wiki:u/adebrech/PlanetSims

Version 11 (modified by adebrech, 9 years ago) ( diff )

Mass Loss Rates

Value Mdot (g/s)
Aniso 6.1x109
lambda=5 5.6x1011
lambda=15 3x107
Mp = MJ 9.9x109
Mp = 0.5MJ 1.5x109
Mp = 0.25MJ > 1.67x109 (not well resolved) 3.7x108
Tp = 5x104 K Negative throughout (not resolved) 3.4x108
Tp = 5x103 K 1.4x109
Tamb = 3 K 4.5x109
Tamb = 25 K 4.5x109
Tamb = 50 K 4.5x109

Compare Aniso with net value from planet paper:

I find it suspicious that all three values for Tamb are the same (to 6 decimal places), but less than Tamb = 100 K (for Aniso).

Calculations with VisIt

Focusing on mass loss rate at the moment. Testing on Run5 (Aniso) so I can compare to Jonathan's paper. Created all the expressions I believe are required (see attachment:expressions.xml), with the mass flux defined as density times radial velocity2 (properly scaled). When I perform a weighted query on the mass flux over a spherical isosurface (defined in terms of distance from origin), however, I get a value about 3 magnitudes too large (1.14*1013). Also tried calculating flux with the mesh flux operator, but velocity times density doesn't seem to be the appropriate quantity. It does make nice pictures, though:

First is at r = 0.1, second is at r = 0.05 (Plot is pseudocolor of operator/Flux/Mesh, SphericalSlice centered at origin). Note that values are negative because the normals of the spherical slice point inwards.

Fixed scaling of expressions. Now summing over an isosurface gives the expected values; it would be easy enough to use min/max to calculate inward and outward flux separately, if desired.

Parameter Space for Planet Sims

Successfully recreated Run 5 (Aniso) from Jonathan's simulations, after meeting and fixing some of the code. Most runs are taking 5-8 hours with 1 level AMR (did check, optimization flag is set to level 3 in makefile). Used momenta to compare relative strength of outflows, made density movies.

In comparison to Aniso, changing lambda performs as expected - with lambda = 5, there is a marked increase in the strength of the wind, and with lambda = 15 there is very little to no wind (size of planet changes as well).

Increased size of the domain to capture sonic radius; bigger outflow, as expected (note that the density scale is different than the rest of the plots, in order to discern the planet):

Run longer (1.5x almost reaches steady state):

Changing the mass of the planet wasn't quite as clear. Overall, it doesn't appear to have affected the wind very much. Again, the planet size is related to the mass (since density is also specified). It is hard to see a wind on the smallest (.25 MJ) planet, but momentum plots appear to support its existence.

Decreased size of domain to better resolve near planet; weaker outflow:

Changing the temperature of the planet also affects the radius (through lambda). A 5x hotter planet appears to have a stronger outflow at the end, although it takes a significant amount of time to gain significant size (perhaps should run this simulation for longer). A planet with .5 T doesn't appear to be very different, qualitatively, from the original run.

Decreased size of domain to resolve planet; actually has weaker outflow (mass loss rate 1 order of magnitude smaller):

Finally (for the moment), changed ambient temperature to lower values. Would also like to explore higher values to see if they have much effect. The 50K, 25K, and 3K runs all have similar maximum momenta, approx. 2x the Aniso run, which seems to suggest a stronger outflow, but the density plots look nearly identical.

What parameters are best to plot?

Sonic Radii

For a few of the parameters, I also plotted contours on the sonic radius. Respectively, Run5 (original), lambda=5 with larger domain, Mp=.25MJ with smaller domain, Tp=5x104 K with smaller domain, and Tamb=50K.

Added lambda=5, run longer (1.5x):

Experimenting with AstroBEAR

Ran OutflowWind simulations with varying parameters. Thickness and wind velocity don't appear to have any effect - perhaps an effect of rescaling? Velocity speeds up the simulation as a whole. The rest are as follows:

Reference (default parameters, 100 frames, 2 units computational time (conversion?)):

Radius increases size of wind, but otherwise has little effect, at least at small values (expected).

Radius of 2:

Increasing density appears to decrease relative density at the center of the front of the outflow (slightly counter-intuitive, but makes sense).

Density of 5: Density of 10: Density of 20:

Increasing temp. creates a more diffuse cloud at the end (predictably). Temp = 0 appears to cause buggy behavior.

Temp = 0: Temp = 1: Temp = 5:

Also tried a run at 1000 frames, but definitely overkill for time.

Attachments (34)

Note: See TracWiki for help on using the wiki.