Accurate prediction of crop yield before harvest is critical to food security and importation. The calculated ten explanatory factors and autumn crop yield data were used as data sources in this research. Firstly, a Redundancy Analysis (RDA) was employed to carry out explanatory factors and feature selection. The simple effects of RDA were used to evaluate the interpretation rates of the explanatory factors. The conditional effects of RDA were adopted to select the features of the explanatory factors. Then, the autumn crop yield was divided into the training set and testing set with an 80/20 ratio, using Support Vector Regression (SVR), Random Forest Regression (RFR), and deep neural network (DNN) for the model, respectively. Finally, the coefficient of determination (R²), the root mean square error (RMSE), the mean absolute error (MAE), and the mean absolute percentage error (MAPE) were used to evaluate the performance of the model comprehensively. The results showed that the interpretation rates of the explanatory factors ranged from 54.3% to 85.0% (p = 0.002), which could reflect the autumn crop yields well. When a small number of sample training data (e.g., 80 samples) was used, the DNN model performed better than both SVR and RF models.

Traditional NDVI-T_s space is triangular or trapezoidal, but Liu et al. (2015) discovered that the NDVI-T_s space was bi-parabolic when the study area was covered with low biomass vegetation. Moreover, the numerical value of the indicator was considered in most of the study when the drought conditions in the space domain were evaluated. In addition, quantitatively assessing the spatial-temporal changes of the drought was not enough. In this study, first, we used MODIS NDVI and T_s data to reexamine if the NDVI-T_s space with “time” and a single pixel domain is bi-parabolic in the Shaanxi province of China, which is vegetated with low biomass to high biomass. This is compared with the triangular NDVI-T_s space and one of the well-known drought indexes called the temperature-vegetation index (TVX). The results demonstrated that dry and wet edges exhibited a parabolic shape again in scatter plots of T_s and NDVI in the Shaanxi province, which was linear in the triangular NDVI-T_s space. The Temperature Vegetation Dryness Index (TVDI_c) was obtained from bi-parabolic NDVI-T_s andTVDI_t was obtained from the triangular NDVI-T_s space and TVX were compared with 10-cm depth relative soil moisture. By estimating the 10-cm depth soil moisture, TVDI_c was better than TVDI_t, which were all apparently better than TVX. Second, combined with MODIS data, the drought conditions of the study area were assessed by TVDI_c between 2000 to 2016. Spatially, the drought in the Shaanxi Province between 2000 to 2016 were mainly distributed in the northwest, North Shaanxi, and the North and East Guanzhong plain. The drought area of the Shaanxi province accounted for 31.95% in 2000 and 27.65% in 2016, respectively. Third, we quantitatively evaluated the variation of the drought status by using Gradient-based Structural Similarity (GSSIM) methods. The area of the drought conditions significantly changed and moderately changed at 5.34% and 40.22%, respectively, between 2000 and 2016. Finally, the possible reasons for drought change were discussed. The change of precipitation, temperature, irrigation, destruction or betterment of vegetation, and the enlargement of opening mining, etc., can lead to the variations of drought.

This paper employs the Normalized Difference Water Index (NDWI), Automated Water Extraction Index (AWEI), Modified Normalized Difference Water Index (MNDWI) and Near-Infrared (NIR) threshold methods to extract the boundaries of Hongjiannao Lake, Zasak reservoir and Changjiagou reservoir. The real boundary of the water bodies was obtained by visual interpretation from high-resolution imagery. The Digital Shoreline Analysis System (DSAS) is further used to calculate the net shoreline movement (NSM) between the real lake boundary and the lake boundary extracted by the NDWI, AWEI, MNDWI and NIR threshold methods. We quantitatively evaluated the accuracy of each water body extraction method by NSM, which was competed with kappa coefficient (κ) and edge detection. The results showed that the average of the |NSM| of the NDWI, MNDWI, AWEI and NIR threshold methods are 12.77 m, 16.76 m, 28.65 m and 31.43 m, respectively. The κ is 0.9869, 0.9855, 0.9747 and 0.9736, respectively and the correct extracted (T) ratios for the edge detection are 84.06%, 82.68%, 56.8% and 53.52%, respectively. The mean value of |NSM| for NDWI is the smallest, while κ and T are the highest. It indicates that the accuracy of NDWI is the highest in Hongjiannao lake. The smaller the |NSM| is, the larger κ and the higher T are. This shows that the |NSM| is consistent with the commonly used accuracy verification method such as κ and edge detection. The results of water extraction from the other two reservoirs also support this conclusion. Therefore, the accuracy verification of NSM can reflect the spatial position information and has reliability. It provides a new approach in verifying the accuracy of water body extraction methods. The optimum water extraction index for different study areas is different. The NIR band threshold method and AWEI can accurately extract water boundaries with a small amount of aquatic vegetation. However, NDWI and MNDWI are more suitable for extracting water bodies located in complex terrains.