Weird soil depths in EFAS version 5 data

Datasets and Usage
, ,
1

Hello.

I am trying to download the soil moisture content data from the EFAS hydrological reanalysis version 5 product: (River discharge and related historical data from the European Flood Awareness System)

This product gives sub-daily soil moisture data for 3 soil layers at sub-daily time steps for multiple years.

The soil layers information is in the variable soil depth, which gives the depth of the bottom of each layer from the surface. This is where I noticed something strange. The official documentation (Model Output - Copernicus Emergency Management Service - CEMS - ECMWF Confluence Wiki) gives the units as meters.

But when I upload the netCDF file in Python using xarray and actually check it, the soil depth for layer 1 goes till 50 meters, and the soil depth for layer 2 goes till 1400 m (1.4 km !!!). This does not seem realistic, as I have never seen a hydrological model or a land surface model working with these high depths of soil, unless the focus is on the deep layers.

Is it possible that the units of soil depth is millimeters instead of meters? Then a depth of 50mm (5cm) and 1400 mm (140 cm) would make much more sense.

Even ERA5-Land has soil water moisture data for similar depths (7 cm, 28 cm etc.)

Can you please let me know if I am missing something?

Here is a plot for soil depths for layer 2 for a part of Croatia, and you see the values go till 1400:

image
image584×438 31.6 KB

So the main question is: is it 1400 meters? or cm? or mm?

With kind regards,
Karan

2

Dear Karan_Mahajan,

Thank you for getting in touch with us and raising this issue.

Indeed, you are correct that there is an error in our documentation. The unit for soil depth in this case (i.e. using LISFLOOD model for EFAS) should be millimetres (mm). You interpretation on how the depth is defined is correct; the file gives the depth of the bottom of each layer from the surface - the soil depth for layer 1 goes to 50 mm and the soil depth for layer 2 goes till 1400 mm, and so on.

We apologise for the error in our soil depth documentation for EFAS. It seems we attempted to describe soil depth as a grib parameter (as it is used in multiple in-house applications) on this documentation page. Some of the applications that use soil depth are archived in MARS with the unit meters (m), but this is not the case for LISFLOOD soil depth which is used in EFAS and is measured in millimetres (mm).

We will review our documentation and update it as soon as possible.

Many thanks again for your attention to detail and for bringing this to our attention.

Kind regards,

Karen, CEMS-Flood Support

2 Likes