Abstract:Satellite imagery has been widely used to map urbanization processes. To address the urgent need for urban landscape mapping that goes beyond urban footprint analysis, the local climate zone (LCZ) scheme has been increasingly used to reveal the urban forms and functions important to urban heat islands and micro-climates across the globe. As with most supervised classification strategies, proper application of training data is critical for the success of LCZ classification models. However, the collection and application of LCZ training areas brings with it two challenges that may affect mapping success. First, because digitizing training areas is a time-consuming task, there is a broad effort in the LCZ mapping community to create a crowdsourced data collection among different experts. However, this strategy likely leads to inconsistencies in labels that could weaken models. Second, the LCZ labeling process typically involves the delineation of large zones from which multiple training samples are drawn, but those samples are likely spatially autocorrelated and lead to overly optimistic estimates of model accuracy. Although both effects -- inconsistent labeling and spatial autocorrelation -- are theoretically possible, it is unknown whether they substantially affect accuracy. We investigated both issues, specifically asking: (i) how do the discrepancies of LCZ labeling by different experts impact broad-scale LCZ mapping? (ii) to what extent does spatial correlation affect model prediction power? We used two classifiers (Random Forests and ResNets) to map eight metropolitan areas in the US into LCZs, comparing training areas drawn by different or consistent interpreters, and data splitting strategy using rules that allow or reduce spatial autocorrelation. We found large discrepancies among results built from crowdsourced training areas digitized by different experts; improving the consistency of labels can lead to substantial improvements in LCZ classification accuracy. Second, we found that spatial autocorrelation can boost the apparent accuracy of the classifier by 16% to 21%, leading to erroneous interpretation of mapping results. The two effects interplay as well: spatial autocorrelation in the raw data can lead to an underestimation of the model's predictive error when modeling with crowdsourced training areas of high inconsistency. Due to the uncertainty in the labeling process and spatial autocorrelation in derived training data, broad-scale LCZ mapping results should be interpreted with caution.

A Co-Training Approach to the Classification of Local Climate Zones with Multi-Source Data.

Machine learning-assisted mapping of city-scale air temperature: Using sparse meteorological data for urban climate modeling and adaptation

Urban surface classification using self-supervised domain adaptive deep learning models and its application in urban environment studies

High Spatial-Resolution Classification of Urban Surfaces Using a Deep Learning Method

Local Climate Zone Mapping by Coupling Multilevel Features With Prior Knowledge Based on Remote Sensing Images

Mapping Local Climate Zones in the Urban Environment: the Optimal Combination of Data Source and Classifier.

Local Climate Zone Classification by Seasonal and Diurnal Satellite Observations: An Integration of Daytime Thermal Infrared Multispectral Imageries and High-Resolution Night-Time Light Data

Self-Training Classification Framework with Spatial-Contextual Information for Local Climate Zones

Application of training data affects success in broad-scale local climate zone mapping

Local climate zone mapping using remote sensing: a synergetic use of daytime multi-view Ziyuan-3 stereo imageries and Luojia-1 nighttime light data

Local Climate Zone Classification with Different Source Data in Xi'an, China

Assessment of Local Climate Zone Classification Maps of Cities in China and Feasible Refinements

Mapping and analyzing the local climate zones in China’s 32 major cities using Landsat imagery based on a novel convolutional neural network

Integrating Satellite and Street-Level Images for Local Climate Zone Mapping.

Advances of Local Climate Zone Mapping and Its Practice Using Object-Based Image Analysis

SAR and Multi-Spectral Data Fusion for Local Climate Zone Classification with Multi-Branch Convolutional Neural Network

Deep learning-based local climate zone classification using Sentinel-1 SAR and Sentinel-2 multispectral imagery

Deep Transfer Model Based Local Climate Zone Classification Using SAR/Multispectral Images

Multicascaded Feature Fusion-Based Deep Learning Network for Local Climate Zone Classification Based on the So2Sat LCZ42 Benchmark Dataset

Local Climate Zone in Xi’an City: A Novel Classification Approach Employing Spatial Indicators and Supervised Classification

Local Climate Zone Classification via Semi-Supervised Multimodal Multiscale Transformer