Soil organic matter (SOM) and pH are essential soil fertility indictors of paddy soil in the middle-lower Yangtze Plain. Rapid, non-destructive and accurate determination of SOM and pH is vital to preventing soil degradation caused by inappropriate land management practices. Visible-near infrared (vis-NIR) spectroscopy with multivariate calibration can be used to effectively estimate soil properties. In this study, 523 soil samples were collected from paddy fields in the Yangtze Plain, China. Four machine learning approaches—partial least squares regression (PLSR), least squares-support vector machines (LS-SVM), extreme learning machines (ELM) and the Cubist regression model (Cubist)—were used to compare the prediction accuracy based on vis-NIR full bands and bands reduced using the genetic algorithm (GA). The coefficient of determination (R²), root mean square error (RMSE), and ratio of performance to inter-quartile distance (RPIQ) were used to assess the prediction accuracy. The ELM with GA reduced bands was the best model for SOM (SOM: R² = 0.81, RMSE = 5.17, RPIQ = 2.87) and pH (R² = 0.76, RMSE = 0.43, RPIQ = 2.15). The performance of the LS-SVM for pH prediction did not differ significantly between the model with GA (R² = 0.75, RMSE = 0.44, RPIQ = 2.08) and without GA (R² = 0.74, RMSE = 0.45, RPIQ = 2.07). Although a slight increase was observed when ELM were used for prediction of SOM and pH using reduced bands (SOM: R² = 0.81, RMSE = 5.17, RPIQ = 2.87; pH: R² = 0.76, RMSE = 0.43, RPIQ = 2.15) compared with full bands (R² = 0.81, RMSE = 5.18, RPIQ = 2.83; pH: R² = 0.76, RMSE = 0.45, RPIQ = 2.07), the number of wavelengths was greatly reduced (SOM: 201 to 44; pH: 201 to 32). Thus, the ELM coupled with reduced bands by GA is recommended for prediction of properties of paddy soil (SOM and pH) in the middle-lower Yangtze Plain.

Accurate monitoring of soil salinization plays a key role in the ecological security and sustainable agricultural development of arid regions. As a branch of artificial intelligence, machine learning acquires new knowledge through self-learning and continuously improves its own performance. The purpose of this study is to combine Sentinel-2 Multispectral Imager (MSI) data and MSI-derived covariates with measured soil salinity data and to apply three machine learning algorithms in modeling to estimate and map the soil salinity in the study sample area. According to the convenient transportation conditions, the study area and sampling quadrat were set up, and the 5-point method was used to collect the soil mixed samples, and 160 soil mixed samples were collected. Kennard–Stone (K–S) algorithm was used for sample classification, 70% for modeling and 30% for verification. The machine learning algorithm uses Support Vector Machines (SVM), Artificial Neural Network (ANN), and Random Forest (RF). The results showed that (1) the average reflectance of each band of the MSI data ranged from 0.21–0.28. According to the spectral characteristics corresponding to different soil electrical conductivity (EC) levels (1.07–79.6 dS m⁻¹), the spectral reflectance of salinized soil in the MSI data ranged from 0.09–0.35. (2) The correlation coefficient between the MSI data and MSI-derived covariates and soil EC was moderate, and the correlation between certain MSI data sets and soil EC was not significant. (3) The SVM soil EC estimation model established with the MSI data set attained a higher performance and accuracy (R² = 0.88, root mean square error (RMSE) = 4.89 dS m⁻¹, and ratio of the performance to the interquartile range (RPIQ) = 1.96, standard error of the laboratory measurements to the standard error of the predictions (SEL/SEP) = 1.11) than those attained with the soil EC estimation models established with the RF and ANN models. (4) We applied the SVM soil EC estimation model to map the soil salinity in the study area, which showed that the farmland with higher altitudes discharged a large amount of salt to the surroundings due to long-term irrigation, and the secondary salinization of the farmland also caused a large amount of salt accumulation. This research provides a scientific basis for the simulation of soil salinization scenarios in arid areas in the future.

The spatio-temporal distribution soil pH is critical for understanding the productivity and long-term sustainability of our agri-ecosystem. This study quantified the spatio-temporal distribution of paddy soil pH in Jiangxi province, China, and the potential driver of the change between 1980 and 2010. Data from the Soil Survey Information of Jiangxi province (1980s) and Jiangxi Soil Testing and Fertilizer Recommendation study (2010s) were collected and categorized into six pH ranges from strongly-acidic to strongly-alkaline with unit pH differences. Changes were calculated from the maps developed using the Pedological Knowledge base for 1980s data (without geolocation) and geostatistical methods for the 2010s data (geolocated). An overall 0.6-unit decrease and a major shift of soil pH from weakly-acidic (54% → 18%) to acidic (35% → 74%) was observed over the province in a scattered fashion with concentration in the central part and the Poyang Lake area. About half of the area under paddy cultivation went through acidification by at least one pH unit and 7% by at least 2 pH units, while 40% of the area remained unchanged. Excessive fertilizer application and acid-rain intensity contributed to the acidification. Thus, a more knowledge-based and comprehensive fertilizer management should be adopted to make paddy production sustainable in the province.

The coordination relationship between land urbanization and population urbanization is crucial for achieving sustainable development under economic transition. Moreover, the balance between land urbanization and population urbanization is essential to guarantee the urbanization process of an entire city. This paper empirically analyzes the interaction between land urbanization and population urbanization in Nanchang from 2002 to 2017 based on the coupling coordination model (CCM). The impacts of the coordination degree on coordinated development are quantified by multivariate linear regression (MLR). The results show the following: (1) The indices of land urbanization and population urbanization in Nanchang showed an upward trend, and therefore the coordination degree in Nanchang increased significantly from 2002 to 2017. (2) The coordinated development of urbanization underwent two stages: disordered and moderately disordered. (3) The urban population proportion and the supporting capability of agricultural production had a positive impact on coordinated development. Meanwhile, the results also show that per capita education expenditures and the per capita public green area had negative impacts on the coordination degree, while economic development and the urban industrial structure were positive contributors to the coordination degree. Finally, this paper proposes that policies should be formulated to achieve coordinated development of urbanization. It can be concluded that the results regarding coordinated development of urbanization can help decision makers formulate effective measures to achieve coordinated development in the future.

High‐quality precipitation data are vital for hydrological applications and climate change research. In this study, we evaluated two satellite‐based precipitation products from Chinese Fengyun (FY) 2G (G is one of the third batch of operational geostationary satellites in the Chinese FY‐2 series) and the Global Precipitation Measurement (GPM) against the rain gauge‐based precipitation data, respectively, over the Yunnan‐Kweichow Plateau, China, at different temporal scales (e.g., monthly, daily, and hourly). And the main findings of this study were as follows: (1) FY2G Quantitative Precipitation Estimation (QPE) and Integrated Multisatellite Retrievals for GPM (IMERG) shared similar spatial precipitation patterns with those of rain gauge data, while the FY2G QPE general underestimated the precipitation (bias, −30%–0%), and IMERG obviously overestimated the precipitation (bias, 0%–60%) at monthly scale over the Yunnan‐Kweichow Plateau; (2) the FY2G QPE correlated significantly better (Correlation coefficient, CC, ~0.80) than IMERG (CC ~0.20) against rain gauge data at daily scale, meanwhile, the average daily bias value of IMERG (~30%) was around 6 times larger than that of FY2G QPE (~5%); (3) The FY2G QPE (CC ~0.7, bias ~ − 8%, root‐mean‐square error ~2.0 mm/hr, mean absolute error ~ 0.6 mm/hr) also generally outperformed the IMERG (CC ~0.1, bias ~13%, root‐mean‐square error ~3.0 mm/hr, mean absolute error ~ 1.0 mm/hr), at hourly scale; (4) The FY2G QPE significantly underestimated the precipitation in the period between 16:00 and 20:00 with the probability of detection (POD) decreased from 0.8 to 0.5, while the IMERG significantly overestimated the precipitation in the period between 4:00 and 12:00, at diurnal scale; and (5) one of the reasons resulting the significant overestimations of IMERG might due to its weak abilities in detecting the precipitation events with POD ~0.2 and false alarm ratio (FAR) ~0.7 at diurnal scale, which performed worse than those of the FY2G QPE (POD ~0.8 and FAR ~0.4). These findings would provide valuable recommendations for using these satellite‐based precipitation products in various application fields, such as hydrology, agriculture, and meteorology over the YKP, and also for improving the algorithms of GPM.