Search

Many countries have been constructing modern ground transportation projects. This raises questions about the impacts of such projects on development of impervious surfaces, yet there have been few attempts to systematically analyze these impacts. This paper attempts to narrow this information gap using the Hangzhou Bay Bridge project, China, as an exploratory case study. Using remotely sensed data, we developed a framework based on statistical techniques, wavelet multi-resolution analysis and Theil-Sen slope analysis to measure the changes in impervious surfaces. The derived changes were then linked to the bridge project with respect to socio-economic factors and land use development activities. The findings highlight that the analytical framework could reliably quantify the area, pattern and form of new urban area and urban intensification. Change detection analysis showed that urban area, GDP and the length of highways increased moderately in the pre-Hangzhou Bay Bridge period (1995–2002) while all of these variables increased more substantially during (2002–2009) and after (2009–2013) the bridge construction. The results indicate that the development of impervious surfaces due to new urban area came at the expense of permeable surfaces in the urban fringe and within rural regions, while urban intensification occurred mainly in the form of the redevelopment of older structures to modern high-rise buildings within existing urban regions. In the context of improved transportation infrastructure, our findings suggest that new urban area and urban intensification can be attributed to consecutive events which act like a chain reaction: construction of improved transportation projects, their impacts on land use development policies, effects of both systems on socio-economic variables, and finally all these changes influence new urban area and urban intensification. However, more research is needed to better understand this sequential process and to examine the broader applicability of the concept in other developing regions.

Large-scale crop mapping is of fundamental importance to tackle food security problems. SAR remote sensing has lately received great attention for crop type mapping due to its stability in the revisit cycle and is not hindered by cloud cover. However, most SAR image-classification studies focused on the application of backscattering characteristics with machine learning models, while few investigated the potential of the polarization decomposition and deep-learning models. This study investigated whether the radar polarization information mined by polarization decomposition, the patch strategy and the approaches for combining recurrent and convolutional neural networks (Conv2d + LSTM and ConvLSTM2d) could effectively improve the accuracy of crop type mapping. Sentinel-1 SLC and GRD products in 2020 were collected as data sources to extract VH, VV, VH/VV, VV + VH, Entropy, Anisotropy, and Alpha 7-dimensional features for classification. The results showed that the three-dimensional Convolutional Neural Network (Conv3d) was the best classifier with an accuracy and kappa up to 88.9% and 0.875, respectively, and the ConvLSTM2d and Conv2d + LSTM achieved the second and third position. Compared to backscatter coefficients, the polarization decomposition features could provide additional phase information for classification in the time dimension. The optimal patch size was 17, and the patch-based Conv3d outperformed the pixel-based Conv1d by 11.3% in accuracy and 0.128 in kappa. This study demonstrated the value of applying polarization decomposition features to deep-learning models and provided a strong technical support to efficient large-scale crop mapping. [ABSTRACT FROM AUTHOR]