Parameterising snow density in microclimate model

[Shane Morris]

Hi everyone,

This is not a question but just a post to make people aware of something I was not aware of: that the snow density function is by default set to a maritime area and you can change it to other environments.

I encountered this issue when modelling the snow depth (using micro_era5) for a site in northern Sweden and could not understand why it kept underestimating the snow depth over long periods of time by  ~30%. The issue was that the default exponential function for the densfun parameter is  c(0.5979, 0.2178, 0.001, 0.0038). The help file states Sturm et al. 2010 J. of Hydromet. 11:1380-1394  (https://doi.org/10.1175/2010JHM1202.1) is the source of this. If you look at this paper it has the different exponential functions for the different snow environments. Once I switched mine to the taiga function my snow model worked perfectly. 

Here are the functions for all the sites in the paper 

Alpine = c(0.5975, 0.2237,0.0012, 0.0038)

Maritime = c(0.5979, 0.2578,0.001, 0.0038)

Prairie = c(0.594, 0.2332,0.0016, 0.0031)

Taiga = c(0.2170, 0.2170, 1e-5, 1e-5) 

Tundra = c(0.3630, 0.2425,0.0029,  0.0049)

Hope this helps,

Shane.