Sentinel-1A C-SAR and Sentinel-2A MultiSpectral Instrument (MSI) provide data applicable to the remote identification of crop type. In this study, six crop types (beans, beetroot, grass, maize, potato, and winter wheat) were identified using five C-SAR images and one MSI image acquired during the 2016 growing season. To assess the potential for accurate crop classification with existing supervised learning models, the four different approaches namely kernel-based extreme learning machine (KELM), multilayer feedforward neural networks, random forests, and support vector machine were compared. Algorithm hyperparameters were tuned using Bayesian optimization. Overall, KELM yielded the highest performance, achieving an overall classification accuracy of 96.8%. Evaluation of the sensitivity of classification models and relative importance of data types using data-based sensitivity analysis showed that the set of VV polarization data acquired on 24 July (Sentinel-1A) and band 4 data (Sentinel-2A) had the greatest potential for use in crop classification.

Optical remote sensing is one of the most attractive options for generating crop cover maps because it enables computation of vegetation indices, which are useful for assessing the condition of vegetation. The Sentinel-2A Multispectral Instrument (MSI), which is a multispectral sensor with 13 bands covering the visible, near infrared and short-wave infrared (SWIR) wavelength regions, offers a vast number of vegetation indices. Spectral indices, which are combinations of spectral measurements at different wavelengths, have been used in the previous studies and they sometimes contributed to improve classification accuracies. In this study, 91 published spectral indices were calculated from the MSI data. Additionally, classification algorithms are essential for generating accurate maps and the random forests classifier is one of which possesses the five hyperparameters were applied. The improvements in classification accuracies were confirmed achieving an overall accuracy of 93.1% based on the reflectance at 4 bands and 8 spectral indices.

The classification maps are required for the management and the estimation of agricultural disaster compensation; however, those techniques have yet to be established. Some supervised learning models may allow accurate classification. In this study, the Random Forest (RF) classifier and the classification and regression tree (CART) were applied to evaluate the potential of multi-temporal TerraSAR-X dual-polarimetric data, on the StripMap mode, for the classification of crop type. Furthermore, comparisons of the two algorithms and polarizations were carried out. In the study area, beans, beet, grasslands, maize, potato and winter wheat were cultivated, and these crop types were classified using the data set acquired in 2009. The classification results of RF were superior to those of CART, and the overall accuracies were 0.91–0.93.

In this paper, an Urban Light Index (ULI) is constructed to facilitate analysis and quantitative evaluation of the process of urbanization and expansion rate by using DMSP/OLS Nighttime Light Data during the years from 1992 to 2010. A unit circle urbanization evaluation model is established to perform a comprehensive analysis of the urbanization process of 34 prefecture-level cities in Northeast China. Furthermore, the concept of urban light space is put forward. In this study, urban light space is divided into four types: the core urban area, the transition zone between urban and suburban areas, suburban area and fluorescent space. Proceeding from the temporal and spatial variation of the four types of light space, the pattern of morphologic change and space-time evolution of the four principal cities in Northeast China (Harbin, Changchun, Shenyang, Dalian) is analyzed and given particular attention. Through a correlation analysis between ULI and the traditional urbanization indexes (urban population, proportion of the secondary and tertiary industries in the regional GDP and the built-up area), the advantages and disadvantages as well as the feasibility of using the ULI in the study of urbanization are evaluated. The research results show that ULI has a strong correlation with urban built-up area (R² ＝ 0.8277). The morphologic change and history of the evolving urban light space can truly reflect the characteristics of urban sprawl. The results also indicate that DMSP/OLS Nighttime Light Data is applicable for extracting urban space information and has strong potential to urbanization research.

With China as the study area, MODIS MOD14A1 and MCD12Q1 products were used to derive daily crop residue burning spots from 2014 to 2015. After vectorization of crop residue burning pixels and with the use of fishnet, burning density distribution maps were eventually completed. Meanwhile, the daily air quality data from 150 cities in 2014 and 285 cities in 2015 were used to obtain daily and monthly PM2.5 distribution maps with the Kriging interpolation. The results indicate that crop residue burning occurs in a seasonal pattern, and its spatial distribution is closely related to farming activities. The annual PM2.5 in China decreased 11.81% from 2014 to 2015, and the distribution of PM2.5 in China's east and north is always higher than in China's west and south. Furthermore, the changes in PM2.5 exhibit a hysteresis after crop residue burning in summer and autumn-winter. Regarding summer crop residue burning in China's middle-east, the r between crop residue burning spots and PM2.5 is 0.6921 (P < 0.01) in 2014 and 0.5620 (P < 0.01) in 2015, while the correlation coefficient of autumn-winter crop residue burning in China's northeast is slightly lower with an r of 0.5670 (P < 0.01) in 2014 and 0.6213 (P < 0.01) in 2015. In autumn-winter, crop residue burning can induce evident PM2.5 increase in China's northeast, and that is more obvious than summer crop residue burning in China's middle-east. Furthermore, when data of summer and autumn-winter crop residue burning from 2014 to 2015 are compared, we can see that the change in number of crop residue burning spots significant changes PM2.5 in these regions. Both the summer and autumn-winter crop residue burning areas presented spatial consistency with high PM2.5. By contrast, the results from many aspects indicated that the crop residue burning in spring did not cause a notable change of PM2.5.

Carbon dioxide (CO₂) is the most important greenhouse gas (GHG) in the atmosphere and is the greatest contributor to global warming. CO₂ concentration data are usually obtained from ground observation stations or from a small number of satellites. Because of the limited number of observations and the short time series of satellite data, it is difficult to monitor CO₂ concentrations on regional or global scales for a long time. The use of the remote sensing data such as the Advanced Very High Resolution Radiometer (AVHRR) or Moderate Resolution Imaging Spectroradiometer (MODIS) data can overcome these problems, particularly in areas with low densities of CO₂ concentration watch stations. A model based on temperature (MOD11C3), vegetation cover (MOD13C2 and MOD15A2) and productivity (MOD17A2) of MODIS (which we have named the TVP model) was developed in the current study to assess CO₂ concentrations on a global scale. We assumed that CO₂ concentration from the Thermal And Near infrared Sensor for carbon Observation (TANSO) aboard the Greenhouse gases Observing SATellite (GOSAT) are the true values and we used these values to check the TVP model accuracy. The results indicate that the accuracy of the TVP model is different in different continents: the greatest Pearson’s correlation coefficient (R²) was 0.75 in Eurasia (RMSE = 1.16) and South America (RMSE = 1.17); the lowest R² was 0.57 in Australia (RMSE = 0.73). Compared with the TANSO-observed CO₂ concentration (XCO₂), we found that the accuracy throughout the World is between −2.56~3.14 ppm. Potential sources of TVP model uncertainties were also analyzed and identified.