[Qgis-user] problems reprojecting in equal earth

Nicolas Cadieux nicolas.cadieux at archeotec.ca
Wed Aug 21 17:52:01 PDT 2019


Hi,

I figured it out! (Well... you figured it out:)  I created a world grid 
with 1 degree longitudes.  I then densified this grid with the value 10 
000 (Vector/Geometry tools/densify by count). Then, I selected the 
longitude line I wanted to split (-44). I then used the Split with lines 
found in processing.  Reprojection still gave me those big polygons but 
this time, you can select them by clicking slightly right of the 
polygons (on the right side of the screen) on the reprojected project.  
Then, you can delete the problematic polygons.  Results are in the 
Google Drive.

https://drive.google.com/open?id=1eL6ws8Jc-PugMM2gxV79j0k-TGoWcw6B

Thanks for your help!

Nicolas

On 2019-08-21 3:26 p.m., Alex M wrote:
> I had this issue once also, I think the clipping method is the most
> reliable. If I recall the odd/invalid polygons depended on which scale
> of Natural Earth you use. Try downloading a different scale. I remember
> actually trying to fix the polygons and Russia was just too complicated
> to deal with at the time.
>
> This is a common problem anytime you wrap the dateline with world data,
> and you might find more solutions using those search terms.
>
> Thanks,
> Alex
>
> On 8/21/19 11:10, Nicolas Cadieux wrote:
>> Good idea Calvin,
>> Anyone else with a simpler, faster  solution?  If not, I will try.
>> Nicolas
>>
>>> Le 21 août 2019 à 13:43, C Hamilton <adenaculture at gmail.com> a écrit :
>>>
>>> The way I more or less solved something like this was by pre-clipping the Natural Earth data at the boundaries of the projection and made two different vector layers You then will not get the wrap around. If need be you can shift the longitude of one of layers by 180 degrees and then attempt to merge the two pieces together. For countries that now have a cut line down the middle you can then merge the pieces together. It can be time consuming, but will work. I don't know if there is an easier way.
>>>
>>> Calvin
>>>
>>>> On Wed, Aug 21, 2019 at 1:17 PM Nicolas Cadieux <nicolas.cadieux at archeotec.ca> wrote:
>>>> Hi,
>>>>
>>>> These are the nasty polygons.  I am hoping you will see the image.
>>>>
>>>> <djfpggkfmkkepjkp.jpg>
>>>>
>>>>> On 2019-08-19 12:26 p.m., Nicolas Cadieux wrote:
>>>>> Hi,
>>>>>
>>>>> Trying to help another Qgis user that posted earlier.
>>>>>
>>>>> We have Natural Earth vectorial Data that we what to reproject in Equal Earth.  We want Austria to be in the middle of the map so we created a Equal Earth Project using  custom CRS  using +proj=eqearth +datum=WGS84 +wktext +lon_0=136 with the help of https://proj.org/operations/projections/eqearth.html. Using  lon_0=136 puts Australia in the middle and limits the distortion in that longitude.
>>>>>
>>>>> Data was reprojected and saved in the new projection (vector/Data Management tools/Reproject Layer).  Project is also in the Equal Earth projection so no reprojection on the fly is happening behind the scenes.
>>>>>
>>>>> Question: How so we take care of the nasty polygons that appear when we add the "+lon_0=136" parameter?  See the zip.
>>>>>
>>>>> You can find the data here: https://drive.google.com/open?id=1eL6ws8Jc-PugMM2gxV79j0k-TGoWcw6B
>>>>>
>>>>> Qgis 3.8.1 on Windows 10_64
>>>>>
>>>>> Thanks
>>>>>
>>>>> Nicolas
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Qgis-user mailing list
>>>>> Qgis-user at lists.osgeo.org
>>>>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>>>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>>> _______________________________________________
>>>> Qgis-user mailing list
>>>> Qgis-user at lists.osgeo.org
>>>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
>> _______________________________________________
>> Qgis-user mailing list
>> Qgis-user at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
> .
>


More information about the Qgis-user mailing list