The rising machine learning (ML) models have become the preferred way for landslide detection based on remote sensing images, but the performance of these models in a sample-free area are rarely concerned in many studies. In this study, we used a cross-validation method (training model in one area and validation in another) to compare the model portability of trained ML models applied in an “off-site” area, as a consideration of the landslide detection ability of these models in sample-free areas. We integrate nighttime light imagery, multi-seasonal optical Landsat time-series and digital elevation data, and we employed support vector machines (SVM), artificial neural networks (ANN) and random forest (RF) models to classify the satellite imagery and identify landslides. Samples of two scenarios generated from two subareas of the Jiuzhaigou disaster-stricken region are used for the cross-application and accuracy evaluation of three ML models. The results revealed that when the trained models are applied in areas outside those in which they were developed, the landslide identification accuracy of these three models has declined. Especially for the SVM and ANN models, the accuracy is greatly reduced and there appears a seriously imbalanced user’s and producer’s accuracy. However, although the performance of the RF model is lower than that of SVM and ANN models in their local area, the RF model exhibits stable portability, and retains the original performance and achieves a satisfactory balance between overestimation and underestimation in “off-site” areas. An additional validation from a new area proved that the landslide detection performance of the RF model with stable portability is higher than that of the SVM and ANN models in “off-site” areas. The results suggest that evaluating the model portability through cross-application can be a useful way to determine the most suitable model for landslide detection in “off-site” areas with a similar geographic environment to model development areas, so as to maximize the accuracy of landslide detection based on limited samples.

Landslide dam, a common disaster around the world, always forms landslide lake because of blocking the river. Landslide lake is always with potentially immense hazard from rapid release of a large of water masses. Nowadays, with the development of remote sense technology, remote sensing big data has been widely applied to disaster monitor, which has broad prospect in landslide lake monitor. This study provided the processing flow of remote sensing image and flow of technology system in the landslide lake monitor based on remote sensing big data, which have been successfully applied to Baige landslide lake monitor. The result shows remote sensing technology as an objective, fleet and dynamic change of landslide lake with large monitor range and meets the requirement of all-weather dynamic monitor of landslide lake, which will provide basic information of landslide lake for decisionmaking departments to make disaster prevention and reduction. Lastly, the research points out the disadvantage of landslide lake monitor based on remote sensing big data. Drone images and aerial images will be an important supplement to remote sensing big data for landslide lake monitor.