Deforestation has become a significant contributing factor to climate change\nand, due to this, both classifying the drivers and predicting segmentation maps\nof deforestation has attracted significant interest. In this work, we develop a\nrotation equivariant convolutional neural network model to predict the drivers\nand generate segmentation maps of deforestation events from Landsat 8 satellite\nimages. This outperforms previous methods in classifying the drivers and\npredicting the segmentation map of deforestation, offering a 9% improvement in\nclassification accuracy and a 7% improvement in segmentation map accuracy. In\naddition, this method predicts stable segmentation maps under rotation of the\ninput image, which ensures that predicted regions of deforestation are not\ndependent upon the rotational orientation of the satellite.\n