Skip to content
DraconiticResearch
Search Sources API
226,742 studies31 sourcesUpdated 51 min ago
PreprintCC BY 4.0EarthArXiv

Uncertainty-aware deep learning for national lightning hazard and risk mapping to support sustainable disaster risk reduction in Bangladesh

Abstract

Bangladesh has one of the world's highest lightning death tolls, borne mostly by farmers and fishers, yet national hazard maps rest on interpolation or tabular machine learning without independent validation. We developed a raster-native deep-learning framework reproducing monthly lightning stroke density from the WWLLN Global Lightning Climatology (WGLC, 2012-2025) on a 5-arc-minute grid from 18 meteorological, aerosol and surface predictors, comparing U-Net, Attention U-Net, ConvLSTM and a ConvLSTM-U-Net hybrid under a predefined protocol. On unseen years (2023-2025), every architecture outperformed monthly climatology, and a 12-member ensemble cut the mean squared error by 20 % and raised the hotspot critical success index from 0.15 to 0.27, whereas a convective available potential energy (CAPE) × precipitation proxy failed on hotspots (0.08). Recalibrated ensemble spread gave 90 % coverage of 95 % intervals. The ensemble matched the independent TRMM-LIS/ISS-LIS satellite climatology (r = 0.53) about as well as the ground record did, and Integrated Gradients ranked CAPE first in all 12 runs. A change-point test found a 2016 detection step (×1.96) in the observed record absent from the meteorological drivers, so no trend is claimed. With hazard, exposure and vulnerability combined under propagated uncertainty, risk peaks in the north-eastern haor basin, where 50 upazilas (15.5 million people) have probability ≥ 0.9 of being high or very high risk. District deaths follow pre-monsoon risk (ρ = 0.40) more closely than annual risk (ρ = 0.19), supporting harvest-timed warnings and shelters targeted to these upazilas under Sustainable Development Goal targets 1.5, 11.5 and 13.1.

Related

  1. Corporate governance and Environmental, Social, and Governance practices as drivers of alignment with the sustainable development goals: Evidence from Bangladesh
    Emon Kalyan Chowdhury
    Innovation and Green DevelopmentPeer-reviewedCC BYFull textEnvironmental science & pollutionDOAJ
  2. Knowledge, attitudes, and practices of farmers towards sustainable agriculture in the Barind Tract of Bangladesh
    A K M Kanak Pervez, Md Mahedi, Ashfaq Ahmad Shah +2
    Discover AgriculturePeer-reviewedCC BYFull textEnvironmental science & pollutionDOAJ
  3. Economic complexity and energy uncertainty in the G7: Implications for sustainable development and energy security
    Tomiwa Sunday Adebayo, Berna Uzun, Oktay Özkan
    Innovation and Green DevelopmentPeer-reviewedCC BYFull textEnvironmental science & pollutionDOAJ
  4. Reduction of uncertainty in near-term climate forecast by combining observations and decadal predictions
    Rémy Bonnet, Julien Boé, Emilia Sanchez-Gomez +1
    Earth System DynamicsPeer-reviewedCC BY 4.0Full textClimate scienceOceanography & cryosphereHAL
  5. Estimating aman and aus rice area in Bangladesh using sentinel-1 imagery and machine learning algorithms
    H. M. Hamidur Rahman, Hasan Mahmud, Md Mahfuzul Hasan +1
    PLOS OnePeer-reviewedCC BY 4.0Full textClimate scienceHydrology & soilsPLOS
BibTeX
@misc{faruque2026uncertaintyaware,
  title = {{Uncertainty-aware deep learning for national lightning hazard and risk mapping to support sustainable disaster risk reduction in Bangladesh}},
  author = {Fardin Bin Faruque and Kazi Tahmid Parveg and Jannatul Ferdous},
  howpublished = {EarthArXiv},
  year = {2026},
  month = nov,
  doi = {10.31223/x57f83},
  url = {https://eartharxiv.org/repository/view/16095/},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - Uncertainty-aware deep learning for national lightning hazard and risk mapping to support sustainable disaster risk reduction in Bangladesh
AU  - Fardin Bin Faruque
AU  - Kazi Tahmid Parveg
AU  - Jannatul Ferdous
PY  - 2026
DA  - 2026/11/13
PB  - EarthArXiv
DO  - 10.31223/x57f83
UR  - https://eartharxiv.org/repository/view/16095/
AB  - Bangladesh has one of the world's highest lightning death tolls, borne mostly by farmers and fishers, yet national hazard maps rest on interpolation or tabular machine learning without independent validation. We developed a raster-native deep-learning framework reproducing monthly lightning stroke density from the WWLLN Global Lightning Climatology (WGLC, 2012-2025) on a 5-arc-minute grid from 18 meteorological, aerosol and surface predictors, comparing U-Net, Attention U-Net, ConvLSTM and a ConvLSTM-U-Net hybrid under a predefined protocol. On unseen years (2023-2025), every architecture outperformed monthly climatology, and a 12-member ensemble cut the mean squared error by 20 % and raised the hotspot critical success index from 0.15 to 0.27, whereas a convective available potential energy (CAPE) × precipitation proxy failed on hotspots (0.08). Recalibrated ensemble spread gave 90 % coverage of 95 % intervals. The ensemble matched the independent TRMM-LIS/ISS-LIS satellite climatology (r = 0.53) about as well as the ground record did, and Integrated Gradients ranked CAPE first in all 12 runs. A change-point test found a 2016 detection step (×1.96) in the observed record absent from the meteorological drivers, so no trend is claimed. With hazard, exposure and vulnerability combined under propagated uncertainty, risk peaks in the north-eastern haor basin, where 50 upazilas (15.5 million people) have probability ≥ 0.9 of being high or very high risk. District deaths follow pre-monsoon risk (ρ = 0.40) more closely than annual risk (ρ = 0.19), supporting harvest-timed warnings and shelters targeted to these upazilas under Sustainable Development Goal targets 1.5, 11.5 and 13.1.
ER  -