College of Earth Sciences��Jilin University��Changchun 13006��China

Porosity is one of the most important properties of oil and gas reservoirs�� The porosity data that come from well log are only available at well points�� It is necessary to use other method to estimate reservoir porosity�� Seismic data contain abundant lithological information�� Because there are inherent correlations between reservoir property and seismic data��it is possible to estimate reservoir porosity by using seismic data and attributes�� Probabilistic neural network is a powerful tool to extract mathematical relation between two data sets�� It has been used to extract the mathematical relation between porosity and seismic attributes�� Firstly�� a seismic imped- ance volume is calculated by seismic inversion�� Secondly�� several appropriate seismic attributes are extracted by using multi- regression analysis�� Then a probabilistic neural network model is trained to obtain a mathematical relation between porosity and seismic attributes�� Finally��this trained probabilistic neural network model is im- plemented to calculate a porosity data volume�� This methodology could be utilized to find advantageous areas at the early stage of exploration�� It is also helpful for the establishment of a reservoir model at the stage of reservoir development��