Use contrast as measure of effect size glmmTMB

Question

I'm investigating the effect of an experimental treatment with the response variable being Weight (Wt). I have 2 control treatments (A,B) & 2 experimental treatments (C,D). I'm looking to see if the effect of treatment D relative to the control treatments increases with temperature (med_Hobo), and the same for treatment C relative to treatment D. (Additional fixed effects: Atl.Pac, NS, and random effect of Site)

Contrast matrix:

mat<-rbind(c(-0.5,-0.5,0,1),
           c(0,0,-1,1))
Cmat<-ginv(mat)

Model:

WW<-glmmTMB(Wt~med_Hobo*Treatment*Atl.Pac*NS+(1|Site)+ (1|Site:Treatment), 
            data=WetWtsEnv,contrasts=list(Treatment=Cmat),family=Gamma(link="log"))

Using emmeans and emtrends I can demonstrate that the mean contrast 1 is significant while the mean constrast 2 is not. But I'm struggling to assess any change in the contrast with Temperature. In the figure below, Treatment D (blue) clearly diverges with Temperature, how do I demonstrate this in numbers/figures? E.g. I'd expect contrasts at med_Hobo=10 to be 1 and increase to a factor of ~2 at med_Hobo=30, with a corresponding increase in c.i.

Weight predictions based on glmmTMB model

Thanks in Advance!