@NiraYadav0529 I checked the residuals of the ammonium model (line 77 of Soil_nutrients_exp1.R) and this is showing the funnelling pattern (variation increases with the mean):
Make sure you are looking at these plots and evaluating whether you need to transform your response variable. In this case I think log10() will be appropriate but you have negative values that need to be addressed before doing this. Remember discussing the detection limit for the instrument and how to deal with values below this limit?
@NiraYadav0529 I checked the residuals of the ammonium model (line 77 of Soil_nutrients_exp1.R) and this is showing the funnelling pattern (variation increases with the mean):
Make sure you are looking at these plots and evaluating whether you need to transform your response variable. In this case I think log10() will be appropriate but you have negative values that need to be addressed before doing this. Remember discussing the detection limit for the instrument and how to deal with values below this limit?