|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | [gislist] Ocean circulation model interpolation |
| Date: |
01/30/2006 01:35:01 PM |
| From: |
Jesse Cleary |
|
|
Hi Listers
I need some data format brainstorming help. I've been asked to
interpolate and serve Sea Surface Height output from a basin scale ocean
circulation model. The model generates height output for nodes in a
triangular network. This data will be available to me as a NetCDF file
with x,y,z,t attributes at each grid node. I need to think carefully
about what format I translate this into for possible web service as well
as future analysis. I would prefer a continuous surface visualization
to a contour representation, but would consider both. This process also
needs to be automated since the model runs can appear hourly.
Some limitations:
I will be serving this with MapServer, which eliminates TIN from final
consideration, unless it is used as the way into another format.
The number of model output nodes is now ~60k but will be expanding to ~500k.
Topology is important - both the value points themselves as well as
facet/polygon membership need to be considered in our interpolation method.
I have local installs of ArcGIS 9.0, ArcINFO, MatLab and a Linux server
with GMT, GDAL/OGR, & Postgres/PostGIS DB
I can flatten the NetCDF into hourly time slices as my node value file.
I have already created a polygon shapefile that represents the static
model network. So, the question is how to get the values at each node
point/time to reference this polygon shapefile in order to interpolate
values into a sea surface height raster/contour for service by
MapServer...and automate the entire process to run every hour.
Likely possibilities seem to be:
1. ESRI 3D coverage to store model network polygons and then insert new
point layers each hour for interpolation. I've limited knowledge about
coverages but am willing to try it out.
2. TIN (static network polygon + value point input) with further
conversion to ESRI grid.
3. some kind of point to raster interpolation that can draw upon my
underlying network polygon topology
4. A clever suggestion for one of you?!
Thanks for any ideas or additional resources y'all might suggest...
Jesse Cleary
--
Jesse Cleary
Department of Marine Sciences
UNC Chapel Hill
17-6 Venable Hall
jcleary@email.unc.edu
(919) 962-4323
_______________________________________________
gislist mailing list
gislist@lists.geocomm.com
http://lists.geocomm.com/mailman/listinfo/gislist
_________________________________
This list is brought to you by
The GeoCommunity
http://www.geocomm.com/
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
