Thesis Defense Advice

Go through it with someone who also know the field and try to anticipate questions they would ask

>>8261711
Just practice like a bazillion times. For my comps I recorded audio of myself giving my presentation and then listened to it all day while doing my lab work. It also helps to anticipate questions and just include the answers as part of your explanation to nip them in the bud. This will lead to the committee members asking fewer questions.

>>8261804
Interesting... so you had it mostly scripted out? I was largely planning on winging it, as I understand the material quite well, but this makes a lot more sense.

Also... How does one deal with a question to which they don't know the answer? Are you pretty much fucked if that happens? Anyone got any foolproof contingency plans?

>>8261711
Post-grad in Remote Sensing here. What area? Surface, atmosphere?

>>8261822
Surface, agricultural-related. Yourself?

>>8261849
Atmosphere.

Just have a good look at some of the very basic stuff that you haven't really looked at in a while.

What actually is a Lambertian surface, and what's the ansatz behind it?
Why exactly does the NDVI work, why is the spectral response vegetation such that it reflects green?
Do you work on retrievals? If so, why (if at all) can you neglect aerosol scattering? How does Rayleigh scatter work, what simple physical model can describe it?

These kind of things that you might use on a daily basis, but would struggle explaining quickly.

Still, have a chat with your colleagues NOW and ask them if there's anything they can tell you that would help you prepare.

>>8261857
Thanks, that is some great advice.

And those questions are really helpful. I actually just added a couple of clarifying sentences to my thesis based on them. I should really go chat up some atmosphere people, it's nice to have someone from the other end of the field's perpective.

If you have any more insights, I'm all ears

>>8261884
Not sure, that's about the best advice I can give on such a general topic. Viva's differ from one country to another, sometimes even from one uni to another.

Make sure you don't fumble the blatantly basic stuff. There will probably not be many questions on that basic level, that's why it's important to be confident with those (again, I have no idea how your viva's look like, so I could just be talking shit here).

Oh, and last but not least. Learn to say 'I don't know' when appropriate. Fumbling around and grasping for straws is embarrassing and dangerous - if you don't know the answer, say so, rather than trying to piece vague details together. They can smell your bullshit from miles and know EXACTLY that you have no clue.

>>8261915
Nice. All of that advice will come in handy, I'm sure.

Do you happen to do anything in the way of ray tracing?

It's not really related to my studies, but I've been wondering lately what kind of programming or software packages one could use to visualize light scattering and absorbtion in the context of its interaction an assemblage of objects, like simplified wheat stalks, that form a crop surface.

I know of some reflectance models that do that kind of thing, but none that are able to provide any form of visualization.

>>8261972
>Do you happen to do anything in the way of ray tracing?

I work on certain aspects of radiative transfer, which is a bit different from ray tracing. So I'm not doing any 3D or monte carlo stuff.

What you want is something like this:
http://www.geo.uzh.ch/microsite/3dveglab/
or
http://www.cesbio.ups-tlse.fr/dart/#/

>>8262000
Damn, those do appear to be exactly what I was looking for at first glance. I've heard of DART, it looks pretty slick... I'll need to dig into them this evening. Thanks.

Which radiative transfer model(s) do you work with? DART looks pretty slick.

>>8262048
Fuck, I repeated myself. Disregard that part.

>>8262048
Since I'm in atmospheres, I can't really do 3D because it takes too long, so I'm using run-of-the-mill discrete ordinate solvers for 1D RT, and then I pimp them for the sake of science and try to make them work better for our applications.

>>8262076
That sounds pretty heady. I respect the work you atmosphere guys do.

I dug into that 3D Veg Lab link, it's the coolest thing I've seen in a long time. Those guys at the ESA really have their shit together. You ever fired up their software package? I may need a dedicated video card for that one.

>>8262109
Can't say I know anything about these software packages. But going from experience with scientific software in general, it'll take you a week to figure out how to run it anyway. Not sure if it has GPU support, but you'll see once you fiddle with it..

>>8262112
Cool - that being the case, I bow to your superior research skills. I've been digging around for this type of thing for a while. Makes sense that it would exist for trees. Owe you a beer.

>>8262119
No worries, mate. I remember this only because I talked to a guy at a conference once who is working on this one (librat): https://librat.wikispaces.com/

Keep working in remote sensing, if you enjoy it. Good job prospects, international career, interesting challenges.

>>8262124
I definitely do enjoy it - I've been considering the PhD route as I have a few ideas that would allow me to expand upon my Master's research, but there is a lot of money to be made in the industry right now, tough call. Do you ever regret going the academic route?

>>8262136
Not really, I mean, one can always quite academia and go industry, it's usually harder the other way around.

Money is shit, yes, but I enjoy the freedom that comes with the job. I was in industry before I did the PhD, so I know both universes.

Whatever works out for you, anon.