During the past decades, multiple remote sensing data sources, including nighttime light images, high spatial resolution multispectral satellite images, unmanned drone images, and hyperspectral images, among many others, have provided fresh opportunities to examine the dynamics of urban landscapes. In the meantime, the rapid development of telecommunications and mobile technology, alongside the emergence of online search engines and social media platforms with geotagging technology, has fundamentally changed how human activities and the urban landscape are recorded and depicted. The combination of these two types of data sources results in explosive and mind-blowing discoveries in contemporary urban studies, especially for the purposes of sustainable urban planning and development. Urban scholars are now equipped with abundant data to examine many theoretical arguments that often result from limited and indirect observations and less-than-ideal controlled experiments. For the first time, urban scholars can model, simulate, and predict changes in the urban landscape using real-time data to produce the most realistic results, providing invaluable information for urban planners and governments to aim for a sustainable and healthy urban future. This current study reviews the development, current status, and future trajectory of urban studies facilitated by the advancement of remote sensing and spatial big data analytical technologies. The review attempts to serve as a bridge between the growing “big data” and modern urban study communities.

Soil organic carbon (SOC) is an important soil property that has profound impact on soil quality and plant growth. With 140 soil samples collected from Ebinur Lake Wetland National Nature Reserve, Xinjiang Uyghur Autonomous Region of China, this research evaluated the feasibility of visible/near infrared (VIS/NIR) spectroscopy data (350–2,500 nm) and simulated EO-1 Hyperion data to estimate SOC in arid wetland regions. Three machine learning algorithms including Ant Colony Optimization-interval Partial Least Squares (ACO-iPLS), Recursive Feature Elimination-Support Vector Machine (RF-SVM), and Random Forest (RF) were employed to select spectral features and further estimate SOC. Results indicated that the feature wavelengths pertaining to SOC were mainly within the ranges of 745–910 nm and 1,911–2,254 nm. The combination of RF-SVM and first derivative pre-processing produced the highest estimation accuracy with the optimal values of R_t (correlation coefficient of testing set), RMSE_t and RPD of 0.91, 0.27% and 2.41, respectively. The simulated EO-1 Hyperion data combined with Support Vector Machine (SVM) based recursive feature elimination algorithm produced the most accurate estimate of SOC content. For the testing set, R_t was 0.79, RMSE_t was 0.19%, and RPD was 1.61. This practice provides an efficient, low-cost approach with potentially high accuracy to estimate SOC contents and hence supports better management and protection strategies for desert wetland ecosystems.

This study investigates spatial dependence and mechanisms of regional development in Greater Beijing, China by employing spatial statistical techniques. We have detected positive, strengthening global spatial autocorrelation from 1978 to 2001, and found such strengthening is the result of newly formed/extended clusters in the area. The local analysis recognizes local regimes of two-tier urban-rural spatial structure at the beginning of the reform period. While the urban-rural divide was lessening due to the reform, a north-south divide has emerged because of local natural conditions and development trajectories. Regarding mechanisms of regional development, ordinary least squares analysis is constrained by the existence of significant spatial autocorrelation among spatial units. Analytical results reveal that an error spatial regression model is a more appropriate alternative due to possible mismatch between boundaries of the underlying spatial process and the spatial units where data are organised. In 1995 and 2001, the signs of all the regression coefficients remained the same for both OLS and spatial models. However, their magnitude and significance change. Specifically, foreign direct investment and fixed-asset investment became less influential in the spatial model, while local government spending emerged as more influential. Este estudio investiga la dependencia especial y los mecanismos de desarrollo regional en el Gran Beijing, China, empleando técnicas estadísticas espaciales. Hemos detectado una autocorrelación espacial global en aumento y positiva desde 1978 al 2001, y hallado que dicho aumento es el resultado de clusters formados/ampliados recientemente en el área. El análisis local reconoce regimenes locales de estructura espacial urbana-rural de dos niveles al inicio del periodo de reforma. Mientras que la separación urbana-rural fue disminuyendo con la reforma, ha aparecido una separación norte-sur por condiciones naturales locales y trayectorias de desarrollo. Respecto de los mecanismos de desarrollo regional, el análisis de mínimos cuadrados ordinario (OLS) se ve restringido por la existencia de una autocorrelación espacial significativa entre unidades espaciales. Los resultados de análisis revelan que un modelo de regresión del error espacial es una alternativa más apropiada debido a una posible disparidad entre los límites del proceso espacial subyacente y las unidades espaciales donde están organizados los datos. En 1995 y en 2001, los signos de todos los coeficientes de regresión permanecieron iguales para OLS y modelos espaciales. Sin embargo, su magnitud y significancia cambian. En particular, la inversión extranjera directa y la inversión en activos fijos pasaron a ser menos influyentes en el modelo espacial, mientras que el gasto del gobierno local apareció más influyente.