Answers are grounded in this paper and cite their sources.
In post-conflict cities, transformation of urban green infrastructure (UGI) may be spatially heterogeneous and poorly captured by single spectral indicators. We assessed post-2022 change in UGI in Irpin, Ukraine, after the local hostilities of February–March 2022, using summer composites of Sentinel-2, Sentinel-1 SAR, ERA5-Land and ESA WorldCover data for 2019–2024. Seven indicators were analysed: NDVI, EVI, NDMI, NBR, Sentinel-1 VH backscatter, NDBI and BSI. Annual analysis showed that city-wide summer NDVI, EVI and NDMI in 2022 remained within the 2019–2021 interannual range (standardized anomalies −0.41, +0.90 and + 0.62 SD), indicating no anomalous city-wide change in greenness. Nevertheless, NBR and Sentinel-1 VH identified localized disturbance signals within pre-conflict tree cover. A conservative multi-indicator disturbance proxy (agreement of at least two of three indicators) covered 1.9% of the reference tree-cover area, compared with 2.3–3.3% for individual indicators; sample-based visual validation confirmed 68.0% of mapped disturbance locations but revealed substantial omission (producer's accuracy 1.8%), indicating that the proxy captures a high-agreement core rather than the total extent of disturbance. Spectral signals of potential urban land-surface transformation (ULST) covered 27.2% of the study area by 2024 (11.5–39.0% across tested thresholds); validation indicated moderate accuracy (overall accuracy 63.8%) and an estimated transformed area of 39.5% (95% CI 29.3–50.2%). ERA5-Land showed no pronounced regional hydroclimatic anomaly in 2022, although local effects cannot be excluded. City-wide greenness can therefore remain stable while localized structural disturbance and extensive land-surface transformation occur, so NDVI-only assessments may overestimate the stability of post-conflict UGI. A multi-indicator optical–SAR framework with explicit uncertainty assessment provides a more reliable, though still preliminary, basis for urban recovery planning.
No takes yet. Share an insight, caveat, or question.
Strashok et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: