4

Is it possible in QGIS to exclude some values from the symbology and define only 1 class with 1 color for all the other values?

I have hard time to find how to do it with the Singleband pseudocolor render type. It seems QGIS only allows 2 classes minimum (why not 1?) The problem is that the 2 classes appear in the legend afterward while in the legend I want only 1 class. I have a raster with values from 0 to 100 and I would like to exclude values from 0 to 29 and draw value from 30 to 100 in one green color.

In ArcGIS there is a convenient option where one can group certain values into one class at the symbology level (without touching anything in the original data).

Is there a way to produce he same effect in QGIS without reclassifying the data?

enter image description here

1 Answer 1

3

For some specific interval with one legend item, you can leave one class in Symbology and use configure Transparency setting (a bit workaround). In transparency settings you can exclude specific values (0 - 400 in my example):

enter image description here

If you need to display only 1 class with values less than defined, it can be done with Discrete interpolation (Value <= 500 in exaple):

enter image description here

5
  • 1
    This is incorrect, The OP wants to show the values of more than or equal certain values, what you are showing is the oppsite. You are showing the less that or equal. Have a look at the header of the classification, it shows Value <=.
    – ahmadhanb
    Oct 3, 2018 at 7:04
  • Yes you are right, sorry I will edit my question.
    – Oto Kaláb
    Oct 3, 2018 at 7:08
  • 1
    The only way to remove the hidden class is from the legend from Print composer -> Add legend -> Uncheck Auto update -> Select the hidden class and click on the red minus symbl (-), and the hidden class will be removed.
    – ahmadhanb
    Oct 3, 2018 at 7:16
  • My solution is fixed. Yes, this is another approach (not only option) which only remove the class from legend, not from the map item (the data stay symbolized), so still you have to symbolize data with transparent/white color with Discrete interpolation.
    – Oto Kaláb
    Oct 3, 2018 at 7:29
  • 1
    +1 from me for having an efficient approach in simple way.
    – ahmadhanb
    Oct 3, 2018 at 7:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.