RE: ODT Discussion - dates

[Kerstein, Alan]

Steve,

 

Gradient multiplication by maps, being a multiplicative random process, inherently introduces important aspects of intermittency, as shown in LEM Part 6.  ODT eddy correlations are another intermittency mechanism, but it's not yet determined whether this makes ODT more accurate overall than LEM with regard to intermittency (a good fundamental question for someone to study).  Triplet maps with unequal sizes of the three map images would be inherently more intermittent (consider the limit in which one or two of the images is/are very small, corresponding to a very high gradient amplification factor) so this provides the freedom to tune LEM (or ODT) intermittency.  This would be another worthwhile study, recently rendered feasible by the adaptive mesh implementation.

 

These points could be of general interest, so I'm cc'ing the google group.

 

Alan

---

From: Steve Krueger [mailto:arcus...@earthlink.net]
Sent: Thursday, February 12, 2009 10:06 AM
To: Kerstein, Alan
Subject: Re: ODT Discussion - dates

I suggested that John could contribute to implementing better treatments of these, especially (1), and perhaps (2). It is less obvius to me how to implement (2) in LEM. It would seem to require correlated intermittency in space and time, which ODT does nicely, but LEM can't do without some type of memory.

On Feb 10, 2009, at 9:13 AM, Kerstein, Alan wrote:

 

 

Did your discussion with Patrick resolve his concerns?

---

From: Steve Krueger [mailto:arcus...@earthlink.net]
Sent: Monday, February 09, 2009 9:13 AM
To: Kerstein, Alan
Subject: Re: ODT Discussion - dates

that sounds good. I'd rather get you when you're FOA. (like FOB...).

I had a nice chat with Patrick Chuang after the POST workshop about LEM, EMPM, cluscoll.

He had some good questions, including:

1. Should we sample from a pdf of entrained blob sizes in EMPM?

2. How could we represent intermittency in the EMPM, etc?

Steve