TL;DR
This study analyzes the spatial and temporal evolution of color steel buildings in Lanzhou City from 2013 to 2021 using remote sensing data and geospatial analysis. The findings reveal a dynamic pattern of increase and decrease in these structures, highlighting their role as indicators of urban development stages and providing insights for sustainable urban planning.
What Problem Does This Paper Address?
Rapid urbanization in China has led to significant changes in land use and urban expansion. Color steel plate buildings, a common and cost-effective building material, have become prevalent in temporary structures, industrial parks, and urban villages. While their presence is widespread, understanding their spatial distribution, temporal evolution, and influencing factors is crucial for urban planning, land management, and addressing potential safety and environmental concerns. Existing research often focuses on extraction techniques or static spatial analysis, lacking a comprehensive understanding of the dynamic, spatiotemporal evolution of these buildings, especially in rapidly developing cities like those in Northwest China. This paper aims to fill this gap by analyzing the spatiotemporal characteristics of color steel buildings in Lanzhou City.
Methods
The study employs a multi-faceted approach combining remote sensing data with various geospatial analysis techniques:
- Data Extraction: Color steel building data was extracted from Google Imagery (2013) and GF-2 satellite imagery (2014-2021). A self-attention mechanism U-Net neural network model was utilized for automated extraction from GF-2 imagery, demonstrating high accuracy.
- Spatial Analysis:
- Center of Gravity Analysis: This method tracks the movement of the spatial centroid of color steel buildings over time to understand expansion trends and directions.
- Standard Deviation Ellipse: This technique visualizes the spatial distribution orientation, dispersion, and diffusion direction of the buildings.
- Compactness Index: Measures the shape characteristics of the building distribution, indicating morphological compactness.
- Patch Density (PD) and Landscape Percentage: These metrics assess landscape composition and the clustering degree of color steel buildings.
- Weighted Kernel Density Analysis: Depicts the clustering intensity of features, accounting for the varying areas of buildings.
- Extreme Point Detection Model: Identifies hotspots of high-density color steel buildings.
- Predictive Analysis: The FLUS (Future Land Use Simulation Model) was used to predict the suitability and future distribution of color steel plate buildings based on influencing factors like vegetation coverage, population density, and road distance. The Kappa coefficient was used to evaluate the accuracy of these predictions.
Datasets
- Google Imagery: Used for manual vectorization of color steel buildings in 2013.
- GF-2 Satellite Imagery: Used for automated extraction of color steel buildings from 2014 to 2021.
Key Insights and Ideas
- Dynamic Evolution: Color steel buildings are not static; their numbers and distribution change significantly over time, reflecting urban development cycles and policy interventions.
- Urban Development Indicator: The presence and evolution of color steel buildings can serve as an indicator of urban development stages, particularly in transitional periods or areas undergoing rapid change.
- Policy Impact: The study suggests that government policies, such as shantytown renovation and urban management improvements, have influenced the decrease in color steel buildings in certain areas.
- Spatial Heterogeneity: The distribution and evolution patterns of color steel buildings vary significantly across different districts within Lanzhou, reflecting localized development dynamics.
- Predictive Potential: Geospatial models can effectively predict the future distribution of these structures, aiding in urban planning and resource allocation.
Key Results
- Overall Trend: The number of color steel buildings in Lanzhou City increased from 20,497 in 2013 to a peak of 27,318 in 2017, then decreased to 26,425 in 2021.
- Area: The total area of color steel buildings also followed a similar trend, peaking in 2017.
- District-Level Variations: Chengguan and Qilihe districts had the highest distribution of color steel buildings. Anning District showed a decreasing trend, while Xigu and Honggu districts exhibited increasing trends in both number and area.
- Directionality: The overall spatial distribution and expansion of color steel buildings in Lanzhou showed a tendency towards the northwest and southeast directions.
- Aggregation: While some districts showed higher aggregation, the overall aggregation of color steel buildings in Lanzhou was relatively low, indicating a degree of scatteredness and increasing expansibility.
- Extraction Accuracy: The U-Net model achieved high accuracy (mean Kappa of 82.00%) in extracting color steel buildings from satellite imagery.
- Prediction Accuracy: The FLUS model demonstrated high simulation accuracy (Kappa coefficient of 0.797189) for predicting color steel building distribution.
Limitations and Future Work
- Data Granularity: While GF-2 imagery provides high resolution, the study acknowledges that finer-grained data might reveal more nuanced patterns.
- Influencing Factors: While several factors were considered for prediction, future work could explore a broader range of socioeconomic and environmental variables.
- Temporal Resolution: The study covers specific years; a more continuous temporal analysis could provide a smoother understanding of the evolution.
- Future Research: The authors suggest further research into the impact of urban villages and shantytowns with color steel plate buildings on residents' lives and safety, and the exploration of more detailed remote sensing data systems. They also highlight the need for developing new methods and integrating various data sources to characterize urban spatial morphology and development.
Summary updated