# Selecting projection for comparison with satellite data **URL:** \ **Category:** General\ **Created:** [December 12, 2023, 9:02am UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372 "2023-12-12T09:02:40Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![atmoboran](https://avatars.discourse-cdn.com/v4/letter/a/ecccb3/32.png) [@atmoboran](https://openradar.discourse.group/u/atmoboran)\ **Post date:** [December 12, 2023, 9:02am UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/1 "2023-12-12T09:02:40Z")
Thanks for the hint. Yes it is used for further processing, so this should be redone. Firstly I want to put satellite data (Meteosat) over it. And the satellite data is in lat lon, but I can also download it in other projections. So I thought of maybe using the Lambert cylindrical equal-area projection (EPSG:9835), but somehow osr, doesn’t recognize it. proj = osr.SpatialReference() proj.ImportFromEPSG(9835) → proj stays empty Do you have suggestions for a crs, that makes sense to use here? ---
**Author:** ![kmuehlbauer](https://yyz2.discourse-cdn.com/free1/user_avatar/openradar.discourse.group/kmuehlbauer/32/439_2.png) [@kmuehlbauer](https://openradar.discourse.group/u/kmuehlbauer)\ **Post date:** [December 12, 2023, 9:21am UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/2 "2023-12-12T09:21:23Z")
@atmoboran I’m no expert on projections, so I’d appreciate any suggestions here. AFAICT EPSG:9835 is just referring to the projection method, but is no projection. You would need to pick some projection which best suits your downstream use cases. A search for `equal area` brought up several projections including these two **EPSG:3035** and **EPSG:19986**. HTH, Kai P.S. We can swap this projection question out to a new topic if you like. ---
**Author:** ![atmoboran](https://avatars.discourse-cdn.com/v4/letter/a/ecccb3/32.png) [@atmoboran](https://openradar.discourse.group/u/atmoboran)\ **Post date:** [December 12, 2023, 9:30am UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/3 "2023-12-12T09:30:10Z")
> P.S. We can swap this projection question out to a new topic if you like. Yes that makes sense. ---
**Author:** ![atmoboran](https://avatars.discourse-cdn.com/v4/letter/a/ecccb3/32.png) [@atmoboran](https://openradar.discourse.group/u/atmoboran)\ **Post date:** [December 12, 2023, 9:50am UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/4 "2023-12-12T09:50:55Z")
Why is 32636 always used in the recipes? Is this one equal-area? ---
**Author:** ![kmuehlbauer](https://yyz2.discourse-cdn.com/free1/user_avatar/openradar.discourse.group/kmuehlbauer/32/439_2.png) [@kmuehlbauer](https://openradar.discourse.group/u/kmuehlbauer)\ **Post date:** [December 12, 2023, 1:21pm UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/5 "2023-12-12T13:21:35Z")
It’s actually **EPSG:32632** (UTM Zone 32N). It covers most of Germany. And you are right it’s not really equal area but the distorsions are minimized compared to normal tranverse mercator projection. ---
**Author:** ![atmoboran](https://avatars.discourse-cdn.com/v4/letter/a/ecccb3/32.png) [@atmoboran](https://openradar.discourse.group/u/atmoboran)\ **Post date:** [December 12, 2023, 2:50pm UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/6 "2023-12-12T14:50:27Z")
So do I understand correctly, that the interpolation method you suggest in ([Georeferencing single sweep data and converting to numpy array - #2 by kmuehlbauer](https://openradar.discourse.group/t/georeferencing-single-sweep-data-and-converting-to-numpy-array/353/2)) is only “totally” correct if the data is on an equal area projection? And 32632 is basically close enough to equal area? ---
**Author:** ![kmuehlbauer](https://yyz2.discourse-cdn.com/free1/user_avatar/openradar.discourse.group/kmuehlbauer/32/439_2.png) [@kmuehlbauer](https://openradar.discourse.group/u/kmuehlbauer)\ **Post date:** [December 12, 2023, 3:02pm UTC](https://openradar.discourse.group/t/selecting-projection-for-comparison-with-satellite-data/372/7 "2023-12-12T15:02:10Z")
Nothing is “totally” correct, as Earth is no sphere, unfortunately. I’m using the UTM Zone32N in that example as it covers Germany nicely and it’s used by several agencies. Sorry, but my knowledge in projections is somewhat limited.