Search

ABSTRACTMethods for fine-grained sample collection are essential for detecting land cover changes at large scales. The complexity of wetland types increases the difficulty of obtaining training samples for high-precision wetland changes, while existing methods mainly focus on coarse-grained classification of urban areas, ignoring the physical growth cycle of vegetation. To solve the above problems, we propose a method for phenological knowledge transfer-based fine grained land cover change sample collection (PKT). Taking the Yellow River Delta as an example, the experimental results are shown as follows. (1) The overall accuracy of the results of the PKT method is 77.03%, and k is 0.42, which is better than the results of the other methods. (2) The PKT method is able to obtain the area of wetland change more accurately and can identify the wetland type changes in the area of change. (3) Making full use of multisource data and fine-grained category information can effectively improve the accuracy of change training samples. (4) Changes in coastal wetlands are the result of the interaction between natural factors and human activities. (5) Further restoration and management of wetlands can be carried out in terms of appropriate protective measures and restrictions on construction behavior.

The United Nations adopted 17 Sustainable Development Goals (SDGs) to address societal, economic and environmental sustainability issues. The efficiency of SDGs monitoring could be improved by essential variables (EVs), which can help to better deal with massive data, interdisciplinary knowledge and workloads. However, in practice, effectively combining EVs with SDGs monitoring remains challenging. In this paper, we proposed a refining method of essential SDGs variables (ESDGVs) to land degradation. Firstly, we selected northwest China as our experimental region and extracted a group of variables related to land degradation from SDG indicators based on the DPSIR framework. Next, we identify the essential ones using a combined qualitative and quantitative methods with the criteria of feasibility, spatialization, and relevance which considered the issues of data acquisition, monitoring scale, and closeness to the land degradation. Finally, we analysed the monitoring role of ESDGVs. Results show that, compared to conventional observations, ESDGVs facilitate the monitoring and evaluation of regional SDGs with reduced efforts. And both climate and human activities have a facilitating or inhibiting effect on land degradation processes. In the future, we hope to have more mature data sets and consider adding more SDG indicators for ESDGVs’ refinement.

The distribution and dynamic changes of regional or national population data with long time series are very important for regional planning, resource allocation, government decision-making, disaster assessment, ecological protection, and other sustainability research. However, the existing population datasets such as LandScan and WorldPop all provide data from 2000 with limited time series, while GHS-POP only utilizes land use data with limited accuracy. In view of the limited remote sensing images of long time series, it is necessary to combine existing multi-source remote sensing data for population spatialization research. In this research, we developed a nighttime light desaturation index (NTLDI). Through the cross-sensor calibration model based on an autoencoder convolutional neural network, the NTLDI was calibrated with the same period Visible Infrared Imaging Radiometer Suite Day/Night Band (VIIRS-DNB) data. Then, the geographically weighted regression method is used to determine the population density of China from 1990 to 2020 based on the long time series NTL. Furthermore, the change characteristics and the driving factors of China’s population spatial distribution are analyzed. The large-scale, long-term population spatialization results obtained in this study are of great significance in government planning and decision-making, disaster assessment, resource allocation, and other aspects.

Assessing Sustainable Development Goal (SDG) indicator 15.3.1, which refers to the proportion of degraded land to total land area, and analysing its status and drivers is essential for the development of policies to promote the early achievement of SDG target 15.3 of Land Degradation Neutrality (LDN). In this study, Northeast China was selected as the study area, and the progress of indicator 15.3.1 was assessed based on the perspective of Net Primary Productivity (NPP) calculated by the CASA model. WorldPop population spatial distribution data were used as a proxy for human activities, combined with climate data to calculate the effects of changes in temperature, precipitation and population spatial distribution on vegetation NPP based on the partial correlation coefficient method and the Geodetector method. The results showed that 92.81% of the areas that passed the test of significance showed an increasing trend in vegetation NPP from 2000 to 2020. The vegetation NPP was affected by a combination of temperature, precipitation and population. The effects of temperature and precipitation on spatial differences in NPP for various vegetation types were significantly greater than those of population, but in high-density population zones, the effects of population on spatial differences in NPP were generally greater than those of temperature and precipitation. Precipitation was the main driver for spatial variation in NPP in deciduous broad-leaved forests, cultivated vegetation and thickets, while temperature was the main driver for spatial variation in NPP in evergreen coniferous forests. Generally, the warming and wetting trend in Northeast China contributed to the accumulation of NPP in cultivated vegetation, thickets, steppes and grasslands. The sensitivity of NPP to temperature and precipitation in deciduous broad-leaved and deciduous coniferous forests varied according to geographical location.

Evaluating the ecological vulnerability of the northwestern region of Liaoning Province, a typical ecologically vulnerable area in China, and exploring its inherent patterns help formulate effective measures for ecological protection and restoration. Existing studies in this region have focused on the county, city, or provincial level, failing to capture the overall ecological vulnerability. Based on the “sensitivity-degradation” model framework, this study employed the Analytic Hierarchy Process, Principal Component Analysis, and geographical detector methods to evaluate the ecological vulnerability in the northwestern region of Liaoning Province from 2000 to 2022 and analyzed the possible driving factors. The results showed that, from 2000 to 2022, areas with moderate or higher vulnerability accounted for an average of 64.7% of the total area, while severely and extremely vulnerable areas accounted for 32.5%. These severely and extremely vulnerable areas are mainly distributed along both banks of the Liuhe River and in the northern part of Liaoning Province, where it adjoins the Horqin Sandy Land. From 2000 to 2022, the overall ecological vulnerability in the northwestern region of Liaoning Province showed a decreasing trend, with the proportion of severe vulnerability significantly decreasing from 35.6% to 8.6%. Vegetation and land use were the primary factors leading to long-term changes in ecological vulnerability. This research can also provide an example for the similar issues in other study areas.

Studying urban heat islands holds significance for the sustainable development of cities. This comprehensive study analyzed the temporal characteristics of a Surface Urban Heat Island and Canopy Layer Heat Island by employing Moderate-Resolution Imaging Spectroradiometer image data spanning from 2003 to 2020 over Beijing, China. Leveraging the Gaussian capacity model, the geometrical characteristics of the Surface Urban Heat Island and Canopy Layer Heat Island, such as intensity, center, direction, and range, were examined among three different timescales of day, month, and year. Results indicate that the intensities of the Surface Urban Heat Island and Canopy Layer Heat Island tend to have bigger seasonal variations during winter nights and summer daytime. In addition, at night the centers of Surface Urban Heat Island and Canopy Layer Heat Island are mainly concentrated in the range of 116.3°~116.4° E in longitude and 39.90°~39.95° N in latitude, while during the daytime they are more scattered, mainly in the range of 116.2°~116.5° E in longitude and 39.7°~40.0° N in latitude. In the hot season, the center of the heat island moves east to north, while in the cold season it moves west to south. Monthly average ellipse areas of Surface Urban Heat Island and Canopy Layer Heat Island vary more during the day than that at night, the maximum daytime differences were 2662 km2 and 2293 km2, while the maximum nighttime differences were 484 km2 and 265 km2. Overall, the average area is increasing, with the heat island center moving eastward and deflecting towards the northeast-southwest direction. The expansion of urban areas will continue to influence the movement and extent of heat islands. The study offers insights to inform strategies for mitigating urban heat islands.

Studying driving factors of the urban heat island phenomenon is vital for enhancing urban ecological environments. Urban functional zones (UFZs), key for planning and management, have a substantial impact on the urban thermal environment through their two-dimensional (2D)/three-dimensional (3D) morphology. Despite prior research on land use and landscape patterns, understanding the effects of 2D/3D urban morphology in different UFZs is lacking. This study employs Landsat-8 remote sensing data to retrieve the land surface temperature (LST). A method combining supervised and unsupervised classification is proposed for UFZ mapping, utilizing multi-source geospatial data. Subsequently, parameters defining the 2D/3D urban morphology of UFZs are established. Finally, the Pearson correlation analysis and GeoDetector are used to analyze the driving factors. The results indicate the following: (1) In the Fifth Ring Road area of Beijing, the residential zones exhibit the highest LST, followed by the industrial zones. (2) In 2D urban morphology, the percentage of built-up landscape (built-PLAND) and Shannon’s diversity index (SHDI) are the main factors influencing LST. In 3D urban morphology, building density, the sky view factor (SVF), and the area-weighted mean shape index (shape index) are the main factors influencing LST. Therefore, low-density buildings with simple and dispersed shapes contribute to mitigating LST, while fragmented distributions of trees, grasslands, and water bodies also play important roles in alleviating LST. (3) In the interactive detection results, all UFZs show the highest interaction detection results with the built-PLAND. (4) Spatial variations are observed in the impact of different UFZs on LST. For instance, in the residential zones, industrial zones, green space zones, and public service zones, the SVF is negatively correlated with LST, while in the commercial zones, the SVF exhibits a positive correlation with LST.

Notable achievements have been made in remote sensing images change detection with sample-driven supervised deep learning methods. However, the requirement of the number of labeled samples is impractical for many practical applications, which is a major constraint to the development of supervised deep learning methods. Self-supervised learning using unlabeled data to construct pretext tasks for model pretraining can largely alleviate the sample dilemma faced by deep learning. And the construction of pretext task is the key to the performance of downstream task. In this work, an improved contrastive self-supervised pretext task that is more suitable for the downstream change detection is proposed. Specifically, an improved Siamese network, which is a change detection-like architecture, is trained to extract multilevel fusion features from different image pairs, both globally and locally. And on this basis, the contrastive loss between feature pairs is minimized to extract more valuable feature representation for downstream change detection. In addition, to further alleviate the problem of little priori information and much image noise in the downstream few-sample change detection, we propose to use variational information bottleneck theory to provide explicit regularization constraint for the model. Compared with other methods, our method shows better performance with stronger robustness and finer detection results in both quantitative and qualitative results of two publicly available datasets.

Fine-scale population spatial distribution plays an important role in urban microcosmic research, influencing infrastructure placement, emergency evacuation management, business decisions, and urban planning. In the past, nighttime light (NTL) data were generally used to map the spatial distribution of the population at a large scale because of their low spatial resolution. The new generation of Luojia1-01 NTL data can be used for fine-scale social and economic analysis with its high spatial resolution and quantitative range. However, due to the geometry and background noise of the data themselves, the accuracy of the original NTL data is still low. Points-of-interest (POI) also can be used to map the population spatialization, but the indicative relationship between the POI and population is not clear, especially in rural and urban areas with different landscape structures. To solve the above-mentioned problems, this study proposes an improved nighttime light (INTL) index to better use the Luojia1-01 NTL data. Meanwhile, a zonal classification model based on INTL and impervious surface area is proposed to distinguish urban and rural areas. Compared with previous research and existing datasets, our result had the highest accuracy (R² = 0.86). This study explains that the INTL index is applicable to population spatialization research with the emergence of high-resolution and multispectral NTL satellite data. Moreover, the zonal classification model in this research can significantly improve the accuracy of population spatialization in rural areas. This study provides a possible way to use NTL and POI data in other social and economic spatialization research.

Rapid population growth has had a significant impact on society, economy and environment, which will challenge the achievement of the United Nations Sustainable Development Goals (SDGs). Spatially accurate and detailed population distribution data are essential for measuring the impact of population growth and tracking progress on the SDGs. However, most population data are evenly distributed within administrative units, which seriously lacks spatial details. There are scale differences between the population statistical data and geospatial data, which makes data analysis and needed research difficult. The disaggregation method is an effective way to obtain the spatial distribution of population with greater granularity. It can also transform the statistical population data from irregular administrative units into regular grids to characterize the spatial distribution of the population, and the original population count is preserved. This paper summarizes the research advances of population disaggregation in terms of methodology, ancillary data, and products and discusses the role of spatial disaggregation of population statistical data in monitoring and evaluating SDG indicators. Furthermore, future work is proposed from two perspectives: challenges with spatial disaggregation and disaggregated population as an Essential SDG Variable (ESDGV).

Ordos City is an important energy supply city for Chinese provinces and cities, providing a secure energy supply for China while also generating corresponding environmental pollution. Examining the spatiotemporal patterns of net primary productivity (NPP) in Ordos City and its driving factors is relevant to the realization of the carbon emission policy in Inner Mongolia. This study was undertaken to analyze NPP and its driving factors in Ordos City from 2000 to 2019 using NPP data, CO2 spatial grid data, meteorological data and statistical yearbook data accordingly. The NPP in Ordos City increased significantly from 2000 to 2019, mainly showing low values of NPP in the northwest and high values in the southeast. The usable grassland area and annual mean precipitation had a significant positive correlation with NPP, whereas the other factors had a more significant negative correlation. The usable grassland area had the largest influence on NPP, and fixed asset investment had the smallest influence on NPP. The total NPP–anthropogenic factor regression model and the mean NPP–natural factor regression model constructed allow for the prediction of NPP. Anthropogenic carbon emissions, population growth and usable grassland area were the main causes of NPP changes. Planting and protecting green plants and scientific and effective energy extraction plans are measures that enhance the degree of carbon sequestration in Ordos City.

The characteristics of a wide variety of scales about objects and complex texture features of high-resolution remote sensing images make deep learning-based change detection methods the mainstream method. However, existing deep learning methods have problems with spatial information loss and insufficient feature representation, resulting in unsatisfactory effects of small objects detection and boundary positioning in high-resolution remote sensing images change detection. To address the problems, a network architecture based on 2-dimensional discrete wavelet transform and adaptive feature weighted fusion is proposed. The proposed network takes Siamese network and Nested U-Net as the backbone; 2-dimensional discrete wavelet transform is used to replace the pooling layer; and the inverse transform is used to replace the upsampling to realize image reconstruction, reduce the loss of spatial information, and fully retain the original image information. In this way, the proposed network can accurately detect changed objects of different scales and reconstruct change maps with clear boundaries. Furthermore, different feature fusion methods of different stages are proposed to fully integrate multi-scale and multi-level features and improve the comprehensive representation ability of features, so as to achieve a more refined change detection effect while reducing pseudo-changes. To verify the effectiveness and advancement of the proposed method, it is compared with seven state-of-the-art methods on two datasets of Lebedev and SenseTime from the three aspects of quantitative analysis, qualitative analysis, and efficiency analysis, and the effectiveness of proposed modules is validated by an ablation study. The results of quantitative analysis and efficiency analysis show that, under the premise of taking into account the operation efficiency, our method can improve the recall while ensuring the detection precision, and realize the improvement of the overall detection performance. Specifically, it shows an average improvement of 37.9% and 12.35% on recall, and 34.76% and 11.88% on F1 with the Lebedev and SenseTime datasets, respectively, compared to other methods. The qualitative analysis shows that our method has better performance on small objects detection and boundary positioning than other methods, and a more refined change map can be obtained.