Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 31, 2026ISPRS annals of the photogrammetry, remote sensing and spatial information sciencesOpen Access

Assessing the Feasibility of Landsat-Driven NO 2 Prediction: A Spatial Cross-Validation Framework

View Full Paper
Ask AI
Bookmark
Share

Authors

ATAmir TahooniUniversity of TehranAKAta A. KakroodiUniversity of TehranMKMajid KiavarzUniversity of Tehran

Discussion

Loading...

Member takes

Implication

Randomized trial assesses nitrogen dioxide prediction in Tehran, highlighting limitations of models.

Key Points

  • This research aims to evaluate the feasibility of predicting ground-level nitrogen dioxide (NO₂) using Landsat imagery and other variables.
  • Developed Random Forest and XGBoost models using 21 predictors from Landsat 8/9, ASTER DEM, and OpenStreetMap in Tehran, Iran.
  • Employed three cross-validation strategies, including traditional 10-fold and rigorous spatial methods like leave-one-station-out (LOSO) and cluster-based CV.
  • Analyzed model performance metrics focusing on validation method effects.
  • 10-fold CV yielded optimistic R² values (0.21-0.26), whereas LOSO CV showed a pooled RMSE of ≈45.5 μg/m³ and a negative mean R², suggesting poor predictions at unseen locations.
  • Both algorithms exhibited comparable accuracy, but XGBoost demonstrated greater resilience against overfitting.
  • Landsat-derived proxies showed broad patterns in NO₂ estimation but struggled with fine-scale local variations.

Cite This Study

Tahooni et al. (2026) studied this question.

synapsesocial.com/papers/6a1bd03d5783ba022b6fc0b2https://doi.org/10.5194/isprs-annals-x-4-w8-2025-751-2026
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Regionally adaptive representation learning for surface-level nitrogen dioxide reconstruction from satellite and auxiliary data2026
  2. 2Cross-sectional Analysis and Machine Learning Modeling of Ground-Truth and Satellite-Derived NO₂ Concentrations in Temperate Climate Zone2026
  3. 3Developing a Machine Learning Approach for Estimating Urban NO2 Concentrations Using Multi-Source Remote Sensing Data2026
  4. 4Cross-City Transfer Learning for Sentinel-5P-Driven NO 2 Prediction in Data-Sparse Urban Environments2026
  5. 5Estimating surface-level NO2 concentrations in the Madrid region using Sentinel-5P observations and ground-based meteorological data with machine learning approaches2026 · 1 citations