The land surface temperature (LST), defined as the radiative skin temperature of the ground, plays a critical role in land surface systems, from the regional to the global scale. The commonly utilized daily Moderate Resolution Imaging Spectroradiometer (MODIS) LST product at a resolution of one kilometer often contains missing values attributable to atmospheric influences. Reconstructing these missing values and obtaining a spatially complete LST is of great research significance. However, most existing methods are tailored for reconstructing clear-sky LST rather than the more realistic cloudy-sky LST, and their computational processes are relatively complex. Therefore, this paper proposes a simple and effective real LST reconstruction method combining Thermal Infrared and Microwave Remote Sensing Based on Temperature Conservation (TMTC). TMTC first fills the microwave data gaps and then downscales the microwave data by using MODIS LST and auxiliary data. This method maintains the temperature of the resulting LST and microwave LST on the microwave pixel scale. The average Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R² of TMTC were 3.14 K, 4.10 K, and 0.88 for the daytime and 2.34 K, 3.20 K, and 0.90 for the nighttime, respectively. The ideal MAE of the TMTC method exhibits less than 1.5 K during daylight hours and less than 1 K at night, but the accuracy of the method is currently limited by the inversion accuracy of microwave LST and whether different LST products have undergone time normalization. Additionally, the TMTC method has spatial generality. This article establishes the groundwork for future investigations in diverse disciplines that necessitate real LSTs.

Land subsidence prediction remains challenging. Conventional grid-based or sequence-only neural networks struggle to represent these spatial dependencies and often lack structured mechanisms for incorporating region-level deformation priors and local physical consistency. This study develops PEGNet, a Peridynamics-inspired and emergent-feature-conditioned spatio-temporal graph neural network for land subsidence prediction. PEGNet uses a per-node GRU for temporal encoding and an EF-conditioned GAT for spatial aggregation. At the node level, historical InSAR deformation and auxiliary hydroclimatic observations are augmented with a PD-derived state and an EF neighborhood context, allowing a per-node GRU to learn temporal evolution and produce a base estimate of future incremental deformation. At the edge level, a fixed spatial graph supports EF-conditioned graph attention, where edge representations combine spatial distance, PD-derived bond-strain information, and the EF same/cross-region gate to generate a spatial residual prediction. At the training and prediction level, the temporal and spatial branches are fused through a learned gate, cumulative deformation is reconstructed from the last observed value, and training combines a data-fitting loss with a PD-inspired local consistency term weighted by distance and EF relations. EF therefore conditions both the network and the coupled regularizer, although it introduces no separate loss term, and the PD component remains a local consistency mechanism rather than a complete Peridynamic solver. PEGNet is evaluated using 60 months of PS-InSAR observations in Tongzhou District, Beijing, four overlapping purged test windows covering nine unique target months and five random seeds. It achieves an incremental-deformation RMSE of 2.153 ± 0.007 mm and a reconstructed cumulative-deformation RMSE of 2.631 ± 0.072 mm. These results outperform the deterministic baselines and remain comparable to the unconstrained GRU-GAT model. Mechanism diagnostics indicate that EF mainly redistributes graph attention, whereas the PD components improve local consistency without producing a substantial global accuracy gain. Overall, the comparable aggregate performance and improved local consistency support the use of PEGNet for conditional subsidence monitoring and scenario analysis.