Continuing my previous post about Idea for Geophysics Project?, I’d like to write down whats has been going around inside my head few weeks back. Though, the are no progress at all regarding those 2 previous projects that I wrote, but hey, maybe somebody stumbled upon my blog and then get inspired😛 HAHAHAHA.
Anyway, this project still about Borehole Seismic (VSP), because I was stuck with this stuff recently. Lets summarize it then,
- High Resolution Radon for wavefield separation
- VSP Multiple Migration
Sometimes, even in parametric wavefield decomposition was done in frequency domain. If there is any aliasing in the data, this will make issues in wavefield separation. Jovanovic summarize this in her thesis in Univ. of Houston (P SV Polarization Filtering), and publish this in SEG (2004). Though I think if the moveout between even are still close, even radon domain cannot separate it clearly, but high resolution radon might help to separate this.
I faced sometimes in Indonesia VSP data, wavefield separation of VSP data cannot be done easily. The shear wave contamination in compressional wavefield cannot easily be separated. To my understanding it can be because due to aliasing or something else, so that maybe high resolution radon can help with this.
The field of interferometric is really growing right now. Especially in utilizing VSP multiple for imaging, some are using VSP downgoing wavefield for imaging, and some are using VSP ghost multiple. One thing that is really improvement of using multiple migration is to make VSP image wider using ghost multiple just like the one that is summarized by Zhiyong Jiang et al (TLE, 2004 or Min Lou (SPG India, 2008). As we all know, VSP is lacking of illumination, low/narrow fold coverage, narrow aperture. This VSP multiple can help to overcome the situation. My expectation was just at least I have some improvement in edge reflection image of VSP migration result.
Min Lou (SPG India, 2008) summarized steps that are required to do this imaging. When I red his paper, I was thinking a short way to do it. They summarized about mirror imaging, how you mirror your velocity model above the free surface, and put your receiver in the mirror velocity model. I was thinking by mirroring this, when you doing raytracing, you can raytrace your multiple reflection in the same reflection plane as your primary reflection. Thus, without doing any numerical programming, I can do VSP multiple migration just by make mirror velocity model🙂 I haven’t tried, but is this what the paper said?
Too bad, because we are using ghost multiple, this method may not be applicable for land data, where in Indonesia VSP in onshore is more acquired than offshore. Even in the offshore, I think it is required that the water coloumn is thick enough to generate some multiple. But I think this multiple imaging is promising.
That’s all folks, hope I can do something with the Idea🙂
* fiuhhh, my English is so bad😦