Résumé
For spatial processes such as environmental ones, it is of great importance to understand how extreme events tend to occur jointly or separately. After a brief reminder on the concept of dependence of extremes, a madogram is introduced that conveys the relevant information of asymptotic dependence. In order to determine if the field is asymptotically independent at several lags, a multiple test procedure is derived, based on the bootstrap distribution of the number of times the null hypothesis is rejected. It is then tested on three classical spatial models and finally applied to two environmental datasets.