The reconstruction of the Normalized Difference Vegetation Index (NDVI) is a crucial prerequisite for numerous spatiotemporal continuous studies. To address the limitations posed by satellite temporal resolution and challenging atmospheric conditions, the combination of Synthetic Aperture Radar (SAR) and optical images from diverse sources has proven to be effective and widely employed. In this study, we employ the Spatial-Temporal Savitzky-Golay algorithm to rectify MODIS NDVI maps and eliminate interruptions caused by noise. Random Forest and Gradient Boosting Decision Trees serve as a dual filter to select SAR indices with the highest impact on NDVI reconstruction, ensuring that the chosen indices encapsulate the most valuable information. Subsequently, we conduct a series of ablation experiments and develop a deep learning network named Residual-in-Residual Dense Block NDVI Reconstruction net (RDNRnet). This network effectively mitigates the impacts of MODIS coarse resolution and speckle noises in SAR data. We also evaluate the network performance in reconstructing NDVI across all seasons and land cover types. Our findings highlight that the modified dual-polarimetric SAR vegetation index and the standard deviation of the VV band are the most crucial SAR indices. The predictions for summer exhibit the highest performance, with a coefficient of determination (R2) reaching 0.9757. Optimal performances by land cover type are observed in forests, paddy fields, and dry farming fields, all with R2values exceeding 0.9580. Our adaptive NDVI reconstruction solution demonstrates robust performance across different data availability scenarios, effectively catering to all seasons and land cover types.

Rivers are important elements of the earth’s ecosystem, and their spatial distribution information is critical for the study of hydrological and biogeochemical processes. Moderate Resolution Imaging Spectroradiometer (MODIS) imagery has been widely used for river mapping due to its high temporal resolution and long-term observation records, which are essential to capture the rapid fluctuation of rivers. However, when the conventional hard classification methods are used, the accuracy of the river maps produced from MODIS data (and especially for those rivers with narrow widths) is often limited because mixed pixels are common in MODIS imagery, due to coarse spatial resolution. In this article, a cascaded spectral–spatial information combined deep convolutional neural network (CNN) model for super-resolution river mapping (DeepRivSRM) is proposed to produce Landsat-like fine-resolution river maps from MODIS images. In DeepRivSRM, a CNN made up of a spectral unmixing module and a super-resolution mapping (SRM) module is introduced to handle the spectral and spatial information simultaneously. Moreover, for training the DeepRivSRM model, an adaptive cross-entropy loss function incorporating the fraction information of the rivers is designed to improve the performance of the DeepRivSRM model for small rivers. The proposed method was evaluated with MODIS images from three test sites and was compared with hard classification, a conventional SRM method, and a CNN-based SRM method. The results show that DeepRivSRM can generate more accurate river maps by effectively learning the subpixel-scale spatial–spectral information in MODIS imagery.

We study how to learn a semantic parser of state-of-the-art accuracy with less supervised training data.We conduct our study on WikiSQL, the largest hand-annotated semantic parsing dataset to date.First, we demonstrate that question generation is an effective method that empowers us to learn a state-ofthe-art neural network based semantic parser with thirty percent of the supervised training data.Second, we show that applying question generation to the full supervised training data further improves the state-of-the-art model.In addition, we observe that there is a logarithmic relationship between the accuracy of a semantic parser and the amount of training data.

In this study, the aboveground biomass (AGB) was estimated for the forest in Xishuangbanna (Southwestern China) area by combining forest field investigation data and Geoscience Laser Altimeter System (GLAS) data, onboard the Ice, Cloud and land Elevation Satellite (ICESat). The tree height prediction models at GLAS footprint level were developed from GLAS waveform variables and filed survey plot tree height using multiple linear stepwise regression method, and then were used to estimate the biomass for all GLAS footprints of forest in the study area according the allometric equations deduced by the fixed field sites. The AGB estimate result was in a reasonable range after compared with the forest inventory plot data.

In this study, the aboveground biomass (AGB) was estimated for the forest in Daxinganlin (Northeastern China) area by combining CCD data from Small Satellite for Disaster and Environment Monitoring and Forecast (HJ-1) and Geoscience Laser Altimeter System (GLAS) data, onboard the Ice, Cloud and land Elevation Satellite (ICESat). The AGB biomass prediction models at GLAS footprint level were developed from GLAS waveform variables and filed survey plot biomass in Changqing (CQ) forest farm, and then were used to estimate the biomass for all GLAS footprints of forest in the study area. The footprint forest biomass and the spectral reflectivity parameters and vegetation indices from HJ-1 satellites CCD data were then used to establish biomass prediction model to predict forest AGB in the entire study site using multiple linear regression method. Compared with the Zhuanglin (ZL) field investigated data, the mean error is 5.74 ton/ha.

Interferometric SAR (InSAR)data have been used to measure canopy height. Polarimetric interferometric SAR (PolInSAR) data can be used to derive canopy height without using ground surface elevation data. But in most cases, only single polarization InSAR data are available and the elevation of ground surface in the forested areas is needed to get the height of the scattering phase center. On contrary, the elevation of canopy surface is relatively easy to obtain by Stereo imagery. In this study the feasibility of the estimation of forest height using SRTM-DEM and ASTER- GDEM was investigated. The ASTER-GDEM was firstly resampled to the pixel size of SRTM-DEM (3 arc-second) and then was registered to SRTM- DEM using the points selected from their aspect maps. The results showed that the registration is necessary because the geolocation error at east-west direction is about half of the pixel size. The relationship between the forest height and the elevation difference was analyzed. The results showed that the elevation difference between registered ASTER-GDEM and SRTM-DEM is positively correlated with the forest height. Although there are some problems when the terrain is rough, it provides us a way to estimate the height of mature forest in flat terrain.

Synthetic Aperture Radar is the only instrument that can provide consistent remote sensing data for south China with persistent cloud cover and rain. In this paper, the potential of multi-temporal ERS, ENVISAT-1 and PALSAR data for land cover/land use classification and urban change detection was investigated at a test area in Fuzhou city, the capital of Fujian province in southeastern China. Both SAR backscatter intensity characteristics and interferometric data were analyzed. Two methods were carried out in the urban dynamics change detection: one was post-classification comparison, and the other one was multi-temporal image ratioing. The results from both classification and urban change detection were validated by field survey data and showed promising application of PALSAR and ESA SAR data in southeastern China.