The Weather Forecasting Forum

Discussions about the latest in meteorology, programs, tools, and techniques
It is currently Sat Dec 16, 2017 4:38 pm

All times are UTC




Post new topic Reply to topic  [ 14 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Strange map behaviour
PostPosted: Wed Nov 09, 2011 2:38 pm 
Offline
Senior Member

Joined: Tue Nov 25, 2003 4:22 pm
Posts: 198
Location: Suoi Tien, Vietnam
Hello

The last 2 years I have been very busy with moving from Holland to Vietnam and building a house there. So no time for D.A. Now the house is more or less finished so I have some time to work with DA again. I have V 2.09 Pro. I also have a 23” monitor now, so one of the first things I did was making maps that fill the screen. During this I noticed something strange.

I saw that the line of the isobars were on different places when the size of the map changed. I have tested with a fixed height of 960 and a width of 640, 960, 1280 and 1500. The center of the map is lat 12 and long 109 (Vietnam). The scale is 2500.

I should expect the lines on the same place, no matter what the size of the map is. The pressure should not be depending on the map size.

Am I thinking in a wrong way? Is there something wrong with my installation? Does others see the same behaviour, perhaps on other parts of the world?

Greetings

Fred


Top
 Profile  
 
PostPosted: Wed Nov 09, 2011 3:52 pm 
Offline
Moderator

Joined: Sun Nov 23, 2003 9:55 pm
Posts: 30
Location: Perth, Scotland, UK
Fred,
Can you post a couple of example pictures (not full size!)

Chris


Top
 Profile  
 
PostPosted: Wed Nov 09, 2011 6:49 pm 
Offline
Moderator

Joined: Wed Jan 06, 2010 1:17 pm
Posts: 57
Location: Lochaber, Scotland
What's the screen resolution and is it widescreen?
I know that things look much different on my own screen if I have it set to a normal resolution instead of the 900x1600 that it should be.
Also, do other images look OK when at full screen size?


Top
 Profile  
 
PostPosted: Thu Nov 10, 2011 9:54 am 
Offline
Senior Member

Joined: Tue Nov 25, 2003 4:22 pm
Posts: 198
Location: Suoi Tien, Vietnam
It's a 23"screen, set on 1920*1080.

It's not that the map looks strange, the isobars are on a complete different place, depending on the size of the map. It's easy trying. Generate the 4 maps with the size a gave in the first post and for each map hit the SLP button (and watch the diffences I hope).

All my USB-ports went down a couple of days ago, so tonight I try in another way to post my maps.


Top
 Profile  
 
PostPosted: Thu Nov 10, 2011 12:45 pm 
Offline
Senior Member

Joined: Tue Nov 25, 2003 4:22 pm
Posts: 198
Location: Suoi Tien, Vietnam
I just made new maps. I attach here the 680*960 and the 960*960. Look specific to the 10 and 11 lines.


Attachments:
960x960.jpg
960x960.jpg [ 162.59 KiB | Viewed 4540 times ]
680x960.jpg
680x960.jpg [ 146.74 KiB | Viewed 4540 times ]
Top
 Profile  
 
PostPosted: Thu Nov 10, 2011 3:08 pm 
Offline
Moderator

Joined: Wed Jan 06, 2010 1:17 pm
Posts: 57
Location: Lochaber, Scotland
Interesting one, Fred.
No idea what's happening there. :confused:


Top
 Profile  
 
PostPosted: Thu Nov 10, 2011 8:36 pm 
Offline
Moderator

Joined: Sun Nov 23, 2003 9:55 pm
Posts: 30
Location: Perth, Scotland, UK
Fred,
I can reproduce the problem. I generated both a 680 and a 960 map using your co-ordinates and then plotted SLP using data from my last download - I agree the isobars do not match.

I too, have no explanation :confused:

Chris


Top
 Profile  
 
PostPosted: Fri Nov 11, 2011 2:14 am 
Offline
Member

Joined: Thu Jul 15, 2004 3:42 pm
Posts: 33
Location: Kansas
I believe that the number of grid points DA uses in the analysis does not change when you change magnifications/scales/dimensions. So, zooming out, or changing one dimension, will result in different grid spacing. Additionally, the additional observations on the edges of the grid/display in the zoomed out version will result in a different analysis, particularly near the edges if these are data sparse areas with few observations nearby.

For example, using the same dimensions that you did in your example with the two graphics, gives me 18 grid points east/west across Kansas in both images, but 7 grid points north/south on the 960x680 and 9 grid points north/south on the 960x960. The individual grid boxes are square on the 960x960 image and rectangular on the 960x680 image.

It might be helpful to plot the observations that were used so we can see what the observation spacing is across the map. It could explain quite a bit. With a dense observation network across the map area, one would expect little variation due to your change in dimensions. With a sparse observation network, much larger differences should exist between the two.

You could also try a different analysis scheme (Barnes, Cressman) and see if there is such a large discrepancy.


Top
 Profile  
 
PostPosted: Fri Nov 11, 2011 3:51 am 
Offline
Senior Member

Joined: Tue Nov 25, 2003 4:22 pm
Posts: 198
Location: Suoi Tien, Vietnam
I’m not convinced yet. I try to explain my reasoning.

I attach 2 new maps, 680*960 and 1280*960. They are made with the same data as the previous maps. This time I used Cressman for analysis. This hardly gave differences, only at the edges of the map.

On both maps the plots are on the same place of the country, so I think this part of the grid is OK. In my opinion it also means that analysis of pressure, temperature etc. over the whole grid will be the same.

When plotting the lines are on different places, depending on the size of the map. So, again in MY opinion, it looks like that during plotting, the program does not look to the place from the map versus the right place in the grid.

I am very curious what more experienced meteorologists will find when they try the same for their location. I am still a beginner in this field.


Attachments:
plot 1280.gif
plot 1280.gif [ 54.87 KiB | Viewed 4513 times ]
plot 680.gif
plot 680.gif [ 46.89 KiB | Viewed 4513 times ]
Top
 Profile  
 
PostPosted: Fri Nov 11, 2011 6:58 am 
Offline
Member

Joined: Thu Jul 15, 2004 3:42 pm
Posts: 33
Location: Kansas
The sea level pressure analyses on my attached maps look quite a bit better than yours. But then I don't think the map area is as large for mine, and there is a much greater density of observations and more uniform spacing.


Attachments:
960by960.gif
960by960.gif [ 57.35 KiB | Viewed 4509 times ]
680by960.gif
680by960.gif [ 56.3 KiB | Viewed 4509 times ]
Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 14 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group