# Using Raviart-Thomas elements on a surface mesh

**URL:** https://community.freefem.org/t/using-raviart-thomas-elements-on-a-surface-mesh/3614
**Category:** General Discussion
**Created:** [December 4, 2024, 4:42pm UTC](https://community.freefem.org/t/using-raviart-thomas-elements-on-a-surface-mesh/3614 "2024-12-04T16:42:58Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![yowie](https://avatars.discourse-cdn.com/v4/letter/y/e9bcb4/32.png) [@yowie](https://community.freefem.org/u/yowie)
#### Post date: [December 4, 2024, 4:42pm UTC](https://community.freefem.org/t/using-raviart-thomas-elements-on-a-surface-mesh/3614/1 "2024-12-04T16:42:58Z")

</div>

Hey y’all!

As a part of my master thesis, I’m trying to implement a mixed finite element method for a diffusion equation on a _curved_ surface. This means that I have to use `RT0` finite elements on a `meshS`. The problem is that functions in such a finite element space only have **two** components (e.g. `[v1, v2]`), whereas the global x-y-z coordinate system is obviously **three** -dimensional. In particular, this means that I can’t simply use `dx`,`dy`,`dz` to compute the surface divergence of such a function. How would I go about using `RT0` elements on a `meshS`, and is it possible to compute the surface divergence?

Thanks in advance for your time!