New Feature : Two-Way Coupling

[joeschoonover]

Two-way coupling between WAM and IPE is the goal for FY 2018. There are three main tasks, from which subtasks can be broken out :

• Add local arrays to WAM for accepting IPE arrays.
• Add routines with IPE to calculate terms needed by WAM.
• Update Mediator

[joeschoonover]

Issue NOAA-SWPC/WAM-IPE#54 is a subtask for this issue

[timfullerrowell]

Notes for 2-way coupling below.
Includes questions on rotation of vectors (?).
Tim needs to check equations and signs work both ends of the flux tube.

This is what we currently do:

Neutral winds from WAM in geographic eastward, northward and vertical directions U(1), U(2), U(3) are passed from WAM to IPE (ms-1)

The mediator interpolates the neutral wind values to the IPE grid, keeping the directions still geographic

In IPE on the IPE grid, we rotate these vectors to the magnetic east, magnetic equatorward/down, and parallel to magnetic field, to generate neutral winds on the IPE grid in the magnetic directions: Un(1), Un(2), Un(3) (m s-1)
Do we have the code in IPE that does this?

We need to calculate the ion drag momentum tendencies and Joule heating in IPE to pass back to WAM. To do this we need to:

a) Calculate the ion velocities in the IPE grid in the magnetic east and equatorward/ down directions (see below) Vi(1), Vi(2) (m s-1)

b) Use the ion velocity parallel to B from FLIP, which is the weighted average of the O+ and H+ field aligned velocities (parallel to B), weighted by the mass mixing ratio of the ions – Vi(3) (m s-1)

Required to pass to Subroutine Calculate_Tendencies_for_WAM:

Global real parameters on IPE grid from IPE lower boundary to ~1000 km

Ion-neutral collision frequency: vin (s-1)
Total plasma density: ne (m-3)
Mean ion mass: mi (amu)
Parallel ion velocity (see above): Vi(3) (m s-1)
Neutral wind in magnetic directions (see above) Un(1), Un(2), Un(3) (m s-1)
Neutral mass density rho (k gm-3))
Height of the grid points above Earth’s surface (to only calculate tendencies to ~1000 km) (m)

Point of discussion:
Do we use the electric field and total magnetic field from foot point value or allow for height dependence?

Electric field in magnetic east E(1) and south/down E(2) (V m-1)
Total magnetic field B (Tesla)

Calculate ions velocities perpendicular ion velocities Vi(1) and Vi(2):

Constants
Elementary charge:

q (C)

Ion gyro frequency:

wi = q * B / mi

Ratio of ion-neutral collision frequency and ion gyro frequency :

k = vin / wi

Vi(1) = E(2) / (B * (1. + k2))
+ k * Un(2) / (1. + k2)
+ k * E(1) / (B * (1. + k2))
+ k2 * Un(1) / (1. + k**2)

Vi(2) = - E(1) / (B * (1. + k2))
- k * Un(1) / (1. + k2)
+ k * E(2) / (B * (1. + k2))
+ k2 * Un(2) / (1. + k**2)

We will then need to rotate the ion velocities Vi(1), Vi(2), and Vi(3) from magnetic to geographic directions east and north to V(1) and V(2) ( I think this is one of the AVAILABLE apex transformations?)

Eastward momentum tendency EMT (m s-2):

EMT = mi * ne * vin * (U(1) – V(1)) / rho

Northward momentum tendency NMT (m s-2):

NMT = mi * ne * vin * (U(2) – V(2)) / rho

Joule heating JHT (J kg-1 s-1)

JHT = ne * vin * ((U(1) – V(1))**2 + (U(2) – V(2))**2) / rho

Subroutine returns these three tendencies.

[timfullerrowell]

Naomi, (Tzu-Wei, Joe)

As we just discussed, to be more specific, sometime next week, could I please get feedback on these questions regarding the 2-way coupling:

1. Should we include height dependence of E and B along flux tube in the 100 to ~1000 km height range.

2. What are the positive directions of the following parameters at each end of the flux tubes in the north and south hemisphere:
   a) Neutral wind parallel to B, and perpendicular to B in the magnetic equatorward/northward/ up/down direction
   b) Direction of B
   c) Direction of E perpendicular in the magnetic equatorward/northward/up/down direction
   d) Direction of parallel and perpendicular ion velocities

[twfang]

@IonospherePlasmasphereElectrodynamics
Naomi, could you please take a look at Tim's questions and see if we can get everything sort out int the coming few days? We were not included in the issue tracer earlier and have missed the messages that Tim sent late week.

[twfang]

@timfullerrowell to answer your questions, for the Apex coordinate, positive B is Northward. Eastward electric field and eastward ion velocity are positive. Downward/equatorward electric field and ion velocity are positive. However, all these need to be verified in the IPE refactor code.

[naomimaruyama]

Hi @twfang et al, I am sorry for the delay in responding. I was totally swamped by the proposal deadline (today). I already answered to @timfullerrowell <https://github.com/timfullerrowell> 's questions to him on tuesday (9/4), but have not had a chance to update on github until today.

…

On Thu, Sep 6, 2018 at 1:55 PM, twfang ***@***.***> wrote: @timfullerrowell <https://github.com/timfullerrowell> to answer your questions, for the Apex coordinate, positive B is Northward. Eastward electric field and eastward ion velocity are positive. Downward/equatorward electric field and ion velocity are positive. However, all these need to be verified in the IPE refactor code. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <https://github.com/SWPC-IPE/WAM-IPE/issues/45#issuecomment-419221022>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ANGc4sKKr28G4yi77HSfpWl4ZdRgBqIyks5uYX3CgaJpZM4QDWpN> .

-- Naomi Maruyama CIRES, Univ. of Colorado Boulder NOAA Space Weather Prediction Center phone: +1 303.497.4857 <%28303%29%20497-4857> email: naomi.maruyama@noaa.gov

[twfang]

For completed list of fields that need to be extracted from IPE for WAM, please refer to the closed issue NOAA-SWPC/WAM-IPE#118.

[twfang]

Need to provide these parameters into WAM, see #5 in GSMWAM-IPE

[twfang]

Ion-neutral collision frequency vin (s-1): ion_neutral_collisionfrequency in IPE_Plasma_Class.F90
Total plasma density: ne (m-3): plasma % electron_density
Mean ion mass: mi (amu): ion_mass_amu in IPE_Plasma_Class.F90
Parallel ion velocity (see above) Vi(3) (m s-1): plasma % ion_velocities(1,i,lp,mp)
Neutral wind in magnetic directions (see above) Un(1), Un(2), Un(3) (m s-1): Ue(1) and Ue(2) are calculated in subroutine Calculate_Field_Line_Integrals
Neutral mass density rho (k gm-3)): the same as the calculation in SWIO
Height of the grid points above Earth’s surface (to only calculate tendencies to ~1000 km) (m)

Constants, Elementary charge:

q (C) Ion gyro frequency wi = q * B / mi : similar to the calculation in omega_op in IPE_Plasma_Class.F90

Ratio of ion-neutral collision frequency and ion gyro frequency: k = vin / wi

Vi(1) and Vi(2) are in magnetic coordinates:

Vi(1) = E(2) / (B * (1. + k^2)) + k * Un(2) / (1. + k^2) + k * E(1) / (B * (1. + k^2)) + k^2 * Un(1) / (1. + k^2)

Vi(2) = - E(1) / (B * (1. + k^2)) – k * Un(1) / (1. +k^2) + k * E(2) / (B * (1. + k^2)) + k^2 * Un(2) / (1. + k^2)

Electric field in magnetic east E(1) and south/down E(2) (V m-1): eldyn % electric_field(2,lp,mp)
Total magnetic field B (Tesla): grid % magnetic_field_strength(i,lp,mp)

We will then need to rotate the ion velocities Vi(1), Vi(2), and Vi(3) from magnetic to geographic directions east and north to V(1) and V(2) ( I think this is one of the AVAILABLE apex transformations?)

Eastward momentum tendency EMT (m s-2):

EMT = mi * ne * vin * (U(1) – V(1)) / rho

Northward momentum tendency NMT (m s-2):

NMT = mi * ne * vin * (U(2) – V(2)) / rho

Joule heating JHT (J kg-1 s-1)

JHT = ne * vin * ((U(1) – V(1))**2 + (U(2) – V(2))**2) / rho

Subroutine returns these three tendencies.

[twfang]

@rmontuoro and @timfullerrowell Could you please take a look at my comment and see if they make sense to you? I have all IPE parameters written in bold. Thank you!

[timfullerrowell]

Tzu-Wei, Are Ue(1) and Ue(2) the neutral wind in magnetic directions the same as Un(1), Un(2)? Un(3) (or Ue(3) if it is the same) should be used directly in IPE; the neutral wind flow parallel to B. The units of rho should probably be kg m-3, but I think that was just a misprint in the spaces. There is a bit of the text missing for the final calculation of tendencies. Tim

…

On Wed, Sep 16, 2020 at 4:18 PM twfang ***@***.***> wrote: Ion-neutral collision frequency: vin (s-1):*ion_neutral_collisionfrequency in IPE_Plasma_Class.F90* Total plasma density: ne (m-3): *plasma % electron_density* Mean ion mass: mi (amu): *ion_mass_amu in IPE_Plasma_Class.F90* Parallel ion velocity (see above) Vi(3) (m s-1): *plasma % ion_velocities(1,i,lp,mp)* Neutral wind in magnetic directions (see above) Un(1), Un(2), Un(3) (m s-1): *Ue(1) and Ue(2) are calculated in subroutine Calculate_Field_Line_Integrals* Neutral mass density rho (k gm-3)): *the same as the calculation in SWIO* Height of the grid points above Earth’s surface (to only calculate tendencies to ~1000 km) (m) Constants, Elementary charge: q (C) Ion gyro frequency: wi = q * B / mi *similar to the calculation in omega_op in IPE_Plasma_Class.F90* Ratio of ion-neutral collision frequency and ion gyro frequency: k = vin / wi Vi(1) and Vi(2) are in magnetic coordinates: Vi(1) = E(2) / (B * (1. + k*2)) + k * Un(2) / (1. + k*2) + k * E(1) / (B * (1. + k*2)) + k*2 * Un(1) / (1. + k**2) Vi(2) = - E(1) / (B * (1. + k*2)) – k * Un(1) / (1. +k*2) + k * E(2) / (B * (1. + k*2)) + k*2 * Un(2) / (1. + k**2) Electric field in magnetic east E(1) and south/down E(2) (V m-1): *eldyn % electric_field(2,lp,mp)* Total magnetic field B (Tesla): *grid % magnetic_field_strength(i,lp,mp)* We will then need to rotate the ion velocities Vi(1), Vi(2), and Vi(3) from magnetic to geographic directions east and north to V(1) and V(2) ( I think this is one of the AVAILABLE apex transformations?) Once we have Vi(1), Vi(2), and Vi(3), the momentum tendencies can be calculated in the geographic grid. — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#12 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AH5BFOFKIFRX7CUB3RLZXLDSGE2SLANCNFSM4JU77SXQ> .

[twfang]

I think so, but I will take a look at these winds again. I have added the tendencies back to the comment.