墨西哥湾天然气水合物矿产资源预测模型及其应用

Mineral resources prediction model and its application about gas hydrate in the Gulf of Mexico

墨西哥湾是世界上研究天然气水合物较深入的海区,调查资料丰富,已在50多处采集到天然气水合物样品,具备建立天然气水合物矿产资源预测模型的条件。选择34处已知天然气水合物矿点和34处已知无矿点作为训练区,建立矿点存在与否的预测模型。模型的相关系数值表明墨西哥湾天然气水合物与盐底辟关系密切。将该模型应用于整个墨西哥湾北部,初步获得了天然气水合物存在可能性概率分布图;概率大于0.7的预测单元包含已知矿点中的30个,利用该阈值圈定了墨西哥湾天然气水合物潜在资源分布区。

The Gulf of Mexico is a sea area of deeper study on gas hydrate in the world, plenty of data, and gas hydrate samples have been collected at more than 50 locations, so it possesses the condition to establish mineral resource prediction model about gas hydrate. 34 known ore occurrences and 34 non-ore occurrences were selected as a training area, and the prediction model of ore occurrence existence-probability was es- tablished. Correlation coefficients indicate that gas hydrate in the Gulf of Mexico has close relation to basin and salt diapir. Such model was used in the whole northern Gulf of Mexico and probability figures of gas hydrate distribution were gained; the prediction cell with probabilities greater than 0.5 contains 31 of known ore occurrences, and this probability value was used to delineate potential resource distribution area of gas hydrate in the Gulf of Mexico.