In current upscaling of in situ surface soil moisture practices, commonly used novel statistical or machine learning-based regression models combined with remote sensing data show some advantages in accurately capturing the satellite footprint scale of specific local or regional surface soil moisture. However, the performance of most models is largely determined by the size of the training data and the limited generalization ability to accomplish correlation extraction in regression models, which are unsuitable for larger scale practices. In this paper, a deep learning model was proposed to estimate soil moisture on a national scale. The deep learning model has the advantage of representing nonlinearities and modeling complex relationships from large-scale data. To illustrate the deep learning model for soil moisture estimation, the croplands of China were selected as the study area, and four years of Visible Infrared Imaging Radiometer Suite (VIIRS) raw data records (RDR) were used as input parameters, then the models were trained and soil moisture estimates were obtained. Results demonstrate that the estimated models captured the complex relationship between the remote sensing variables and in situ surface soil moisture with an adjusted coefficient of determination of ${\overline{\mathrm{R}}}^2$ = 0.9875 and a root mean square error (RMSE) of 0.0084 in China. These results were more accurate than the Soil Moisture Active Passive (SMAP) active radar soil moisture products and the Global Land data assimilation system (GLDAS) 0–10 cm depth soil moisture data. Our study suggests that deep learning model have potential for operational applications of upscaling in situ surface soil moisture data at the national scale.

The vegetation supply water index (VSWI = NDVI/LST) is an effective metric estimating soil moisture in areas with moderate to dense vegetation cover. However, the normalized difference vegetation index (NDVI) exhibits a long water stress lag and the land surface temperature (LST), sensitive to water stress, does not contribute considerably to surface soil moisture monitoring due to the constraints of the mathematical characteristics of VSWI: LST influences VSWI less when LST value is sufficiently high. This paper mathematically analyzes the characteristics of VSWI and proposes a new operational dryness index (surface water content temperature index, SWCTI) for the assessment of surface soil moisture status. SWCTI uses the surface water content index (SWCI), which provides a more accurate estimation of surface soil moisture than that of NDVI, as the numerator and the modified surface temperature, which has a greater influence on SWCTI than that of LST, as the denominator. The validation work includes comparison of SWCTI with in situ soil moisture and other remote sensing indices. The results show SWCTI demonstrates the highest correlation with in situ soil moisture; the highest correlation R = 0.801 is found between SWCTI and the 0–5 cm soil moisture in a sandy loam. SWCTI is a functional and effective method that has a great potential in surface soil moisture monitoring.

Solar-induced chlorophyll fluorescence (SIF) has been shown to be a powerful proxy for photosynthesis and a promising indicator of drought monitoring, but the ability of high-resolution satellite-derived SIF for drought monitoring has not been widely investigated due to a lack of data. The lack of high spatiotemporal resolution satellite SIF hinders the resolution enhancement of SIF derived by downscaling or reconstruction algorithms. The TROPOspheric Monitoring Instrument (TROPOMI) SIF provides an alternative with finer spatiotemporal resolution. We present an operational downscaling method to generate 500 m 16-day SIF (TSIF) using Neural Networks over a local spatiotemporal window. The results showed that our method is very robust against overfitting, and TSIF has a strong spatiotemporal consistency with TROPOMI SIF (TROPOSIF) with ${\mathrm{R}}^2=0.956$ and $\mathrm{RMSE}=0.054{\mathrm{mWm}}^{-2}{\mathrm{sr}}^{-1}{\mathrm{nm}}^{-1}$. Comparison with another SIF product (CASIF) showed a spatiotemporal consistency with TSIF. Comparison with tower gross primary productivity (GPP) from AmeriFlux in California showed a strong correlation with ${\mathrm{R}}^2$ for multiple ecosystems ranging from 0.58 to 0.88. We explored the capacity of TSIF for monitoring a drought event in Henan, China, showing that TSIF is more sensitive to drought and precipitation compared to the Enhanced Vegetation Index. Our TSIF is a very promising indicator for regional drought monitoring.

Due to the inconsistent spatiotemporal spectral scales, a remote sensing dataset over a large-scale area and over long-term time series will have large variations and large statistical distribution features, which will lead to a performance drop of the deep learning model that is only trained on the source domain. For building an extraction task, deep learning methods perform weak generalization from the source domain to the other domain. To solve the problem, we propose a Capsule–Encoder–Decoder model. We use a vector named capsule to store the characteristics of the building and its parts. In our work, the encoder extracts capsules from remote sensing images. Capsules contain the information of the buildings’ parts. Additionally, the decoder calculates the relationship between the target building and its parts. The decoder corrects the buildings’ distribution and up-samples them to extract target buildings. Using remote sensing images in the lower Yellow River as the source dataset, building extraction experiments were trained on both our method and the mainstream methods. Compared with the mainstream methods on the source dataset, our method achieves convergence faster, and our method shows higher accuracy. Significantly, without fine tuning, our method can reduce the error rates of building extraction results on an almost unfamiliar dataset. The building parts’ distribution in capsules has high-level semantic information, and capsules can describe the characteristics of buildings more comprehensively, which are more explanatory. The results prove that our method can not only effectively extract buildings but also perform great generalization from the source remote sensing dataset to another.