Carbon density -> Carbon stocks

[Cesar Terrer]

Dear all,

I need a global map of soil organic carbon STOCKS at 0-15cm from SoilGrids2.0. A 0-30cm layer is readily available, but not 0-15cm. I was suggested to compute 0-15cm using organic carbon densities at 0-5cm and 5-15cm. The layers are in g/dm3, so to transform to stocks in g/dm2 = t/ha, I have multiplied them by thickness in dm and summed them up:

c_stocks_015cm (t/ha) = (c_density_05cm * 0.5) + (c_density_515cm * 1)

I assume carbon stocks at 0-15cm calculated in this way should be ~1/2 the readily available carbon stocks at 0-30cm. However, the resulting 0-15cm is 10x higher. Either I did something wrong, or C density is not in g/dm3, but in dg/dm3. I think the second explanation is quite likely, because in the online viewer carbon densities appear an order of magnitude larger than stocks, whereas given the reported units they should be comparable: https://soilgrids.org/

I have thus divided by 10:

c_stocks_015cm (t/ha) = (c_density_05cm * 0.5) + (c_density_515cm * 1) / 10

However, the resulting layer at 0-15cm is only a tiny bit smaller than the 0-30cm, while it should be ~1/2.

Is anyone experiencing similar issues? Would anyone be able to illustrate how to successfully compute C stocks at 0-15cm?

Thanks much,

Cesar