We put forward a new method to estimae one-dimensional distribution of subsdse paralel fracturesbased on core data. It is model-dependent so that fraCfore distribution pattem must be datennned at first.our estimation is to find out the optimum distribution with a given pattern, having minimum misfitbforeen theoretical and measured Probabilities of meeting fractures in core. Formulation of thetheoretical Probabilities has been deduced by using probability theory. In order to reveal subsurfacefracture distribution unknown in most cases, we mght repeat the procedure by selecting a variety ofdifferent patterns, and then have their optimutn distributions. Among them, the one with the leastminimum misfit can be considered as the best approximaion. Application of the method in two casesleads to much better description of complicated fractures in nthee. Therefore it is of common use, andwill give a more realistic fracture model in reservoirs for petroleum development. We put forward a new method to estimate one-dimensional distribution of subsdse paralel fractures based on core data. It is model-dependent so that fraCfore distribution pattem must be datennned at first. Estimate is to find out the optimum distribution with a given pattern, having minimum misfitbforeen theoretical and measured Probabilities of meeting fractures in core. Formulation of the thetical Probabilities has been deduced by using probability theory. In order to reveal subsurfacefracture distribution unknown in most cases, we mght repeat the procedure by selecting a variety of different patterns, and then have their optimutn distributions. Among them, the one with the leastminimum misfit can be considered as the best approximaion. Application of the method in two casesleads to much better description of complicated fractures in nthee. It it’s common use, andwill give a more realistic fracture model in reservoirs for petroleum development.