Urban Classification Research Articles

Urban Land Use Classification Model Fusing Multimodal Deep Features

Urban land use classification plays a significant role in urban studies and provides key guidance for urban development. However, existing methods predominantly rely on either raster structure deep features through convolutional neural networks (CNNs) or topological structure deep features through graph neural networks (GNNs), making it challenging to comprehensively capture the rich semantic information in remote sensing images. To address this limitation, we propose a novel urban land use classification model by integrating both raster and topological structure deep features to enhance the accuracy and robustness of the classification model. First, we divide the urban area into block units based on road network data and further subdivide these units using the fractal network evolution algorithm (FNEA). Next, the K-nearest neighbors (KNN) graph construction method with adaptive fusion coefficients is employed to generate both global and local graphs of the blocks and sub-units. The spectral features and subgraph features are then constructed, and a graph convolutional network (GCN) is utilized to extract the node relational features from both the global and local graphs, forming the topological structure deep features while aggregating local features into global ones. Subsequently, VGG-16 (Visual Geometry Group 16) is used to extract the image convolutional features of the block units, obtaining the raster structure deep features. Finally, the transformer is used to fuse both topological and raster structure deep features, and land use classification is completed using the softmax function. Experiments were conducted using high-resolution Google images and Open Street Map (OSM) data, with study areas on the third ring road of Shenyang and the fourth ring road of Chengdu. The results demonstrate that the proposed method improves the overall accuracy and Kappa coefficient by 9.32% and 0.17, respectively, compared to single deep learning models. Incorporating subgraph structure features further enhances the overall accuracy and Kappa by 1.13% and 0.1. The adaptive KNN graph construction method achieves accuracy comparable to that of the empirical threshold method. This study enables accurate large-scale urban land use classification with reduced manual intervention, improving urban planning efficiency. The experimental results verify the effectiveness of the proposed method, particularly in terms of classification accuracy and feature representation completeness.

Knowledge Distillation for Urban Tree Image Classification Based on the Diffusion Model

As urban areas continue to emerge as significant contributors to global carbon emissions, the classification of urban street trees has gained increasing importance in carbon sequestration research. By precisely identifying and categorizing different tree species, the carbon absorption capabilities of urban vegetation can be evaluated more accurately. This understanding is essential for addressing the growing environmental challenges posed by carbon emissions, as urban trees play a crucial role in absorbing carbon dioxide and mitigating the effects of climate change. In this study, a Feature Distillation Based on Diffusion Model (FDBD) framework was proposed, utilizing state-of-the-art image classification technology to enhance the accuracy of urban tree species identification. The framework utilized knowledge distillation, a process where a smaller, more efficient “student” model is trained to mimic the performance of a larger, more complex “teacher” model, significantly reducing computational demands while maintaining high accuracy. The model’s effectiveness has been validated through combination experiments with a selected backbone model, achieving promising results. This approach not only enhances the understanding of urban trees’ carbon sequestration potential but also provides crucial insights for policymakers. By facilitating precise tree classification, it empowers urban planners to implement more informed and targeted strategies for carbon reduction, ultimately promoting more sustainable urban environments.

Did the COVID-19 Pandemic Affect Suicide Rates Differently in Urban and Rural South Korea?

Abstract Background The influence of the COVID-19 pandemic on suicide rates remains a subject of debate. Despite increases in suicidal ideation and attempts reported during the pandemic, several recent meta-analyses have not demonstrated corresponding rises in suicide rates. Given the potential for substantial regional and demographic variability in the psychological consequences of COVID-19-related restrictions, this study explores differential suicide rates in urban versus rural contexts within South Korea subsequent to the pandemic outbreak. Methods This investigation employed age-standardized suicide rates from two periods, 2015-2019 and 2020-2022, across 234 administrative divisions (152 urban, 82 rural). A difference-in-differences approach was utilized to evaluate the pandemic’s impact on suicide rates, segmented by urban and rural classifications. A linear regression model was developed to examine the interaction between location (urban/rural) and time period, controlled for annual variations, perceived stress, and health status within each locale. Results Analysis revealed higher suicide rates in rural areas compared to urban counterparts (β = 1.247, p = 0.003), although no significant changes were observed during the pandemic period itself (β=-0.425, p = 0.682). Furthermore, the interaction between urban/rural status and the pandemic period did not reach statistical significance (p = 0.367). Conclusions The findings indicate no significant contribution of the COVID-19 pandemic to increased suicide rates, irrespective of the urban or rural setting. Further research is warranted to dissect the impact of more direct socioeconomic factors on this dynamic. Key messages • There in no evidence of increased suicide rates during the COVID-19 pandemic in South Korea. • The impact of the COVID-19 pandemic on suicide rates did not significantly different between urban and rural areas.

Communities That HEAL Intervention and Mortality Including Polysubstance Overdose Deaths

The HEALing Communities Study (HCS) evaluated the effectiveness of the Communities That HEAL (CTH) intervention in preventing fatal overdoses amidst the US opioid epidemic. To evaluate the impact of the CTH intervention on total drug overdose deaths and overdose deaths involving combinations of opioids with psychostimulants or benzodiazepines. This randomized clinical trial was a parallel-arm, multisite, community-randomized, open, and waitlisted controlled comparison trial of communities in 4 US states between 2020 and 2023. Eligible communities were those reporting high opioid overdose fatality rates in Kentucky, Massachusetts, New York, and Ohio. Covariate constrained randomization stratified by state allocated communities to the intervention or control group. Trial groups were balanced by urban or rural classification, 2016-2017 fatal opioid overdose rate, and community population. Data analysis was completed by December 2023. Increased overdose education and naloxone distribution, treatment with medications for opioid use disorder, safer opioid prescribing practices, and communication campaigns to mitigate stigma and drive demand for evidence-based interventions. The primary outcome was the number of drug overdose deaths among adults (aged 18 years or older), with secondary outcomes of overdose deaths involving specific opioid-involved drug combinations from death certificates. Rates of overdose deaths per 100 000 adult community residents in intervention and control communities from July 2021 to June 2022 were compared with analyses performed in 2023. In 67 participating communities (34 in the intervention group, 33 in the control group) and including 8 211 506 participants (4 251 903 female [51.8%]; 1 273 394 Black [15.5%], 603 983 Hispanic [7.4%], 5 979 602 White [72.8%], 354 527 other [4.3%]), the average rate of overdose deaths involving all substances was 57.6 per 100 000 population in the intervention group and 61.2 per 100 000 population in the control group. This was not a statistically significant difference (adjusted rate ratio [aRR], 0.92; 95% CI, 0.78-1.07; P = .26). There was a statistically significant 37% reduction (aRR, 0.63; 95% CI, 0.44-0.91; P = .02) in death rates involving an opioid and psychostimulants (other than cocaine), and nonsignificant reductions in overdose deaths for an opioid with cocaine (6%) and an opioid with benzodiazepine (1%). In this clinical trial of the CTH intervention, death rates involving an opioid and noncocaine psychostimulant were reduced; total deaths did not differ statistically. Community-focused data-driven interventions that scale up evidence-based practices with communications campaigns may effectively reduce some opioid-involved polysubstance overdose deaths. ClinicalTrials.gov Identifier: NCT04111939.

How educational inequality affects family multichild behavior—evidence from super high schools

In a context characterized by the dual issues of educational inequality and problems with fertility, the internal relationship between these two factors requires further in-depth exploration. On the basis of data drawn from the China Family Panel Studies (CFPS) from 2010 to 2018, this paper investigates whether “super high schools” in cities represent a form of educational inequality, uses a combination of a two-way fixed effect model with the instrumental variable method to discuss the influence of this situation on family fertility and the corresponding transmission mechanism, develops a proportional risk regression model, and expands our understanding of the influence of educational inequality on the second-child birth interval. This study reveals that educational inequality significantly suppresses the number of family births and that the paths of action associated with this effect include both the explicit and implicit costs of education. The inhibitory effect of educational inequality on the number of family births varies according to urban grade, urban or rural classification, and the education levels attained by women. In addition to decreasing the number of family births, educational inequality increases the second-child birth interval. This paper not only enriches related research on family multichild behavior but also provides an educational reference for attempts to address the issue of continuously low fertility rates in China and establish a connection between population and educational economics.

Urban Morphology Classification and Organizational Patterns: A Multidimensional Numerical Analysis of Heping District, Shenyang City

Prior studies have failed to adequately address intangible characteristics and lacked a comprehensive quantification of cultural dimensions. Additionally, such works have not merged supervised and unsupervised classification methodologies. To address these gaps, this study employed multidimensional numerical techniques for precise spatial pattern recognition and urban morphology classification at the block scale. By examining building density, mean floor numbers, functional compositions, and street block mixed-use intensities, alongside historical and contemporary cultural assets within blocks—with assigned weights and entropy calculations from road networks, building vectors, and POI data—a hierarchical categorization of high, medium, and low groups was established. As a consequence, cluster analysis revealed seven distinctive morphology classifications within the studied area, each with unique spatial configurations and evolutionary tendencies. Key findings include the dominance of high-density, mixed-use blocks in the urban core, the persistence of historical morphologies in certain areas, and the emergence of new, high-rise clusters in recently developed zones. The investigation further elucidated the spatial configurations and evolutionary tendencies of each morphology category. These insights lay the groundwork for forthcoming studies to devise morphology-specific management strategies, thereby advancing towards a more scientifically grounded, rational, and precision-focused approach to urban morphology governance.

Multi-attribute diagnosis of urban flood-bearing bodies based on integrated learning with Stacking–GPR–QPSO coupling

Flood-bearing bodies are urban components directly impacted and damaged by disasters. Current methods for attribute identification and diagnosis of flood-bearing bodies, relying on real-time monitoring, are inadequate for pre-disaster forecasting and lack comprehensiveness. To reduce the uncertainty associated with single data sources, a Dual Path Network (DPN) method was employed to extract feature vectors based on multi-source datasets. A meta-classifier was constructed by integrating five base learners using Stacking, optimized by Quantum Particle Swarm Optimization (QPSO)-enhanced Gaussian Process Regression, forming an ensemble learner for predicting urban spatial classification. Utilizing GIS proximity analysis functions, attributes of functional zones, spatial attributes of points of interest (POI), and flood loss were assigned to each flood-bearing body grid. By overlaying urban flood inundation maps, multi-attribute diagnosis of flood-bearing bodies was achieved. The Jinshui District of Zhengzhou, China, is selected as the study area. The results show: (1) Predictions of urban functional zone categories in four other districts of Zhengzhou showed an average accuracy rate of 78.5 % through random sampling point validation. The threshold effect of prediction accuracy at different scales was significant. (2) Simulated flood economic losses for recurrence intervals of 1 year, 5 years, 10 years, 20 years, 50 years, and 100 years exhibited an exponential growth trend. (3) The multiple flood-bearing attributes of each flooded grid can be diagnosed. Finally, the model was effectively verified by simulating and comparing historical data from the “7·20” flood event in Zhengzhou.

Where is rural? An analysis of the agreement between quantitative measures of rurality

Government agencies and researchers often rely on quantitative measures to map where rural places are and who belongs to rural communities. To better understand the consistency, or lack thereof, across different quantitative rural measures, we compare four common measures applied at the county level for the contiguous United States. Our purpose was to (1) quantify and map the degree of rural agreement among these selected rural measures and (2) describe key characteristics of counties classified as rural through urban. We found that different measures yielded little agreement on rural places with consensus of only 114 of 3108 counties. County variables contributing to good separability in rural to urban classifications included remoteness, population density, percent of the population aged 65 and over, employment in the agricultural sector, and both developed and grassland landcover. Closer examination of the underlying characteristics of the 114 consensus rural counties provides an opportunity to reflect on the nature of popular rural measures and the variables they use. If these counties are not representative of how researchers conceptualize “rural” across US regions, it is motivation for working toward new and/or spatially adaptive rural measures that better represent the communities that policy and research intends to help.

The prediction of urban growth boundary based on the ANN-CA model: An application to Guangzhou

Urban growth boundary (UGB) delineation is critical not only for China's urban planning policies, such as the "three control lines" of the Ministry of Natural Resources, but also for addressing global challenges related to sustainable urban development. This study contributes to the international discourse on urban growth management by developing an innovative artificial neural network-cellular automata (ANN-CA) model, tailored for cities experiencing rapid expansion. Using Guangzhou as a case study, we constructed an impact factor model that incorporates a wide range of factors, including urban spatial terrain, natural environment, current urban land classification, and industrial and economic conditions, along with the layout of modern service networks. The ANN-CA model was then employed to simulate urban spatial expansion and UGB delineation for the year 2030 under various constraints, such as strict protection zones and sustainable development scenarios. Our findings indicate that between 2020 and 2030, Nansha, Panyu, and Zengcheng districts will witness the most significant urban expansion, with respective area increases of 13.81 km2, 8.94 km2, and 5.8 km2, marking them as key growth areas. Furthermore, we propose that future urban expansion in Guangzhou should prioritize the southern and eastern regions, aligning with the city's strategic spatial objectives of "moving east, expanding south, connecting west, and optimizing north." By emphasizing ecological protection and intensive land use, this study provides a robust framework for urban planning in Guangzhou and offers insights applicable to rapidly urbanizing regions worldwide.

Multi-scale feature fusion segmentation and lightweight ensemble deep learning approach for urban slum classification using satellite images

ABSTRACT Although there are some drawbacks to satellite imaging, such as the difficulty in identifying slums in highly populated metropolitan areas, it is nonetheless a useful tool for preliminary screening and can support on-the-ground and other data sources. Hence, we proposed the MSFF-Seg (Multi-Scale Feature Fusion-based Segmentation) model for urban slum classification using satellite images. Hence we proposed the MSFF-Seg model for urban slum classification using satellite images. Initially, three types of corrections are performed in preprocessing such as geometric correction which is corrected by Enhanced Scale Invariant Feature Transform (ESIFT), radiometric correction which is corrected by Ground-based Relative Radiometric Calibration (GRRC), and atmospheric correction which is corrected by homomorphic filter. After completion of preprocessing phase, segmentation is initiated by Tweak Mask RCNN which extracts the features in multilevel for segmentation that increases the segmentation accuracy. The segmented images are fed into an ensemble classifier for urban slum identification. Here, three types of deep learning are used for ensemble learning namely Schematic Convolutional Neural Network (SCNN), Shallow MobileNet, and Lite ShuffleNet. They extract multiple features from the segmented region, based on the extracted features. LGBM performed urban slum classification which classifies the images into two classes such as slums and non-slums. Finally, the performance of this research is evaluated based on several performance metrics like accuracy (98%), precision (96%), recall (94%), and F1 score (97%), which proved that the proposed work achieved superior performance in urban slum classification.

Enhancing urban functional regions classification by integrating multi-source data with a heterogeneous graph neural network

Enhancing urban functional regions classification by integrating multi-source data with a heterogeneous graph neural network

THE ROLE OF HUMAN CAPITAL IN MARRIED WOMEN'S LABOR MARKET PARTICIPATION: A MICRODATA ANALYSIS FROM SAKERNAS

This study aims to analyze the effect of human capital on the tendency of married women to enter the labor market. With 149,508 samples from the National Labor Force Survey, this study uses a binary logistic regression analysis technique with several modifications to grouping samples based on rural – urban categories, and types of work. This study found that in general observation, the tendency to work increased in the group of women with higher education (diploma and university) although not significantly. The same thing was also found in both the rural and urban classifications. Furthermore, in the classification based on the type of work, only in the type of work employees/staff/laborers of secondary and higher education both increase the tendency to work. Women with higher education were found to be more likely to work after marriage, indicated by a higher odds ratio than women with secondary education. The same effect was not found in the type of work self-employed, casual workers, or family workers. The findings of this study imply that on average to the secondary level, education does not affect the participation of married women in the labor market in Indonesia, but only in higher education and formal employment.

The way out for urban household solid waste classification in China: From the perspective of multi-agent based behavior simulation

Rapid industrialization and urbanization drive China's massive urban household solid waste (HSW) generation. Classification and resource recycling of HSW is an effective way to address the “garbage-sieged city” dilemma and promote “urban mineral resources” exploitation. Despite efforts since 2000, HSW sorting and recycling remain at 30%. This study develops a multi-agent based behavior simulation model for HSW classification and recycling, incorporating the interactions and decision-making of key agents, including residents, enterprises, and the government. Taking Suzhou, China as a case study, field surveys were conducted, and behavioral decision-making functions were constructed to simulate micro-behavioral responses and macro-system evolution under different policy incentives, and to predict the implementation effect of domestic waste management approaches, provides the basis for decision-making. The results demonstrate that a combination of waste metering charges, recycling subsidies, and enhanced education program can effectively regulate agent behavior and optimize the system. One year after the implementation of the simulation, the number of “unclassification” residents has dropped from 1.85 million to 250,000 decreased by 86%; the number of “classification deposition” residents has declined from 1.28 million to 1.08 million, reduced by 16%; the number of “selling after classification” residents has risen from 1.05 million to 2.81 million, increased by 63%; the daily waste output per capita has decreased from 1.21 kg to 1.04 kg, reduced by 14%; the recovery rate of resources has been raised from 33.4% to 66.5%. The implementation effects of other policy scenarios are ranked as follows: the metering charge > the recycling subsides > the enhanced education program. The findings provide a basis for decision-making in HSW management. Concerning the metering charge and the recycling subsidies, they can benefit positive interactions encouraged by such economic stimulus. At the same time, it is suggested to strengthen classification and recycling infrastructure, and push forward the integration of HSW classification recycling.

A worldwide review of formal national street classification plans enhanced via an analytical hierarchy process: Street classification as a tool for more sustainable cities

For cities to utilise their maximum liveability potential, their transport infrastructure and overall service provision need to function seamlessly. To this end, urban street eco-systems should be characterised, organised and utilised effectively. But is this happening on a mass scale across the globe? Are our urban street classification schemes forward-thinking and ready to respond to the emerging sustainability and resilience challenges cities face nowadays? This paper aims to answer these questions by examining and decoding the prevailing “formal street classification scheme model” through conducting a detailed worldwide review of formal national street classification plans. Out of 196 countries investigated, 128 official street classification plans were identified, analysed and evaluated. We also used an Analytical Hierarchy Process (AHP) with 20 experts coming from different fields (i.e., academics, policymakers, practitioners) to enhance our results and contribute to developing an index evaluating urban street classification under the prism of sustainability. The outcomes of our work signify that conventional pro-automobile approaches still prevail, thus shaping car-centric conditions, which undermine the role of sustainable modes and reduce the ability of cities to innovate and succeed. It is demonstrated that the road to achieve sustainability and completeness in urban transport systems, considering these car-led plans, is still uphill. Based on that, multi-dimensional classification systems prioritising public and active transport, while appreciating street's urban aspect should be promoted in the future.

Use of LCZs with urban canopy modeling to evaluate urban growth effects on meteorological conditions in the Salt Lake Valley

Urban Canopy Models (UCMs) can be used to represent urban characteristics on a mesoscale without accounting for effects of individual buildings. A WRF-based single-layer urban canopy model (SLUCM) was validated for the greater Salt Lake City area (GSLA) using measurements from two weather stations for a one-week period during summer 2019. A baseline model based on current GSLA properties developed using the Noah land surface model, SLUCM, and refined urban classifications based on WUDAPT Local Climate Zones (LCZs) produced acceptable predictions and computational times. The baseline model was validated using measured meteorological data and clearly showed diurnal heating patterns. Three scenarios representing multi-decade growth were defined by modifying LCZs in different parts of the GSLA. Compared to baseline values, results showed increased urban density increased local nighttime temperatures 5–10 °C depending on the extent of growth, but did not affect daytime temperatures. This was due to changes in LCZ anthropogenic heating rates and sky view factors which were most noticeable during early nighttime hours, consistent with atmospheric heat island behavior. Predicted 2-m relative humidity and 10-m wind speeds changed minimally.

Rural-urban disparities in cardiovascular disease mortality vary by poverty level and region.

To examine rural and urban disparities in cardiovascular disease (CVD) death rates by poverty level and region. Using 2021 county-level population and mortality data for CVD deaths listed as the underlying cause among adults aged 35-64 years, we calculated age-standardized CVD death rates and rate ratios (RR) for 4 categories of counties: high-poverty rural, high-poverty urban, low-poverty rural, and low-poverty urban (referent). Results are presented nationally and by US Census region. Rural and urban disparities in CVD mortality varied markedly by poverty and region. Nationally, the CVD death rate was highest among high-poverty rural areas (191 deaths per 100,000, RR: 1.76, CI: 1.73-1.78). By region, Southern high-poverty rural areas had the highest CVD death rate (256 deaths per 100,000) and largest disparity relative to low-poverty urban areas (RR: 2.05; CI: 2.01-2.09). In the Midwest and West, CVD death rates among high-poverty areas were higher than low-poverty areas, regardless of rural or urban classification. Results reinforce the importance of prioritizing high-poverty rural areas, especially in the South, in efforts to reduce CVD mortality. These efforts may need to consider socioeconomic conditions and region, in addition to rural and urban disparities.

Refined Land Use Classification for Urban Core Area from Remote Sensing Imagery by the EfficientNetV2 Model

In the context of accelerated urbanization, assessing the quality of the existing built environment plays a crucial role in urban renewal. In the existing research and use of deep learning models, most categories are urban construction areas, forest land, farmland, and other categories. These categories are not conducive to a more accurate analysis of the spatial distribution characteristics of urban green space, parking space, blue space, and square. A small sample of refined land use classification data for urban built-up areas was produced using remote sensing images. The large-scale remote sensing images were classified using deep learning models, with the objective of inferring the fine land category of each tile image. In this study, satellite remote sensing images of four cities, Handan, Shijiazhuang, Xingtai, and Tangshan, were acquired by Google Class 19 RGB three-channel satellite remote sensing images to establish a data set containing fourteen urban land use classifications. The convolutional neural network model EfficientNetV2 is used to train and validate the network framework that performs well on computer vision tasks and enables intelligent image classification of urban remote sensing images. The model classification effect is compared and analyzed through accuracy, precision, recall, and F1-score. The results show that the EfficientNetV2 model has a classification recognition accuracy of 84.56% on the constructed data set. The testing set accuracy increases sequentially after transfer learning. This paper verifies that the proposed research framework has good practicality and that the results of the land use classification are conducive to the fine-grained quantitative analysis of built-up environmental quality.

Extreme Weather Injuries and Fatalities, 2006 to 2021

Extreme weather has major implications for state and national health care systems; however, statistics examining weather-related injuries and fatalities are limited. To examine the frequency and regional distribution of major disaster events (MDEs) in the US. This ecologic cross-sectional study of MDEs occurring between January 1, 2006, and December 31, 2021, evaluated US data on all injuries and fatalities included in the National Oceanic and Atmospheric Administration National Centers for Environmental Information Storm Events Database. The data analysis was performed between February 22, 2023, and April 1, 2024. Major disaster events defined as an environmental event that caused either at least 50 injuries or at least 10 deaths. All MDEs were evaluated using descriptive statistics for event type, property damage, and rural or urban classification according to the National Centers for Health Statistics Urban-Rural Classification Scheme for Counties. The location of events according to Administration for Strategic Preparedness and Response (ASPR) region and hospital bed capacity of ASPR regions were also examined. Between 2006 and 2021, 11 159 storm events caused 42 254 injuries and 9760 deaths. Major disaster events accounted for 209 weather events (1.9%) but caused 19 463 weather-associated injuries (46.1%) and 2189 weather-associated deaths (22.4%). The majority of MDEs were caused by extreme heat (86 [41.1%]) and tornadoes (67 [32.1%]). While a larger proportion of MDEs occurred in urban areas (151 [75.1%]) vs rural areas (50 [24.9%]), rural MDEs caused a median of 9 (IQR, 2-16) deaths per event vs 4 (IQR, 0-14) deaths per event in urban areas. The majority of MDEs occurred in either ASPR region 4 (51 [24.5%]) or region 9 (45 [21.6%]). Certain event types, such as fires, wind, and hurricanes or storms, were geographically concentrated, while extreme heat and floods affected regions across the US equally. Urban counties had disproportionately greater hospital bed capacity than rural counties relative to population and MDE distributions. The findings of this ecologic study indicate that while MDEs accounted for a small proportion of all weather events, they were associated with a disproportionate number of injuries and fatalities. Integrating these data into county, state, and regional hazard vulnerability analyses is crucial to ensuring preparedness and mitigating climate risk.

An investigation of inequalities in exposure to PM2.5 air pollution across small areas in Ireland

The link between exposure to air pollution and adverse effects on human health is well documented. Yet, in a European context, research on the spatial distribution of air pollution and the characteristics of areas is relatively scarce, and there is a need for research using different spatial scales, a wider variety of socioeconomic indicators (such as ethnicity) and new methodologies to assess these relationships. This study uses comprehensive data on a wide range of demographic and socioeconomic indicators, matched to data on PM2.5 concentrations for small areas in Ireland, to assess the relationship between social vulnerability and PM2.5 air pollution. Examining a wide range of socioeconomic indicators revealed some differentials in PM2.5 concentration levels by measure and by rural and urban classification. However, statistical modelling using concentration curves and concentration indices did not present substantial evidence of inequalities in PM2.5 concentrations across small areas. In common with other western European countries, an overall decline in the levels of PM2.5 between 2011 and 2016 was observed in Ireland, though the data indicates that almost all small areas in Ireland were found to have exceeded the World Health Organization (WHO)’s PM2.5 annual guideline (of 5 µg/m3), calling for greater policy efforts to reduce air pollution in Ireland. The recent Clean Air Strategy contains a commitment to achieve the WHO guideline limits for PM2.5 by 2040, with interim targets at various points over the next two decades. Achieving these targets will require policy measures to decarbonise home heating, promote active travel and the transition to electric vehicles, and further regulations on burning fossil fuels and enforcing environmental regulations more tightly. From a research and information-gathering perspective, installing more monitoring stations at key points could improve the quality and spatial dimension of the data collected and facilitate the assessment of the implementation of the measures in the Clean Air Strategy.

Machine learning based urban land cover classification using PolInSAR data: a study with ALOS-2 and RADARSAT-2 datasets

A substantial variation in the land cover dynamics has been observed as a consequence of increasing urban expansion. Polarimetric synthetic aperture radar (PolSAR) data is widely being used for land cover studies in urban areas due to its all-weather, day-and-night imaging capabilities. However, in densely built-up areas, challenge arises with buildings having large Azimuth Orientation Angles (AOAs). These buildings are often misclassified as vegetation due to the depolarization of radar signal causing volumetric scattering response from the structures. This study addresses this issue by proposing an approach that integrates polarimetric information with interferometric SAR (InSAR) coherence to improve the differentiation between urban structures and vegetated areas, enhancing the accuracy of urban land-cover classification. Vegetated areas exhibit lower temporal coherence due to changes in the orientation of its leaves and branches caused by wind, seasonal variations, growth phenology, and other factors. In contrast, urban structures, being relatively stable targets, maintain high temporal coherence values. In present research various decomposition and scattering parameters were evaluated, along with PolInSAR coherence derived from L-band (ALOS-2) and C-band (RADARSAT-2), using two machine learning algorithms namely, Random Forest (RF) and Convolutional Neural Network (CNN). The C-band RADARSAT-2 data, particularly with six-component decomposition parameters, performed better, achieving an overall accuracy as 85.85% using RF algorithm. To further improve classification results, optical datasets from Landsat constellation were fused with SAR parameters using Gram-Schmidt fusion technique. This fusion led to significant improvements, achieving an overall accuracy of 94.50% and kappa statistics of 0.92, when CNN algorithm was applied to the fused optical and C-band RADARSAT-2 dataset. These results demonstrate the effectiveness of combining PolInSAR and optical data for more accurate urban land-cover classification, particularly in complex urban environments.