The urban heat island (UHI) effect exerts a great influence on the Earth’s environment and human health and has been the subject of considerable attention. Landscape patterns are among the most important factors relevant to surface UHIs (SUHIs); however, the relationship between SUHIs and landscape patterns is poorly understood over large areas. In this study, the surface UHI intensity (SUHII) is defined as the temperature difference between urban and suburban areas, and the landscape patterns are quantified by the urban-suburban differences in several typical landscape metrics (ΔLMs). Temperature and land-cover classification datasets based on satellite observations were applied to analyze the relationship between SUHII and ΔLMs in 332 cities/city agglomerations distributed in different climatic zones of China. The results indicate that SUHII and its correlations with ΔLMs are profoundly influenced by seasonal, diurnal, and climatic factors. The impacts of different land-cover types on SUHIs are different, and the landscape patterns of the built-up and vegetation (including forest, grassland, and cultivated land) classes have the most significant effects on SUHIs. The results of this study will help us to gain a deeper understanding of the relationship between the SUHI effect and landscape patterns.

The Local Climate Zone (LCZ) scheme provides researchers with a standard method to monitor the Urban Heat Island (UHI) effect and conduct temperature studies. How to generate reliable LCZ maps has therefore become a research focus. In recent years, researchers have attempted to use Landsat imagery to delineate LCZs and generate maps worldwide based on the World Urban Database and Access Portal Tools (WUDAPT). However, the mapping results obtained by the WUDAPT method are not satisfactory. In this paper, to generate more accurate LCZ maps, we propose a novel Convolutional Neural Network (CNN) model (namely, LCZ-CNN), which is designed to cope with the issues of LCZ classification using Landsat imagery. Furthermore, in this study, we applied the LCZ-CNN model to generate LCZ mapping results for China’s 32 major cities distributed in various climatic zones, achieving a significantly better accuracy than the traditional classification strategies and a satisfactory computational efficiency. The proposed LCZ-CNN model achieved satisfactory classification accuracies in all 32 cities, and the Overall Accuracies (OAs) of more than half of the cities were higher than 80%. We also designed a series of experiments to comprehensively analyze the proposed LCZ-CNN model, with regard to the transferability of the network and the effectiveness of multi-seasonal information. It was found that the first convolutional stage, corresponding to low-level features, shows better transferability than the second and third convolutional stages, which extract high-level and more image- or task-oriented features. It was also confirmed that the multi-seasonal information can improve the accuracy of LCZ classification. The thermal characteristics of the different LCZ classes were also analyzed based on the mapping results for China’s 32 major cities, and the experimental results confirmed the close relationship between the LCZ classes and the magnitude of the Land Surface Temperature (LST).

The mapping of urban areas at regional to global scales is a crucial task due to its value for environmental monitoring, habitat and biodiversity conservation, and decision-making. In most current applications, two techniques (i.e., supervised classification and data fusion) are widely applied in large-scale urban mapping. However, the costly training sample collection, inadequate data-source descriptions, and diverse urban characteristics (e.g., shape, size, socioeconomic status, and physical environment) are challenging problems for the urban mapping approaches. In this context, aiming at effectively deriving accurate urban areas at a large scale, we propose a novel ensemble support vector machine (SVM) method which consists of three steps: 1) the automatic generation of training data to reduce labor costs; 2) the construction of an ensemble SVM model to effectively combine the multisource data (including remote sensing and socioeconomic data); and 3) an adaptive patch-based thresholding technique to tackle the diverse urban characteristics. The proposed method is employed to map urban areas of China in 2005 and 2010, and the resulting maps are compared with the existing urban maps for 287 prefecture-level cities. It is found that our results present a satisfactory superiority, especially in challenging small cities, with a significant improvement in median Kappa (0.174 for 2005 and 0.203 for 2010). When incorporating moderate-resolution imaging spectroradiometer multispectral data as an additional source, the Kappa coefficient can be further raised by 0.028 for 2010. In general, the proposed method shows great potential for accurately mapping urban areas at regional, continental, or even global scales in a cost-effective manner.

High Spatial and Spectral Resolution (HSSR) remote-sensing images can provide rich spectral bands and detailed ground information, but there is a relative lack of research on this new type of remote-sensing data. Although there are already some HSSR datasets for deep learning model training and testing, the data volume of these datasets is small, resulting in low classification accuracy and weak generalization ability of the trained models. In this paper, an HSSR dataset Luojia-HSSR is constructed based on aerial hyperspectral imagery of southern Shenyang City of Liaoning Province in China. To our knowledge, it is the largest HSSR dataset to date, with 6438 pairs of 256 × 256 sized samples (including 3480 pairs in the training set, 2209 pairs in the test set, and 749 pairs in the validation set), covering area of 161 km² with spatial resolution 0.75 m, 249 Visible and Near-Infrared (VNIR) spectral bands, and corresponding to 23 classes of field-validated ground coverage. It is an ideal experimental data for spatial-spectral feature extraction. Furthermore, a new deep learning model 3D-HRNet for interpreting HSSR images is proposed. The conv-neck in HRNet is modified to better mine the spatial information of the images. Then, a 3D convolution module with attention mechanism is designed to capture the global-local fine spectral information simultaneously. Subsequently, the 3D convolution is inserted into the HRNet to optimize the performance. The experiments show that the 3D-HRNet model has good interpreting ability for the Luojia-HSSR dataset with the Frequency Weighted Intersection over Union（FWIoU） reaching 80.54%, indicating that the Luojia-HSSR dataset constructed in this paper and the proposed 3D-HRnet model have good applicable prospects for processing HSSR remote sensing images.