最近、TUFLOWのMultidimensional NetCDFファイルを扱っていて、データの投影に関していくつか問題に直面しています。データがミラーリングされている（EとNが入れ替わっているが原点は正しい）場合や、データセットの原点がラスターの異なる位置（例えばラスターの中央ではなく下部）から来ている場合など、いくつかの異なる問題に遭遇しました。 <\/P>
他に同じような経験をした方はいらっしゃいますか？<\/P>
Thanks everyone. I ended up getting additional support back from TUFLOW and they are adding another upvote to improve the compatibility of their NC outputs. In the meantime they provided me with a script that utilises the aforementioned netcdf4 python library which writes the projection files into the NC file. As far as I can tell ESRI seems to ignore the accompanying proj file.I'll look at converting the supplied python script into a ESRI toolbox script and post it here for anyone else that runs into this issue.
Hi James, when we started to work with NetCDFs in Image Services a couple of years ago, we did have a bit of trouble with coordinate system definitions: our science data are recorded on a rotated pole coordinate system and then written into NetCDF format - that 'rotated pole' coordinate reference system was the first thing which had to change.
After using a more commonly understood projection system, we figured out that the way the scientists had defined that system was common in software clients in the climate science world, but couldn't easily be read into a mosaic dataset. After changing the definition to a common text string it started to work very well.
An example of that definition (from https://epsg.io/3851 ) - you want to use the one called "Esri WKT"
PROJCS["NZGD_2000_NZ_Continental_Shelf_2000", GEOGCS["GCS_NZGD_2000", DATUM["D_NZGD_2000", SPHEROID["GRS_1980",6378137.0,298.257222101]], PRIMEM["Greenwich",0.0], UNIT["Degree",0.0174532925199433]], PROJECTION["Lambert_Conformal_Conic"], PARAMETER["False_Easting",3000000.0], PARAMETER["False_Northing",7000000.0], PARAMETER["Central_Meridian",173.0], PARAMETER["Standard_Parallel_1",-37.5], PARAMETER["Standard_Parallel_2",-44.5], PARAMETER["Latitude_Of_Origin",-41.0], UNIT["Meter",1.0]]
Cheers,
Tilmann
PS: after reading the above, I notice that I should clarify that we _generate_ the NetCDFs that we work with using scripts written here, so obviously we were able to modify the code used to do that - whereas I'm not sure whether you can do that with TUFLOW, which I don't know.
James,I am far from being a NetCDF specialist, but the issues you face sound familiar. If you read this page of the ArcGIS Pro Doc:https://pro.arcgis.com/en/pro-app/latest/tool-reference/multidimension/essential-netcdf-vocabulary.htm
you will find a hint regarding NetCDF conventions ... and that within ArcGIS we currently support the Climate and Forecast (CF) and the Cooperative Ocean/Atmosphere Research Data Service (COARDS) conventions. It also states "others might work" ... but to my experience this is exactly where the problems with coordinates and reading data/dimensions come from. The link above also has additional links regarding conventions in it.If you want to dig in deeper, ArcGIS Pro today comes with the pre-installed python library netcdf4 which might be a way to get to the data you want ...RegardsGuenter
サインインしたメンバーは投稿、更新のフォローなどができます。初めてですか？無料アカウントを登録してください。
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.
