Comparative Analysis of the Seasonal Driving Factors of the Urban Heat Environment Using Machine Learning: Evidence from the Wuhan Urban Agglomeration, China, 2020

Atmosphere, Journal Year: 2024, Volume and Issue: 15(6), P. 671 - 671

Published: May 31, 2024

With the ongoing advancement of globalization significantly impacting ecological environment, continuous rise in Land Surface Temperature (LST) is increasingly jeopardizing human production and living conditions. This study aims to investigate seasonal variations LST its driving factors using mathematical models. Taking Wuhan Urban Agglomeration (WHUA) as a case study, it explores characteristics employs Principal Component Analysis (PCA) categorize factors. Additionally, compares traditional models with machine-learning select optimal model for this investigation. The main conclusions are follows. (1) WHUA’s exhibits significant differences among seasons demonstrates distinct spatial-clustering different seasons. (2) Compared geographic spatial models, Extreme Gradient Boosting (XGBoost) shows better explanatory power investigating effects LST. (3) Human Activity (HA) dominates influence throughout year positive correlation LST; Physical Geography (PG) negative Climate Weather (CW) show similar variation PG, peaking transition; Landscape Pattern (LP) weak LST, winter while being relatively inconspicuous summer transition. Finally, through comparative analysis multiple constructs framework exploring features aiming provide references guidance development WHUA regions.

Language: Английский

---

Spatiotemporal changes in urban forest carbon sequestration capacity and its potential drivers in an urban agglomeration: Implications for urban CO2 emission mitigation under China’s rapid urbanization

Ecological Indicators, Journal Year: 2024, Volume and Issue: 159, P. 111601 - 111601

Published: Jan. 26, 2024

Urban forests can absorb carbon dioxide for urban CO2 emission mitigation. However, the potential capacity of forest sequestration (CS) and its drivers remain unclear in agglomerations under rapid urbanization. In our study, net primary productivity (NPP) built-up areas was reconstructed Harbin-Changchun agglomeration (HCUA) from 2000 to 2020 reflect CS, spatial CS patterns were further explored using Geodetector model. Our results showed that HCUA has experienced urbanization over past 20 years. Across gradient, higher new developing than old developed all The increased gradually 2020, especially large areas. skewed toward low (<100 g·m−2) medium value (100–300 class distributions years; however, proportion high (>300 show an overall increasing trend small, low-altitude total 0.35 Mt·C·yr−1 2.06 could offset approximately 2.23 % emissions 2000, 5.08 2020. Natural factors, such as temperature, mainly determined changes distribution. addition, we found morphology build-up area, construction height, population density, gross national product, significantly influence CS. We there may exist threshold area product affecting variation. interaction between natural anthropogenic factors had stronger explanatory power variation study help city managers formulate low-carbon development strategies address negative impacts climate change realize cities.

Language: Английский

---

Seasonal surface urban heat island analysis based on local climate zones

Ecological Indicators, Journal Year: 2024, Volume and Issue: 159, P. 111669 - 111669

Published: Feb. 1, 2024

A fundamental aspect of ensuring urban sustainability is a comprehensive understanding the driving mechanisms behind heat island (UHI) phenomenon. The primary objective this study to investigate spatiotemporal variations and underlying surface (SUHI) in Hefei. employed local climate zone (LCZ) method analyze land morphology spatial structure for 2014 2021. Subsequently, calculations were conducted derive intensity (SUHII), normalized difference built-up index (NDBI), vegetation (NDVI), gravity water (GWI), building fraction (BSF), road density (RD), poi (PD), population (PPD). exploration by which factors influence SUHI was utilizing both Pearson correlation analysis geographic detector models. results revealed that sparsely built (LCZ 9) low plants D) predominantly characterized natural coverage areas, respectively. summer season distinguished most extensive distribution highest SUHII levels. Significantly, consistently exceeded those LCZs when contrasted LCZs. Large lowrise 8) displayed levels, whereas G) exhibited lowest values. NDBI took precedence showed positive with SUHI. Among socio-economic factors, height (BH) demonstrated superior explanatory capability compared other variables. interaction between NDVI maximized explanation under different seasons. findings will serve as critical insights planners policymakers, enabling development scientifically-based efficacious strategies mitigate

Language: Английский

---

Global Mapping of Three-Dimensional (3D) Urban Structures Reveals Escalating Utilization in the Vertical Dimension and Pronounced Building Space Inequality

Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: March 1, 2024

Three-dimensional (3D) urban structures play a critical role in informing climate mitigation strategies aimed at the built environment and facilitating sustainable development. Regrettably, there exists significant gap detailed consistent data on 3D building space with global coverage due to challenges inherent collection model calibration processes. In this study, we constructed structure dataset (GUS-3D), including volume, height, footprint information, 500 m spatial resolution using extensive satellite observation products numerous reference samples. Our analysis indicated that total volume of buildings worldwide 2015 exceeded 1 × 1012 m3. Over 1985 period, observed slight increase magnitude growth (i.e., it increased from 166.02 km3 during 1985–2000 period 175.08 2000–2015 period), while expansion magnitudes two-dimensional (2D) (22.51 103 km2 vs. 13.29 km2) extent (157 133.8 notably decreased. This trend highlights intensive vertical utilization land. Furthermore, identified heterogeneity provision inequality across cities worldwide. is particularly pronounced many populous Asian cities, which has been overlooked previous studies economic inequality. The GUS-3D shows great potential deepen our understanding creates new horizons for studies.

Language: Английский

---

A 100 m gridded population dataset of China's seventh census using ensemble learning and big geospatial data

Earth system science data, Journal Year: 2024, Volume and Issue: 16(8), P. 3705 - 3718

Published: Aug. 16, 2024

Abstract. China has undergone rapid urbanization and internal migration in the past few years, its up-to-date gridded population datasets are essential for various applications. Existing China, however, suffer from either outdatedness or failure to incorporate data latest Seventh National Population Census of conducted 2020. In this study, we develop a novel downscaling approach that leverages stacking ensemble learning big geospatial produce grids at 100 m resolution using seventh census both county town levels. The proposed employs integrate strengths random forest, XGBoost, LightGBM through fusing their predictions training mechanism, it delineates inhabited areas enhance estimation. Experimental results demonstrate exhibits best-fit performance compared individual base models. Meanwhile, out-of-sample town-level test set indicates estimated dataset (R2=0.8936) is more accurate than existing WorldPop (R2=0.7427) LandScan (R2=0.7165) products Furthermore, with area enhancement, spatial distribution intuitively reasonable two products. Hence, provides valuable option producing datasets. holds great significance future applications, publicly available https://doi.org/10.6084/m9.figshare.24916140.v1 (Chen et al., 2024b).

Language: Английский

---

Comprehensive risk assessment of typhoon disasters in China's coastal areas based on multi-source geographic big data

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 926, P. 171815 - 171815

Published: March 19, 2024

Language: Английский

---

Scale-dependent and season-dependent impacts of 2D/3D building morphology on land surface temperature

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 97, P. 104788 - 104788

Published: July 10, 2023

Language: Английский

---

HiMIC-Monthly: A 1 km high-resolution atmospheric moisture index collection over China, 2003–2020

Scientific Data, Journal Year: 2024, Volume and Issue: 11(1)

Published: April 24, 2024

Abstract Near-surface atmospheric moisture is a key environmental and hydro-climatic variable that has significant implications for the natural human systems. However, high-resolution data are severely lacking fine-scale studies. Here, we develop first 1 km high spatial resolution dataset of monthly index collection in China (HiMIC-Monthly) over long period 2003~2020. HiMIC-Monthly generated by light gradient boosting machine algorithm (LightGBM) based on observations at 2,419 weather stations multiple covariates, including land surface temperature, vapor pressure, cover, impervious proportion, population density, topography. This includes six commonly used indices, enabling assessment conditions from different perspectives. Results show good performance, with R 2 values all indices exceeding 0.96 root mean square error absolute within reasonable range. The exhibits consistency situ various temporal regimes, demonstrating broad applicability strong reliability.

Language: Английский

---

Building Height Extraction From High-Resolution Single-View Remote Sensing Images Using Shadow and Side Information

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2024, Volume and Issue: 17, P. 6514 - 6528

Published: Jan. 1, 2024

Extracting building heights from single-view remote sensing images greatly enhances the application of data. While methods for extracting height shadow have been widely studied, it remains a challenging task. The main reasons are as follows: (1) traditional method information exhibits low accuracy. (2) use only to extract results in limited scenarios. To solve above problems, this paper introduces side and complement each other, proposes extraction high-resolution using information. Firstly, we propose RMU-Net method, which utilizes multi-scale features This aims address issues related pixel detail loss imprecise edge segmentation, result significant scale differences within segmentation targets. Additionally, employ area threshold optimize results, specifically tackle small stray patches holes, enhancing overall integrity accuracy extraction. Secondly, that integrates based on an enhanced proportional coefficient model. measuring lengths is improved by incorporating fishing net informed our analysis geometric relationships among buildings. Finally, establish dataset containing images, select multiple areas experimental analysis. demonstrate 91.03% 90.29%. average absolute error (MAE) 1.22, while root mean square (RMSE) 1.21. Furthermore, proposed method's validity scalability affirmed through analyses applicability anti-interference performance extensive areas.

Language: Английский

---

A deep learning-based super-resolution method for building height estimation at 2.5 m spatial resolution in the Northern Hemisphere

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: 310, P. 114241 - 114241

Published: June 4, 2024

Language: Английский

---