Re: [ferret_users] Transformation: curvilinear to rectangular issue

[Szymon Roziewski <szymon.roziewski@xxxxxxxxx>]

Hi Martin,
You are right, I don't believe but etopo2 extends over big area and I need some data in Dutch Straits (on the left) because I work in rotated coordinates and etopo1 doesn't have such data. I would like to add them up to make one bathymetry for Baltic Sea.

2011/4/8 Martin Schmidt <martin.schmidt@xxxxxxxxxxxxxxxxx>

Hi Szymon,
Indeed, seems the island Bornholm is shed into the Baltic. Beware off the tsunami.
I cannot help you to find the error source in your grid transformation. Nevertheless, do you believe in etopo2 in
the Baltic Sea? Depth of less than 10 m in some areas?

You may want to get a better data set from here:

http://www.io-warnemuende.de/topography-of-the-baltic-sea.html

Greetings,
Martin

Szymon Roziewski wrote:

Dear Ferreters,
I have interpolated some data from curvilinear coordinates into regular.
First of all I computed the array of coefficients of a transform and then I did transform.
But what I got as a result somehow embarrassed me because on the left bottom side picture I have a blurred data and I don't know why.
I attach two pictures.
The first one that I called INPUT_etopo2 is an input of the transform on curvilinear grid.
The second is a print of interpolated data onto rectangular grid but with a blur.
The region which I concentrated on was
X=-5.7:5.5
Y=-2.5:10:0
What has caused this smear? What do you think?

Kind regards,
Szymon Roziewski

-- 
Z wyrazami szacunku,
Szymon Roziewski