# Reconstructing multiple shapefile polygon features through time

**URL:** <https://discourse.gplates.org/t/reconstructing-multiple-shapefile-polygon-features-through-time/916>\
**Category:** Q&A\
**Created:** [5 July 2024 23:03 UTC](https://discourse.gplates.org/t/reconstructing-multiple-shapefile-polygon-features-through-time/916 "2024-07-05T23:03:44Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![DCoutts](https://discourse.gplates.org/letter_avatar_proxy/v4/letter/d/bb73d2/32.png) [@DCoutts](https://discourse.gplates.org/u/DCoutts)\
**Post date:** [5 July 2024 23:03 UTC](https://discourse.gplates.org/t/reconstructing-multiple-shapefile-polygon-features-through-time/916/1 "2024-07-05T23:03:44Z")

</div>

I’m trying to reconstruct multiple polygons through time using gplately. They are H3 grid hexagons (variable name “H3”) that I would like to use for time-transgressive prediction. I load them as a feature collection (originally a .shp), then I am hoping to get their reconstructed geometry as well as plot them using gplately.

And have tried reconstructing them through “model.reconstruct(H3, time)” or through the “gplot.get\_feature(H3)”. For get.feature, I have done this as loop through each feature in the collection and it does not seem to help. I get a reconstructed feature geometry, but any plotting does not line up where it should, nor are the coordinates correct.

It seems like regardless, I do not get reconstructed geometry and am returned the original geometry. I can reconstruct other files (e.g., coastlines) with the same methods. All of the H3 features would be anchored on North America at time=0. I have tried setting the plate ID to 101.

Happy to answer any questions you might have. Thanks in advance for your help.  
Danny

---

<div class="post-metadata">

**Author:** ![mchin](https://discourse.gplates.org/user_avatar/discourse.gplates.org/mchin/32/2_2.png) [@mchin](https://discourse.gplates.org/u/mchin)\
**Post date:** [6 July 2024 10:23 UTC](https://discourse.gplates.org/t/reconstructing-multiple-shapefile-polygon-features-through-time/916/2 "2024-07-06T10:23:44Z")

</div>

Hi Danny,

If you show me your code and data, I can try to find a solution.

---

<div class="post-metadata">

**Author:** ![mchin](https://discourse.gplates.org/user_avatar/discourse.gplates.org/mchin/32/2_2.png) [@mchin](https://discourse.gplates.org/u/mchin)\
**Post date:** [8 July 2024 02:33 UTC](https://discourse.gplates.org/t/reconstructing-multiple-shapefile-polygon-features-through-time/916/3 "2024-07-08T02:33:54Z")

</div>

If the reconstruction already returns the original geometries, you need to check if the correct plate IDs have been assigned to the features.

---

<div class="post-metadata">

**Author:** ![DCoutts](https://discourse.gplates.org/letter_avatar_proxy/v4/letter/d/bb73d2/32.png) [@DCoutts](https://discourse.gplates.org/u/DCoutts)\
**Post date:** [8 July 2024 20:45 UTC](https://discourse.gplates.org/t/reconstructing-multiple-shapefile-polygon-features-through-time/916/4 "2024-07-08T20:45:18Z")

</div>

Thanks for this ^ , I had not been setting them quite right. This comment got me headed on the right path after turning in circles for a bit. I then figured out I had to partition the plates not just set plate ID to get the results I wanted. Thanks.
