missing values in 250m soil grids

133 views

Skip to first unread message

Ruhollah Taghizadeh

unread,

Jan 4, 2018, 9:57:12 AM1/4/18

to Global Soil Information

Dear Tom,

first of all, I would like to thank you and your team for valuable effort to produce global soil map 250m.

I am currently working to produce digital soil maps in Iran (e.g. soil class and soil carbon). I want to use 250m digital global maps as covariates. However, there are missing values in the maps: carbon, clay
content, pH, .....

I need your help:
Please let me know how I can deal with these NA values. What is the best way to fill NA values?

Any other suggestions from you would be acknowledged.

Best regards
Ruhollah

Bas Kempen

unread,

Jan 19, 2018, 8:35:13 AM1/19/18

to Global Soil Information

For SoilGrids we used a land mask for our predictions. This grid masks out all non-soil pixels according such as water bodies, cities, roads, mountains (bare rock). So if you want to predict soils and want to use SoilGrids as a covariate, then I think it is OK to have the NA values since these represent areas where there is no soil. At least, that is what I often do.

In the past I did fill NA gaps in grids using ArcGIS. There is the nibble tool that can do this. You might want to look into this. It is a bit of a cumbersome procedure but it works:

1) Run the IsNull tool to identify NA values and set these to 1

2) Run the Con tool: sets NoData identified on basis of IsNull to a value

3) Run the Nibble tool: Input = output of the Con tool; Mask = original raster (SoilGrids)