when I posted my questions this morning -printed below- I have not

emphasized enough the question on the transformation from time to space

domain. The assumptions I have described below are just crude assumptions,

because there can be a change in cloud top height, during the advection

process. Advection can not be described as easy as I have done, because wind

direction and speed changes with altitude etc.

So the assumptions below and the resulting time-space-transformation can

describe the data very well and is therefore widely used, but the results

depend on the atmospheric conditions (wind speed etc.) described above.

Because of these uncertainties, I asked for other concepts.

One interesting concept for example is described in

http://www.bom.gov.au/bmrc/wefor/staff/cnj/Papers/jakob_etal_jgr2004.pdf

Next question refers to the radars nearby: when I use the transformation

above, I get two parallel lines with data. From these lines I can get

information for my variogram for other directions than the direction of the

lines. But these data are very sparse and there are always just particular

leg distances for every direction.

When thinking about this problem I was inspired by the article

J.R.Key (1993) Estimating the area fraction of geophysical fields from

measurements along a transect. Geoscience and Remote Sensing, IEEE

Transactions on ,Volume: 31 , Issue: 5 Pages:1099 - 1102, though Key deals

just with indicator variables and isotropic fields but gives an insight how

many transsects are necessary to describe the underlying field.

Ok, sorry for this huge amount of text !

And thanks for your suggestions in advance !

Klemens

--------

Klemens Barfus

Institute for Hydrology and Meteorology

Technical University of Dresden

Germany

>

--

>

> >Hello list members,

> >

> >1. Does there any concepts or ideas concerning the comparability of time

> >series and spatial data exist ?

> >

> >I deal with cloud top heights measured from satellite as a field and

> >measured by radar as a time series.

> >If I assume, that the clouds are advected / transported over the radar,

> then

> >I can assume a wind speed and with this wind speed I get the horizontal

> >extend of the time series.

> >

> > Now I can calculate spectrum, mean and variance

> >and can compare this to the same data from the satellite field. But are

> >there any other concepts to compare / link time series data with field /

> >spatial data.

> >2. When I get the information about the spectrum of the time series I

> assume

> >isotropy to project these spectrum on a two dimensional field.

> >But when there is a co-located radar nearby, I assume the same wind

> >direction for both radars and transform the time series to space like

> >described above, are there any concepts to infer information about

> isotropy

> >and anisotropy of the underlying field from these two 'line located' data

> ?

> >

> >

> >Thanks for your help, inputs, and ideas in advance !

> >

> >Klemens

> >

> >--------

> >Klemens Barfus

> >Institute for Hydrology and Meteorology

> >Technical University of Dresden

> >Germany

> >

> >

> >

> >

> >------------------------------------------------------------------------

> >

> >* By using the ai-geostats mailing list you agree to follow its rules

> >( see http://www.ai-geostats.org/help_ai-geostats.htm )

> >

> >* To unsubscribe to ai-geostats, send the following in the subject or in

> the body (plain text format) of an email message to sympa@...

> >

> >Signoff ai-geostats

> >

NEU: WLAN-Router f�r 0,- EUR* - auch f�r DSL-Wechsler!

GMX DSL = superg�nstig & kabellos http://www.gmx.net/de/go/dsl