Search

Nighttime light (NTL) is a reliable indicator for measuring urban development, population density, and economic activities. This study sets out to explain the spatio-temporal pattern of urbanization and population distribution using pixel-based NTL data. Changes in NTL intensity in Puglia (Italy) were evaluated from 2014 to 2023, and how NTL is affected by various spatial metrics was investigated. Summer–winter differences at the pixel level were monitored utilizing the Normalized Nighttime Light Index (NNLI), and the relationship between this indicator and tourism was assessed. A consistent increase in NTL across the region in both summer and winter was measured, with variations among provinces. Urban areas with high population density showed greater NTL values, while inland areas showed low NTL intensity. The study also revealed that proximity to urban areas was the most influential factor in predicting NTL intensity. The analysis revealed that differences between summer and winter correlated with tourist activity, especially in coastal municipalities. Leveraging NTL images can support decision-making across various sectors and aid in developing evidence-based urban strategies tailored to local needs and challenges.

The Tibetan Plateau, known for its high elevation and sparse population distribution, heavily depends on gridded population data to enhance disaster prevention and management strategies. This study utilizes multi-source physical geographic and socio-economic factors to delineate the population distribution across the plateau. Using data from the seventh National Census in 2020, we apply three individual machine learning methods (Random Forest, GBDT, and XGBoost) and two multi-model ensemble methods (weighted average ensemble and stacking ensemble) to spatialize the population data into a 100-meter grid. The results reveal that the spatialization accuracy of all models exceeds that of the WorldPop dataset. Specifically, the Random Forest model (RMSE = 4061.09, nRMSE = 44.71 %) and the stacking ensemble model (RMSE = 4094.47, nRMSE = 44.26 %) demonstrate the highest accuracy among the individual and ensemble models, respectively. Emphasizing the importance of integrating multi-source big data, Tencent location-based services data emerges as a crucial variable across all models. This study highlights the effectiveness of ensemble models and multi-source big data in improving population mapping accuracy, especially in regions with complex terrains.

Rapid urban expansion reflects the increasing economic growth in China. As an original conceive for urban sustainability, economic growth contributes to vegetation changes. However, the spatiotemporal heterogeneity of the relationships between these two factors remains unclear in China. After analyzing the spatiotemporal characteristics of the urban expansion in different cities in China from 2002 to 2020, we found a rapid and continuous urban expansion in China, with the mean annual expansion of 1710.96 km2 year−1 (p 60% of cities showed incoordination status and 38.08% of cities showed strongly incoordination status. At pixels levels, these strong incoordination status in urban areas can be fully explained by long-term average NTL digital Number (DN) gradient.Based on a piecewise linear regression model, the proportion of areas with the strong incoordination status was significantly increased by the 1.07% DN−1 (p

• Urbanization dynamics in Puglia, Italy were analyzed using pixel-based NTL data. • Annual changes were examined in winter and summer at regional and provincial scales. • Spatial metrics were used to explain the intensity of NTL. • NNLI was created to monitor seasonal differences and to predict tourist traffic. Nighttime light (NTL) is a reliable indicator for measuring urban development, population density, and economic activities. This study sets out to explain the spatio-temporal pattern of urbanization and population distribution using pixel-based NTL data. Changes in NTL intensity in Puglia (Italy) were evaluated from 2014 to 2023, and how NTL is affected by various spatial metrics was investigated. Summer–winter differences at the pixel level were monitored utilizing the Normalized Nighttime Light Index (NNLI), and the relationship between this indicator and tourism was assessed. A consistent increase in NTL across the region in both summer and winter was measured, with variations among provinces. Urban areas with high population density showed greater NTL values, while inland areas showed low NTL intensity. The study also revealed that proximity to urban areas was the most influential factor in predicting NTL intensity. The analysis revealed that differences between summer and winter correlated with tourist activity, especially in coastal municipalities. Leveraging NTL images can support decision-making across various sectors and aid in developing evidence-based urban strategies tailored to local needs and challenges. [ABSTRACT FROM AUTHOR]

• Population of the Tibetan Plateau in 2020 was mapped at a fine grid resolution of 100 m. • Multi-source data and ensemble model enhanced population mapping in complex terrain. • Ensemble models blend strengths and weaknesses of each machine learning method. • Tencent LBS data directly boosted accuracy and simplifies models. • Ensemble models had higher accuracy than other models and data. The Tibetan Plateau, known for its high elevation and sparse population distribution, heavily depends on gridded population data to enhance disaster prevention and management strategies. This study utilizes multi-source physical geographic and socio-economic factors to delineate the population distribution across the plateau. Using data from the seventh National Census in 2020, we apply three individual machine learning methods (Random Forest, GBDT, and XGBoost) and two multi-model ensemble methods (weighted average ensemble and stacking ensemble) to spatialize the population data into a 100-meter grid. The results reveal that the spatialization accuracy of all models exceeds that of the WorldPop dataset. Specifically, the Random Forest model (RMSE = 4061.09, nRMSE = 44.71 %) and the stacking ensemble model (RMSE = 4094.47, nRMSE = 44.26 %) demonstrate the highest accuracy among the individual and ensemble models, respectively. Emphasizing the importance of integrating multi-source big data, Tencent location-based services data emerges as a crucial variable across all models. This study highlights the effectiveness of ensemble models and multi-source big data in improving population mapping accuracy, especially in regions with complex terrains. [ABSTRACT FROM AUTHOR]

Nighttime light data and land use data derived from remote sensing satellites are commonly used to monitor human activities on the Earth. A thorough understanding of the spatiotemporal interactions between these data enables better characterization and reconstruction of urbanization processes. In this work, a pixelwise fusion analysis using nighttime light data and land use data was performed to explore the spatiotemporal correlation between urban nighttime light luminosity transitions and land use changes. We found that the nighttime luminosity transitions are tightly related to the corresponding land use change types at the pixel level. A significant spatial correlation was identified between the nighttime light luminosity transitions and urban lands with intensive human activity. Relevant policies enacted in Beijing have greatly influenced the local light luminosity and land use type transitions. Pixels with change values in light luminosity in the same range have the same types of land use changes, the reasons for the changes in light luminosity are similar, and the spatial distribution characteristics are the same. This study provides a theoretical basis for quickly assessing changes in urban land use types through transitions in nighttime light luminosity. Simultaneously, the data in our study after integrating nighttime light luminosity and land use information perform well in urban development research and can provide valuable datasets and decision-making references for adjusting and optimizing urban sustainable development policies.

• The incoordination status between economic growth and urban greenness was found in >60% of cities in China. • At pixels levels, only 11.61% of urban areas showed a strong coordination status in China. • Identify the thresholds of long-term average NTL that explain the spatial heterogeneities of incoordination status between economic growth and urban greenness. Rapid urban expansion reflects the increasing economic growth in China. As an original conceive for urban sustainability, economic growth contributes to vegetation changes. However, the spatiotemporal heterogeneity of the relationships between these two factors remains unclear in China. After analyzing the spatiotemporal characteristics of the urban expansion in different cities in China from 2002 to 2020, we found a rapid and continuous urban expansion in China, with the mean annual expansion of 1710.96 km2 year−1 (p < 0.01). However, based on the trends of annual maximum enhanced vegetation index (EVI max) and the nighttime light (NTL), a strong incoordination status (areas with significant increasing NTL (p < 0.05) but significantly decreasing EVI max) between urban greenness and economic growthwas found during this rapid urban expansion period. Spatially, >60% of cities showed incoordination status and 38.08% of cities showed strongly incoordination status. At pixels levels, these strong incoordination status in urban areas can be fully explained by long-term average NTL digital Number (DN) gradient.Based on a piecewise linear regression model, the proportion of areas with the strong incoordination status was significantly increased by the 1.07% DN−1 (p < 0.01) when multi-year average NTL value is <50, but decreased strongly afterward, with the slope of −2.96% DN−1 (p < 0.01). The strong incoordination between urban greenness and economy can be rapidly mitigated with the rapid economic growth. This threshold can be a critical indicator for decision-makers in identifying and analyzing the status of ecological civilization in various regions of China. [ABSTRACT FROM AUTHOR]

[Display omitted] • Pixel-wise fine-grained analysis between NTL data and land use data. • Light brightness will synchronously respond to the changes of land use types. • There is a good spatial correspondence between NTL and urban land use types. • Beijing's development was influenced obviously by policies from 2013 to 2020. Nighttime light data and land use data derived from remote sensing satellites are commonly used to monitor human activities on the Earth. A thorough understanding of the spatiotemporal interactions between these data enables better characterization and reconstruction of urbanization processes. In this work, a pixelwise fusion analysis using nighttime light data and land use data was performed to explore the spatiotemporal correlation between urban nighttime light luminosity transitions and land use changes. We found that the nighttime luminosity transitions are tightly related to the corresponding land use change types at the pixel level. A significant spatial correlation was identified between the nighttime light luminosity transitions and urban lands with intensive human activity. Relevant policies enacted in Beijing have greatly influenced the local light luminosity and land use type transitions. Pixels with change values in light luminosity in the same range have the same types of land use changes, the reasons for the changes in light luminosity are similar, and the spatial distribution characteristics are the same. This study provides a theoretical basis for quickly assessing changes in urban land use types through transitions in nighttime light luminosity. Simultaneously, the data in our study after integrating nighttime light luminosity and land use information perform well in urban development research and can provide valuable datasets and decision-making references for adjusting and optimizing urban sustainable development policies. [ABSTRACT FROM AUTHOR]

• VIIRS DNB data provide more details compared with DMSP/OLS data. • Most protected areas were detected with significant human light disturbance. • Lighted pixels changed differently in different types of protected areas. • Lighted pixels increased significantly from 2012 to 2017. Protected areas, a globally accepted conservation strategy, play a fundamental role in biodiversity and species conservation. There are increasing concerns about the ecological influence of nighttime light within protected areas due to the emergence of more light-related ecological issues. Previous approaches for detecting nighttime light mainly used the traditional data source released by the Defense Meteorological Satellite Program Operational Linescan System (DMSP/OLS), but its coarse spatial resolution and limited radiometric resolution dramatically hamper prompt detection. In this study, we used data from a new source, the Visible Infrared Imaging Radiometer Suite Day-Night Band (VIIRS DNB) to detect nighttime light disturbance within protected areas of mainland China. Protected areas extracted from Landsat 8 Operational Land Imager and the Thermal Infrared Sensor (OLI-TIRS) images served as ground truths to assess detection accuracy. We found that the VIIRS DNB data provided more and better details compared with the traditional DMSP/OLS data. Pixel-based trend analysis clearly indicated that within the protected areas lighted pixels existed extensively and increased significantly from 2012 to 2017. This study provides a new solution to better understand human activities within protected areas. [ABSTRACT FROM AUTHOR]