Spatiotemporal fusion (STF) is considered a feasible and cost-effective way to deal with the trade-off between the spatial and temporal resolution of satellite sensors, and to generate satellite images with high spatial and high temporal resolutions. This is achieved by fusing two types of satellite images, i.e., images with fine temporal but rough spatial resolution, and images with fine spatial but rough temporal resolution. Numerous STF methods have been proposed, however, it is still a challenge to predict both abrupt landcover change, and phenological change, accurately. Meanwhile, robustness to radiation differences between multi-source satellite images is crucial for the effective application of STF methods. Aiming to solve the abovementioned problems, in this paper we propose a hybrid deep learning-based STF method (HDLSFM). The method formulates a hybrid framework for robust fusion with phenological and landcover change information with minimal input requirements, and in which a nonlinear deep learning-based relative radiometric normalization, a deep learning-based superresolution, and a linear-based fusion are combined to address radiation differences between different types of satellite images, landcover, and phenological change prediction. Four comparative experiments using three popular STF methods, i.e., spatial and temporal adaptive reflectance fusion model (STARFM), flexible spatiotemporal data fusion (FSDAF), and Fit-FC, as benchmarks demonstrated the effectiveness of the HDLSFM in predicting phenological and landcover change. Meanwhile, HDLSFM is robust for radiation differences between different types of satellite images and the time interval between the prediction and base dates, which ensures its effectiveness in the generation of fused time-series data.

Spatiotemporal fusion is considered a feasible and cost-effective way to solve the trade-off between the spatial and temporal resolution of satellite sensors. Recently proposed learning-based spatiotemporal fusion methods can address the prediction of both phenological and land-cover change. In this paper, we propose a novel deep learning-based spatiotemporal data fusion method that uses a two-stream convolutional neural network. The method combines both forward and backward prediction to generate a target fine image, where temporal change-based and a spatial information-based mapping are simultaneously formed, addressing the prediction of both phenological and land-cover changes with better generalization ability and robustness. Comparative experimental results for the test datasets with phenological and land-cover changes verified the effectiveness of our method. Compared to existing learning-based spatiotemporal fusion methods, our method is more effective in predicting phenological change and directly reconstructing the prediction with complete spatial details without the need for auxiliary modulation.

The vulnerability of vegetation ecosystems and hydrological systems in high-altitude areas makes their phenology more sensitive and their response to climate change more intense. The Qilian Mountains, an important geographic unit located in the northeastern Tibetan Plateau (TP), has experienced the more significant increases in temperature and precipitation in the past few decades than most areas of the TP. However, under such intense climate change, the temporal and spatial differences in phenology in the Qilian Mountains are not clear. This study explored the spatial and temporal heterogeneity of phenology in the Qilian Mountains from 1982 to 2014 and its response to three temperature indicators, including the mean daily temperature (T_mean), mean daily daytime temperature (T_max), and mean daily nighttime temperature (T_min). The results showed that (1) as the altitude rose from southeast to northwest, the multiyear mean of the start of the growing season (SOS) was gradually delayed mainly from 120 to 190 days, the multiyear mean of the end of the growing season (EOS) as a whole was advanced (from 290 to 260 days), and the multiyear mean of the length of the growing season (LGS) was gradually shortened (from 150 to 80 days). (2) In general, there was an advanced trend in the annual average SOS (0.2 days per decade), a delayed trend in the annual average EOS (0.15 days per decade), and an extended trend in the annual average LGS (0.36 days per decade) over the study period. However, there has been no significant phenological trend in recent years, especially for the SOS after 2000 and the EOS and LGS after 2003. (3) Higher preseason temperatures led to an advanced SOS and a delayed EOS at the regional scale. Moreover, the SOS and EOS were more triggered by T_max than T_min and T_mean. The LGS was significantly positively correlated with annual mean temperature (r = −0.82, p < 0.01).