matroxel responsible for baseline run.

2-4, 8, 9 are implemented at the level of only requiring input yaml file changes.

Others require some modification.

[matroxel]

Notes on detailed numbers.
Probably 1 side derivatives
2 - 0.005 waves
3,4 - 0.005 waves for all three
5 - random assignment of piston/tilt deviation per sca in range of (6nm for piston, same rms for tilt - derive from tilted uniform distribution). dz4 = a0 + a1(x-4086)/4086 + a2(y-4086)/4086.
6 - Same strategy for tilt, define slope in y direction with rms scaled to 5e-3 waves. Constant across individual SCA.
7 - z4(t), z7(t): 10^4 sec
8,9 - 0.015 rms in arcsec, 0.3 in x/y (e1,e2)
10 - same as 8,9 in random 15% of images

[CosmoSheep]

Questions:

2-4, 8, 9 are implemented at the level of only requiring input yaml file changes.

• Does this mean we only need to change the parameters of extra_aberrations in YAML file?

5 - random assignment of piston/tilt deviation per sca in range of (6nm for piston, same rms for tilt - derive from tilted uniform distribution). dz4 = a0 + a1(x-4086)/4086 + a2(y-4086)/4086.

• Does this mean we use random function to assign the parameter in extra_aberrations rather than set it directly? What's the meaning of dz4 = a0 + a1(x-4086)/4086 + a2(y-4086)/4086?

6 - Same strategy for tilt, define slope in y direction with rms scaled to 5e-3 waves. Constant across individual SCA.

• I don't understand how to define the slope in y direction and the reason to do so. Why do we relate z4-z6 aberrations to slope in y direction? It's not simply set the range of aberration equal to 5e-3, is it? Do I need to write the aberration myself in this case?

• Format of codes

[matroxel]

Re: questions - @NewTanc

Does this mean we only need to change the parameters of extra_aberrations in YAML file?

• Yes, any static changes that affect all images just need to be input via the yaml file.

Does this mean we use random function to assign the parameter in extra_aberrations rather than set it directly? What's the meaning of dz4 = a0 + a1(x-4086)/4086 + a2(y-4086)/4086?

• This should also be a constant affect across all exposures for each SCA, but each SCA should be assigned a random value of the coefficients at the start of the simulation (these could be hard coded). The change can be implemented as a deviation to z4 - this is the form that should take, where dz4 is the total affect passed as the number in extra_aberrations for z4 (the 4086 part is to normalize the coefficient - this is the number of 'active' pixels in each dimension on the SCAs).

I don't understand how to define the slope in y direction and the reason to do so. Why do we relate z4-z6 aberrations to slope in y direction? It's not simply set the range of aberration equal to 5e-3, is it? Do I need to write the aberration myself in this case?

• I meant that you can use a dzX similar to that for the tilt, i.e., dz4 = a0(y-N)/N to assign the z4-z6 extra aberration as a function of position across the focal plane (but constant across each SCA, so you'd just pass a value per SCA as extra_abberations based on this function. Here N would be some normalization based on the focal plane coordinates. @hirata10 can probably help you write something, or adapt it from my python adaption of his c code for finding SCA positions: pointing.in_sca() in simulate.py.

Format of codes

• What questions do you have about this?