How to explain the difference of variables in random effects component

Hi all,

I am fitting a very simple linear mixed model by using lme4. It is like 
this:
ModelA<-lmer(TF~ P*A + (1| DistrictID),data, REML=TRUE)
ModelB<-lmer(TF~ P*A + (1+AREA| DistrictID),data, REML=TRUE)
ModelC<-lmer(TF~ P*A + (1+Distance| DistrictID),data, REML=TRUE)

By checking AIC and BIC, it is found that ModelB seems to be the best. 
Could you please tell me how to explain the the impacts of AREA and 
Distance?
The objective is to know the impacts of P and A on TF. AREA and Distance 
are characteristics of the district, therefore, they are added in the 
random effects component.
I am new for mixed model. Could you please help me?

Thanks.

Tetsuya