By effectively observing the land surface and obtaining farmland conditions, satellite remote sensing has played an essential role in agricultural drought monitoring over past decades. Among all remote sensing techniques, optical and thermal remote sensing have the most extended history of being utilized in drought monitoring. The primary goal of this paper is to illustrate how optical and thermal remote sensing have been and will be applied in the monitoring, assessment, and prediction of agricultural drought. We group the methods into four categories: optical, thermal, optical and thermal, and multi-source. For each category, a concise explanation is given to show the inherent mechanisms. We pay special attention to solar-induced chlorophyll fluorescence, which has great potential in early drought detection. Finally, we look at the future directions of agricultural drought monitoring, including (1) early detection; (2) spatio-temporal resolution; (3) organic combination of multi-source data; and (4) smart prediction and assessment based on deep learning and cloud computing.

The Landsat time-series dataset is one of the most widely used datasets for land surface research due to its long time-series and Land Surface Reflectance (LSR) product. Though the United States Geological Survey (USGS) provides Landsat LSR products for later Landsat 4–5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI), no early Landsat 1–5 Multispectral Scanner System (MSS) LSR product is generated currently, limiting the research traced back to the 1970s. Atmospheric correction is one of the necessary preprocesses for generating LSR products. However, it is challenging for MSS images, not only because the image quality is lower and bands are different compared with the current sensors, but also because of the multiple effects of other preprocesses, such as radiometric calibration. Based on the Second Simulation of a Satellite Signal in the Solar Spectrum Vector (6SV) model, we propose a novel framework for generating Landsat 1–5 MSS LSR data of China. Ground-based visibility records are introduced to replace the images-based aerosol optical depth (AOD) to effectively generate MSS LSR data of the 1970s. We evaluate the generated MSS LSR data by the cross-validation of the simultaneous observation of MSS and TM sensors in Landsat 4 and Landsat 5 using Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) surface reflectance product as the truth value. The evaluation result shows that the generated MSS LSR data is comparable with the later Landsat TM LSR product, with slightly larger uncertainties. In addition, it shows that the non-atmospheric factors (e.g., the difference of relative spectral responses of TM and MSS, the georegistration errors, the radiometric calibration uncertainty, and image noises) bring larger uncertainties than the atmospheric factors (e.g., the AOD retrieval method by visibility) to the cross-validation results. We apply the MSS LSR data generated by the proposed framework on time series analysis in the regions of interest (ROIs) of the spectral-stable land cover in China for all the MSS sensors. The application demonstrates the potential and promise of the MSS LSR data generated by the proposed framework.