Objective：Land surface temperature (LST) is a key parameter in the physical process of surface radiant energy balance and water cycle. Obtaining LST data accurately and timely, and mastering its temporal and spatial changes are of great significance to climate change research. Thermal infrared(TIR) measurements are limited in practical applications due to cloud cover and other effects. Passive microwave (PMW) remote sensing measurements can penetrate clouds and are less affected by atmospheric interference, which has the advantage of obtaining all-weather surface radiation information. Among them, the microwave remote sensing data Advanced Microwave Scanning Radiometer for EOS(AMSR-E) can obtain all-weather LST, which can be used as a supplement to the missing LST information in thermal infrared (TIR) products under cloudy conditions. However, the AMSR-E data has the problem of lack of information due to the satellite orbit scanning gap of its own sensor, which causes the obtained AMSR-E LST data to be greatly restricted in practical applications. Therefore, it is necessary to propose an effective method for solving the problem. Method：Based on the superiority of deep learning in solving non-linear problems and the high dynamic variability of LST, this paper proposes a multi-temporal feature-connected convolutional neural network (MTFC-CNN) which uses specific input combinations of multi-temporal information and spatial fusion units. The network structure is based on the characteristics of the temporal and spatial distribution of missing track gaps in AMSR-E LST data and the reconstruction of missing LST values is carried out from the timing information. Result：In the simulation experiment, the 2010 annual data was divided into 8 data subsets in four seasons, day and night. The average root mean square error of the reconstructed LST value is about 1.0k and the coefficient of determination R2 is above 0.88. Compared with the other two methods: spline interpolation (Spline) and time multiple linear regression (Regress), the reconstruction effect of MTFC-CNN method performs better regardless of seasons or day and night, which proves that MTFC-CNN is better than Spline and Regress methods at mining the characteristics of temporal and spatial changes in LST. In real experiments, through comparison with MODIS LST products, the LST value reconstructed at the missing area is basically consistent with it at other areas in temporal and spatial distribution. The reconstruction results show that the LST in mainland China region shows a gradual increasing trend from January to July, and a gradual decreasing trend from August to December. Which is basically consistent with the temperature changes in the four seasons. The change of LST during the day is more significant than that at night. In summer, the temperature in Northwest China is significantly higher than that in other regions. In winter, the temperature in Northeast China is generally lower than that in other regions. At night, the difference between summer and winter is more obvious. The difference in LST changes at night in autumn is relatively close. Conclusion：The experimental results show that the MTFC-CNN method proposed in this paper mines the spatio-temporal variation information of LSTs more effectively than two traditional methods, and achieves better results in reconstructing the orbital gap ah-missing of AMSR-E LST data. It provides the possibility for the reconstruction of missing information from TIR LST data under the cloud. Keywords: Land surface temperature; AMSR-E LST; Reconstruction; Deep learning; MTFC-CNN;