The comprehensive understanding of the occurred changes of permafrost, including the changes of mean annual ground temperature (MAGT) and active layer thickness (ALT), on the Qinghai‐Tibet Plateau (QTP) is critical to project permafrost changes due to climate change. Here, we use statistical and machine learning (ML) modeling approaches to simulate the present and future changes of MAGT and ALT in the permafrost regions of the QTP. The results show that the combination of statistical and ML method is reliable to simulate the MAGT and ALT, with the root‐mean‐square error of 0.53°C and 0.69 m for the MAGT and ALT, respectively. The results show that the present (2000–2015) permafrost area on the QTP is 1.04 × 10⁶ km² (0.80–1.28 × 10⁶ km²), and the average MAGT and ALT are −1.35 ± 0.42°C and 2.3 ± 0.60 m, respectively. According to the classification system of permafrost stability, 37.3% of the QTP permafrost is suffering from the risk of disappearance. In the future (2061–2080), the near‐surface permafrost area will shrink significantly under different Representative Concentration Pathway scenarios (RCPs). It is predicted that the permafrost area will be reduced to 42% of the present area under RCP8.5. Overall, the future changes of MAGT and ALT are pronounced and region‐specific. As a result, the combined statistical method with ML requires less parameters and input variables for simulation permafrost thermal regimes and could present an efficient way to figure out the response of permafrost to climatic changes on the QTP.

There are several publications related to the soil organic carbon (SOC) on the Qinghai-Tibetan Plateau (QTP). However, most of these reports were from different parts of the plateau with various sampling depth. Here, we present the results from a systematic sampling and analysis of 200 soil pits. Most of the pits were deeper than 2 m from an east-west transect across the plateau. The SOC and total nitrogen (TN) pools of the 148 × 10⁴ km², the area of the permafrost zone, for the upper 2 m soils calculated from the vegetation map were estimated to be 17.07 Pg (interquartile range: 11.34–25.33 Pg) and 1.72 Pg (interquartile range: 1.08–2.06 Pg), respectively. We also predicted the distribution of land cover types in 2050 and 2070 using decision tree rules and climate scenarios, and then predicted SOC and TN pools of this region. The results suggested that the SOC and TN pools will decrease in the future. The results not only contribute to the carbon and nitrogen storage and stocks in the permafrost regions as a whole but most importantly, to our knowledge of the possible changes of C and N storage on the QTP in the future.

Permafrost warming leads to greenhouse gas release to the atmosphere, resulting in a positive feedback to climate change. Earth system models indicate that more than 80% of the near‐surface permafrost is projected to disappear by the end of this century, but with a high degree of uncertainty. Here, we apply the Stefan solution to estimate permafrost degradation under future emission scenarios. We find that the most severe future scenario is likely to lead to only a 14% decrease in area extent of the near‐surface permafrost at 3.5 m depth, and an area extent decrease of 1.3% at a depth of 6.0 m. Relative to active layer thickness increases from historical simulations, we find a less than 30% deepening for most permafrost regions by the end of this century. These results imply that the Stefan solution provides near‐surface permafrost area extent degradation estimates that are substantially lower than directly projected by models.