H Chen, C Ning, Z Nan, et al. Correction of Daily Precipitation Data over the Qinghai-Tibetan Plateau with Machine Learning Models[J]. 2017, 39(3): 583—592.[陈浩，宁忱，南卓铜，等. 基于机器学习模型的青藏高原日降水数据的订正研究[J]. 冰川冻土. 2017, 39(3): 583—592.]
Zhao S, Nan Z*, Huang Y, Zhao L. The application and evaluation of simple permafrost distribution models on the Qinghai-Tibet Plateau. Permafrost and Periglacial Processes. 2017, 28(2): 391-404. DOI:10.1002/ppp.1939.
The performance of simple permafrost distribution models widely used on the Qinghai–Tibet Plateau (QTP) has not been fully evaluated. In this study, two empirical models (the elevation model and mean annual ground temperature model) and three semi-physical models (the surface frost number model, the temperature at the top of permafrost model and the Kudryavtsev model) were investigated. The simulation results from the models were compared to each other and validated against existing permafrost maps of the entire QTP and in three representative areas investigated in the field. The models generally overestimated permafrost distribution in the investigated areas, but they captured the broad characteristics of permafrost distribution on the entire QTP, and performed best in areas with colder, continuous permafrost. Large variations in performance occurred at elevations of 3800–4500 m asl and in areas with thermally unstable permafrost. The two empirical models performed best in areas where permafrost is strongly controlled by elevation, such as eastern QTP. In contrast, the three semi-physical models were better in southern island permafrost areas with relatively flat terrain, where local factors considerably impact the distribution of permafrost. Model performance could be enhanced by explicitly considering the effects of elevation zones and regional conditions.
1. Wu X, Nan Z.A multilayer soil texture dataset for permafrost modeling over Qinghai-Tibetan Plateau.In Proceeding of 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS),2016. 4917-4920. (wu et al. 2016, igarss )
2. Wang Y, Nan Z*, Chen H, Wu X.Correction of daily precipitation data of ITPCAS dataset over the Qinghai-Tibetan Plateau with KNN model.In Proceeding of 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS),2016. 593-596. (wang et al. 2016, igarss)
3. Ning C, Wang Y, Nan Z*, Chen H, Liu C.Study on correction of daily precipitation data of the Qinghai-Tibetan plateau with machine learning models.In Proceeding of 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS),2016. 517-520. (ning et al. 2016, igarss)