Issues related to PAWC calculations

When I went to calculate PAWC using soil data from the World Soil Database (HWSD), I found that the OC values of some pixels are more than 10 times higher than others, resulting in the final calculated PAWC above 1. There are also some pixels with null Sand, Slit, Clay and Oc data, which will have an impact on the final result, how should I deal with the anomalies in this part of the data?

Hi @GISerFJH -

I’m not sure what to say about the high OC values in the original data, but since PAWC values should be between 0 and 1, you could change any PAWC values that are > 1 to 1, meaning that all of the water in the soil is available for plants to use.

For the pixels where sand, silt, clay and OC are null, are these water bodies? If so, here’s a section of the User Guide that addresses this.

If the missing pixels are are not water bodies, then you’ll need to look at what kind of land cover/soil is there and determine the appropriate PAWC value to assign. Often soil data is missing where there are glaciers, for example. In the case of glaciers, you might be able to assume that plants cannot grow there (for several reasons), so it is safe to set those areas to a very low (or zero) PAWC value. In other cases where it’s less obvious why there’s no soil data provided, you can set the value to be the same as in the dominant soil type adjacent to the null areas.

~ Stacie

First of all, thank you for your reply, I compared the LUCC land use map with the soil data and found that most of the areas with abnormally high OC values are grasslands; secondly, when I compared the soil data with the areas with null values, I found that some of the grasslands in the plateau area have null values, and the rivers and paddy fields in the plains area have null values as well, is there a need for modification of these data?

In any pixel where any of your spatial inputs have a null value, the model results will be null. So if you want the model to produce results in those pixels, you will need to assign a PAWC value.

I’m not sure if you need to use HWSD for some reason, but you might also consider looking at ISRIC SoilGrids to see if it has better coverage for your study area.

~ Stacie