Archive for the ‘Geostatistics’ tag
Penn State Public Broadcasting released a 5min video giving an introduction of how computers changed the way we deal with spatial information, formerly known as “maps”. Their tagline “the location of anything is becoming everything” is a bit… scary… but I guess we’re on our way.
(via Google Earth Blog)
I’m sitting in Moscone West. As Apple aficionado, this is right where Apple’s Worldwide Developer Conference used to be!
Day 1 of 5 is almost over. I’ve been to a few talks, I’ve been to a few posters. I’ve never been to a conference of a comparable size, so I am a little overwhelmed. The online calendar was really helpful. Scientifically, what was interesting?
- I saw C. Meirovitz’s poster on modelling alluvial fans with TPROGS. He’s trying to use borehole-log data to model hydro-facies in space. Channels that deposited gravel up to bolder size form preferential flow paths, and the simulation of those continuous features is not a trivial task; however, it is important because those are the pathways critical for non-linear processes such as solute transport. Very much related to my own work!
- I have never spent much time learning about fractals, but the concept of “self similarity” popped up at least twice today. Once in F. Schwartz’s talk on modelling prairie potholes, where he uses tho model the size of potholes in simulation by using two parameters (length and water-hight, if I remember correctly), and once in S. Neuman’s talk.
- Using insane amounts of data seems to become quite standard.
- Why are people so crazy about linear regression? And why do people who call themselves scientists refer to what they’re doing as linear regression, if both of their axes are log-scale? Similarly, I’ve seen at least two people who plot some measurement against some measure of scale. For small scales there are a lot of measurements, for large scales there are few measurements. Their argumentation is that measurements on the larger scale result in an average measurement. I am not 100% sure on this. It might also be that there simply are not many measurements on a larger scale, temporal or spatial.
- I saw a map of accumulated daily precipitation intensities, that is accumulated at a given location over a year, then spatially interpolated. That seems like a neat idea. Probably also for temperatures.
- This is nothing new, but well presented presentations are just so much better.
Now I’m off to the “frontiers of geophysics” lecture!
Folks, I am getting ready for the AGU Fall conference, which starts tomorrow (Sunday) evening in San Francisco. I’ve never been to a scientific conference of that size, so I am very curious. I was browsing through the online database of talks and presentations, and it’s just mind-boggling.
My poster’s title is “Effects of Non-Gaussian Spatial Dependence of Hydraulic Conductivity on Hydrodynamic Macrodispersion”. Its ID is H43F-1093 and it will be presented on Thursday, Dec 17, starting at 1:40 PM in the Poster Hall (Moscone South). Come and drop by! It’s nice to meet blog-readers in person!
The key method I am using are spatial copulas. I had set out to explain what this is on this blog here and here, but I have never gotten around to get really to what copulas are. I promise I will continue to write here about copulas as soon as I am back.
There is going to be a meeting of geo-bloggers on Wednesday! Some cool resources do exist already:
- a list online of who is registered
- a shared google.doc from @Boreholegroup of geo-bloggers who present their scientific work at AGU.
- a twitter list by @Allochthonous
I am looking forward to the show!
Starting tomorrow, Monday, I will be attending the “Geostatistics 2008” conference. I am already in Santiago, the first time south of the equator. It’s December, and I had breakfast today on the patio. Wow. Yesterday I toured the city and had one of the best fish soups of my life. Today I went up to Cajon del Maipo, touched the Andes for the first time, enjoyed the fantastic scenery.
Now I am looking forward to a great conference, lots of new experiences and ideas. I will give a talk on tuesday afternoon around 4pm. The show is at the Sheraton Hotel in Santiago, if you’re in the area…
Maybe I can even attract some visitors to this site!
On Friday I’m leaving for Southampton, UK. From Monday to Wednesday there’s the conference “geoEnv 2008“. I’m quite excited: There are many exciting speakers from the geostatistics community who I have never heard live let alone met. The conference seems to be organized really well. And I am going to give a talk on tuesday afternoon, entitled “Applications of Copulas in Geostatistics”. I would be really surprised if anybody who reads this here would actually be in Southampton, but if this should be the case, I sure would be happy to meet you, and buy you a beer!
It turns out that there is a dinner organized in the evening after my talk on restored classic british sailing ship.
Due to this conference, this blog here will likely be updated less frequently than usually. I hope you guys can understand!