Hi all
My data is as follows:
I have seedling height divided <10 cm and >10cm
I collected data across a glacier foreland using square plots. Each plot had one seedling in the centre and I recorded the microsite conditions (% cover of existing vegetation, % cover of litter, % cover gravel etc) of the seedling.

In order to determine whether microsite conditions have an effect on seedling survival (<10 cm:1st season seedling, >10 cm established seedlings) i looked at differences between the plots with seedlings of >10 cm and the plots with <10cm. I used Kruskall wallis tests with Mann whitney post hocs to do this. However, now I want to add a new variable 'distance from glacier' as I want to see whether the effect of microsite conditions on seedling survival is different between those located within 450 m from the glacier and those beyond the 450m. I prefer 'distance from glacier' to be nominal as it suggests different terrain age/difference in terrain characteristics. Within 450 m the area is rocky and beyond the 450 m there is a very high cover of moss.

My question is: Should I divide my plots in two groups (those located <450 m and those >450 m), repeat the Kruskall wallis tests and Mann whitney post hocs and compare the results? Or is there another way/test where I could add the 'distance from glacier' variable and get the results I need? Any recommendations will be greatly appreciated. Hope I was clear enough as English is not my native language. Thank you in advance! Please reply as soon as you can!