# Does Surject2Dto3D could be use on the TreeMesh

**URL:** https://simpeg.discourse.group/t/does-surject2dto3d-could-be-use-on-the-treemesh/121
**Category:** Electromagnetics
**Created:** [June 18, 2020, 9:03am UTC](https://simpeg.discourse.group/t/does-surject2dto3d-could-be-use-on-the-treemesh/121 "2020-06-18T09:03:24Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![zhongmin](https://avatars.discourse-cdn.com/v4/letter/z/74df32/32.png) [@zhongmin](https://simpeg.discourse.group/u/zhongmin)
#### Post date: [June 18, 2020, 9:03am UTC](https://simpeg.discourse.group/t/does-surject2dto3d-could-be-use-on-the-treemesh/121/1 "2020-06-18T09:03:24Z")

</div>

Hi all:  
I am working on 2.5D TEM inversion using Simpeg  
and it will very computational when the number of transmitter loop is large  
so I want to perform the inversion on the 2D mesh and mapping the results on the 3D mesh to perform forward,so I need use Surject2Dto3D in the Maps, it worked well on tensor mesh,  
but it failed on the TreeMesh

what should I do if I use Surject2Dto3D on the Treemesh ?  
or, if I want to inverse a time domain EM data along a profile with many source (like skytem with ~ 500 sources ), how can I save the computation time ? are there any tricks to calculate it fast ？

---

<div class="post-metadata">

### Author: ![sgkang09](https://yyz2.discourse-cdn.com/free1/user_avatar/simpeg.discourse.group/sgkang09/32/8_2.png) [@sgkang09](https://simpeg.discourse.group/u/sgkang09)
#### Post date: [July 9, 2020, 3:41pm UTC](https://simpeg.discourse.group/t/does-surject2dto3d-could-be-use-on-the-treemesh/121/2 "2020-07-09T15:41:30Z")

</div>

HI zhongmin,

That is correct. With the current simpeg time-domain implementation, this will be quite expensive.  
Surject2Dto3D would not work for the Tree mesh since we assumed the mesh is Tensor.

Not sure we have a good solution for your problem at this point with SimPEG. In general, one can set up a different mesh for each of your AEM source to save the cost (seel below referenece).

> **[3-D inversion of airborne electromagnetic data parallelized and accelerated...](https://academic.oup.com/gji/article/196/3/1492/581032)**
>
> Abstract. Airborne electromagnetic (AEM) methods are highly efficient tools for assessing the Earth's conductivity structures in a large area at low cost. Howe

Simulation part of this idea is implemented in [https://github.com/simpeg-research/kang-2019-3D-aem](https://github.com/simpeg-research/kang-2019-3D-aem).

Hope this helps.