• March 15: Studying the ray tracing setup and trying to be able to find a model for atmosphere
• April 7: Rendering mirages and effects in static images
• April 30: Rendering heating effects and mirages in dynamic environment

Work done till now (03/16/2009):

•  Read through some publications to understand the phenomena of mirage and shimmering. Although there are papers that talk of formation of mirages, I have been unable to find a very concrete explanation of the shimmering process, or some model modeling the light path in the heated layer or movement of heated molecules. So, efforts are on  in that direction.
• Got the ray tracing code from Christian, and am playing around with it to see how and what is done in the code.
• Have been playing through with the Blender to model some possible scenario.
  

My general observations and questions:

• Mirages are generated only at certain angles of incidence. So a mirage that is visible to me might not be visible to another camera or person at different positions. Moreover, mirages are a result of the heat shimmering process which is also visible only at certain angles of incidence to the surface. So, for example as we were driving in the car last Sunday which was quite a sunny day, I could notice some shimmering on some parts of the road only at certain positions in the drive and mirages were generated too almost at the same positions. So a continuous change in the angle of incidence does play a role in the generation of mirages on the object.

• The paper on real time visualization of heat shimmering does not take into account this effect because of change in angle of incidence and create a real time sequence, moreover the results shown seem to be more of fluid like effect. At least my visual comparison to heat shimmering and the effect that their videos show does not convince me somehow of the gaseous behavior

• Also, nature of the object is necessary to visualize the shimmering and the mirages. Like mirages are only generated on asphalt road, deserts and oceans... why? Does it also depend on the surface area? In these cases, the propagation of heat is linear ie the heated surface can be taken as a rectangular object, while in cases of a heat toaster, heat from gas cars and buses or a fire the heat shimmering would result in an irregular boundary of the heated area

• A fire seems to be an interesting case, as it would act as the light source apart from being the heat source too... Moreover, the kind of temperature variation would also be interesting to be noted.

• These all effects are very much related to the theories applied to the twinkling of stars as is noted by the atmospheric Scintillation note above and there is research to remove such effects in the remote sensing images recorded, while here we try to recreate such effects that too in the lower layers of atmosphere.

• Generation of mirages and the effects is said to be defined by geometric optics(Trout, etc) and physical optics(Raman, Ray Tracing in non-constant media).

• Non-linear ray tracing is interesting in the sense of its stepwise approach to model a dynamic system such as the system of molecules above the heated surface.

• I noted mirages (very small though) on the road while I was returning from job yesterday (3/12). I did not understand why these are happening at all? Maybe the road was still hot and there was very little variation present in the temperature too.. But, the day was cold throughout and it was cloudy too and I was returning at around 7 in the evening. That was the first time I noticed such an effect during evening.

• Height of shimmering layer, rather area of the shimmering. Source, refractive index variation and before that temperature variations. Why only temperature variation? Refractive indices change with variation in density – density is dependent on temperature and pressure – and in a heated region the pressure almost remains constant (According to Thyagarajan explanation of mirages) – and hence temperature variation being the source of shimmering effects and consequently the mirages.

• Incorporating the direction of temperature gradient for proper generation of mirages, ie below or above the object

• Distributed ray tracing can play a part in the sense of being able to see the variations in a group of rays, rather than single rays... because as observed (my observations.. might not be accurate) it seems that the particles move in bunches.. there seems to be some sort of regular pattern in their movement. I must record the effect on my camera in different environments, just for an observation.

• Well, it is also important to understand how much does the refraction index change in the atmosphere with temperature?

• What I might need is a global temperature condition parameter.. (in ideal conditions..) maybe the temperature through the day can be controlled and the weather of the day is randomised, etc. But, the point of the whole thing is that the object which is supposed to heat itself, would heat at a particular time of day. And the air above it must be heated for that. So, the temperature texture, or surface temperature should be noted maybe at discrete times to estimate the possibility of generation of either mirages or shimmering. Once that is done, one can start the randomised ray tracing of the rays only when particular angles of incidence are achieved. This would also affect the generation of the different type of mirages. (which was not talked of in the Visualization paper I guess, but there were hints that such images can be formed, since basically they implement the model of heat transfer, but there were no results for the same).
• Random refraction and systematic refraction are the two phenomena according to Condon Report.
  Should these be applied? Can be tried actually but will have to see in any case.

What might I need:

• Temperature gradient
• Model of the effects of the temperature gradient (why shimmering is happening anyway?)
• A scene of atmosphere - where do I make the model? and how do I load that into ray tracer...