Scale effect for AWC data
67 views
Skip to first unread message
cerenballi
Jan 12, 2018, 6:56:21 AM1/12/18
to Global Soil Information
Dear Mr Hengl,
I need to use Available Water Holding Content (AWC) from SoilGrids for the calculations of Palmer Drought Severity Index (PDSI). So I downloaded "AWCh3_M_sl6_250m.tif" data. Unfortunately, I could not understand the data range which is between around 5 to 12 for my region (Turkey). However, the range of values should be between 50 to 250 for 100 cm soil depth.
I might need to think about a regional scale to get a real values. Could you please give me an idea or perspective to find a scale effect. Or anything helps?
Thank you in advance.
I will be appreciated.
Regards,
Ceren
Bas Kempen
Jan 23, 2018, 4:12:43 AM1/23/18
to Global Soil Information
Dear Ceren,
Sorry for the delay in the reply. The unit of the SoilGrids AWC layers is %. The range you are referring to I think is mm. Furthermore, the standard layer 6 (sl6) gives the predicted value at 100 cm depth. It is a point prediction. It is not an aggregated value for the 0-100 layer. To obtain this value you should take a weighted average of layers 1 to 6 as explained in the SoilGrids250m journal article (here an example for the 0-30 cm layer is used but a similar function can be applied to the 0-100 cm layer). To convert to mm water just multiply the aggregated AWC in % by 1000.
Hope this helps.
Best regards,
Bas
Sorry for the delay in the reply. The unit of the SoilGrids AWC layers is %. The range you are referring to I think is mm. Furthermore, the standard layer 6 (sl6) gives the predicted value at 100 cm depth. It is a point prediction. It is not an aggregated value for the 0-100 layer. To obtain this value you should take a weighted average of layers 1 to 6 as explained in the SoilGrids250m journal article (here an example for the 0-30 cm layer is used but a similar function can be applied to the 0-100 cm layer). To convert to mm water just multiply the aggregated AWC in % by 1000.
Hope this helps.
Best regards,
Bas
Tomislav Hengl
Jan 25, 2018, 4:11:58 AM1/25/18
to Global Soil Information
Hi Ceren,
The AWC (volumetric fraction) is culculated using the PTF http://gsif.r-forge.r-project.org/AWCPTF.html function, which is explained in detail here: https://github.com/cran/GSIF/blob/master/R/AWCPTF.R
This PTF was unfortunately NOT callibrated using global points (http://www.sciencedirect.com/science/article/pii/S001670611200417X) so I expect problems (it probably gives reasonable estimate of AWC but not for all climatic zones), so please use with care. You might want to take a look also at the EU soilhydrogrids (which fortunately also covers Turkey): https://eusoilhydrogrids.rissac.hu/
As Bas mentioned - to aggregate to standard soil layers e.g. 0-100 cm best use the trapezoidal rule (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0169748#pone-0169748-g001).
HTH,
T. Hengl
On Friday, January 12, 2018 at 12:56:21 PM UTC+1, cerenballi wrote:
Reply all
Reply to author
Forward
0 new messages