We consider the problem of nonparametric estimation of the conditional hazard function for spatial data. More precisely, given a strictly stationary random field , we investigate a kernel estimate of the conditional hazard function of univariate response variable Y_i given the functional variable X_i. The principal aims of this article are to give the mean squared convergence rate and to prove the asymptotic normality of the proposed estimator. Finally, a simulation study and an application on real data are carried out to illustrate, for finite samples, the behavior of our method.