Research

Three connected research themes — GeoAI and multisource remote sensing, urban environmental health, and climate risk and community resilience.

My research program asks a single connected question: how do urbanization, climate extremes, and environmental inequality shape human health, community resilience, and sustainable development — and how can satellite observation and geospatial artificial intelligence answer that question at the scales where decisions are made?

The three themes below build on one another. Methods developed under GeoAI and multisource remote sensing supply the exposure and land-change evidence used in the two applied themes.

GeoAI and multisource remote sensing

I develop geospatial artificial intelligence methods that fuse optical, SAR, LiDAR, hyperspectral, and nighttime-light observations into consistent, decision-ready maps. This work combines machine learning and deep learning with physical understanding of the urban environment, and scales through cloud platforms such as Google Earth Engine.

Key questions

  • How can multisource satellite observations be fused when clouds, shadows, and sensor differences make any single source unreliable?
  • How do we make deep-learning-based mapping transparent enough to support environmental policy?

Methods. Deep learning and machine learning, data fusion, sparse representation, fuzzy-set approaches, Google Earth Engine, spatially explicit modelling

Scale. City to global; annual to intra-year time series

GeoAIremote sensingSARLiDARhyperspectralnighttime lightsdeep learning

Diagram of the UU-Net deep-learning architecture, with separate optical and SAR input branches merged into one road-extraction network

UU-Net fuses optical and SAR imagery to extract roads at large scale. Fig. 2 of Lin et al. (2021), Int. J. Appl. Earth Obs. Geoinf. (CC BY-NC-ND 4.0).

Selected publications

  1. Underestimated mortality risks from compound urban heat and air pollution islands in Global North cities
    Yinyi Lin, Hongsheng Zhang, TC Chakraborty, Shan Wei, Yanjia Cao, and 3 more authors
    npj Urban Sustainability, Jun 2026
  2. RSE
    Incorporating synthetic aperture radar and optical images to investigate the annual dynamics of anthropogenic impervious surface at large scale
    Yinyi Lin, Hongsheng Zhang, Hui Lin, Paolo Ettore Gamba, and Xiaoping Liu
    Remote Sensing of Environment, Jun 2020
  3. JAG
    Leveraging optical and SAR data with a UU-Net for large-scale road extraction
    Yinyi Lin, Luoma Wan, Hongsheng Zhang, Shan Wei, Peifeng Ma, and 2 more authors
    International Journal of Applied Earth Observation and Geoinformation, Dec 2021

All publications →

Urban environmental health

Cities concentrate people and environmental hazards. I quantify urban heat, air pollution, and their compound exposure at the scales where people actually live — buildings, neighbourhoods, and communities — and link them to mortality risk, health inequalities, and the social determinants that make some populations, including older adults, more vulnerable than others.

Key questions

  • How much larger are health risks when urban heat islands and air-pollution islands compound, and who bears them?
  • Which communities face the greatest combined environmental burden, and why?

Methods. Satellite-derived exposure mapping, structural equation modelling, multiscale spatial analysis, health-registry linkage

Scale. Building and community scale to global city samples

urban heatair pollutioncompound exposuremortalityenvironmental inequalityageing populations

World maps and charts showing how often and how intensely urban heat islands and air-pollution islands co-occur in 327 cities, comparing the Global North and Global South

Compound frequency and intensity of urban heat and air-pollution islands in 327 cities, 2003–2020. Fig. 1 of Lin et al. (2026), npj Urban Sustainability (CC BY-NC-ND 4.0).

Selected publications

  1. Underestimated mortality risks from compound urban heat and air pollution islands in Global North cities
    Yinyi Lin, Hongsheng Zhang, TC Chakraborty, Shan Wei, Yanjia Cao, and 3 more authors
    npj Urban Sustainability, Jun 2026
  2. SCS
    Precise mitigation strategies for urban heat island effect in Hong Kong’s new towns using automated machine learning
    Yiyan Li, Hongsheng Zhang, Yinyi Lin, Jing Ling, Huiyuan Xue, and 1 more author
    Sustainable Cities and Society, May 2025
  3. Linking joint exposures to residential greenness and air pollution with adults’ social health in dense Hong Kong
    Ting Zhang, Bo Huang, Sensen Wu, Jie Chen, Yizhen Yan, and 4 more authors
    Environmental Pollution, Dec 2024

All publications →

Climate risk and community resilience

Floods, droughts, and extreme temperatures increasingly arrive together. I map where compound and cascading climate risks concentrate, and use satellite evidence — from disaster-impact records to monthly nighttime lights — to measure how communities are disturbed by shocks and how quickly they recover, with attention to climate vulnerability and environmental justice.

Key questions

  • Where are communities least able to absorb compound floods, droughts, and heat extremes?
  • Can longitudinal satellite observations reveal resilience — disturbance and recovery — directly?

Methods. Disaster-impact spatialization, random forest, fuzzy-logic integration, time-series analysis of nighttime lights, high-resolution risk assessment

Scale. Community to global (multi-kilometre grids); monthly to multi-decadal

floodsdroughtsextreme temperaturecompound riskcommunity resilienceclimate adaptationenvironmental justice

World map of the Community-Climate Resilience Index (CCRI): higher resilience in western Europe, lower across much of South America, Africa, and Oceania

Community-Climate Resilience Index (CCRI) worldwide. From Lin et al. (2026), Earth's Future (CC BY-NC 4.0).

Selected publications

  1. Global Inequality and Potential Trade-Offs in Community Resilience to Compound Climate Extremes
    Yinyi Lin, Hongsheng Zhang, Shan Wei, Peng Gong, and Zhenci Xu
    Earth’s Future, Sep 2026
  2. Promoting sustainable development worldwide in the metacoupled anthropocene
    Qutu Jiang, Zhenci Xu, Nishan Bhattarai, Yinyi Lin, Huijuan Xiao, and 14 more authors
    Nature Communications, Feb 2026
  3. Coastal urbanization may indirectly positively impact growth of mangrove forests
    Shan Wei, Hongsheng Zhang, Zhenci Xu, Guanghui Lin, Yinyi Lin, and 6 more authors
    Communications Earth & Environment, Oct 2024

All publications →

Research vision

My long-term goal is to build high-resolution, explainable, and socially meaningful geospatial evidence — evidence that helps cities and communities adapt to climate extremes, protect public health, and grow more sustainably.

Getting there requires three things at once: observation systems that see every city and settlement consistently; models whose outputs scientists and policymakers can interrogate rather than take on faith; and research questions set by what matters for people — health, safety, and fair access to a livable environment. Each project I take on is chosen to advance at least one of these fronts.