# Questions about "Material model/latent heat melt": input parameter"reference temperature" is useless?

**URL:** https://community.geodynamics.org/t/questions-about-material-model-latent-heat-melt-input-parameter-reference-temperature-is-useless/2070
**Category:** ASPECT
**Created:** [November 4, 2021, 11:58am UTC](https://community.geodynamics.org/t/questions-about-material-model-latent-heat-melt-input-parameter-reference-temperature-is-useless/2070 "2021-11-04T11:58:37Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![Qyunian](https://yyz2.discourse-cdn.com/flex036/user_avatar/community.geodynamics.org/qyunian/32/757_2.png) [@Qyunian](https://community.geodynamics.org/u/Qyunian)
#### Post date: [November 4, 2021, 11:58am UTC](https://community.geodynamics.org/t/questions-about-material-model-latent-heat-melt-input-parameter-reference-temperature-is-useless/2070/1 "2021-11-04T11:58:37Z")

</div>

Hi,  
As the title, I find the change of “input paameter:reference temperature” doesn’t influence the density profile in my model, it’s like that the value of “reference temperature” always be zero in the input.  
Then I read the “source/material model/latent heat melt.cc”, the code as follows:

```auto
const double delta_temp = in.temperature[i] - reference_temperature;
........

 if (this->include_adiabatic_heating ())
    {
       // temperature dependence is 1 - alpha * (T - T(adiabatic))
      temperature_dependence -= (in.temperature[i] -this>get_adiabatic_conditions().temperature(in.position[i])) * thermal_alpha;
     }
    else
       {
         temperature_dependence -= in.temperature[i] * thermal_alpha;
       }
.............

out.densities[i] = (reference_rho + composition_dependence + pressure_dependence)*
temperature_dependence * (1.0 - melt_dependence);

```

density=reference\_density\*(1 - alpha_deltT) ;  
but the code was written as density=reference\_density_(1 - alpha\*T). i.e. " temperature\_dependence -= delt\_temp(not in.temperature[i]) \* thermal\_alpha