Abstract

Independent monitoring of crucial infrastructure under restricted access requires remote sensing methods capable of inferring functional status from externally observable proxies. This study presents a multi-sensor satellite remote sensing analysis for tracking nuclear safety-relevant functions at the Zaporizhzhia Nuclear Power Plant (ZNPP), Europe's largest nuclear power plant that is currently operating under military occupation with constrained international oversight. Our study demonstrates an integrated approach across complementary spatial, spectral, and temporal scales to study key safety functions of nuclear power plants, primarily decay heat removal, cooling water availability, power supply reliability, and physical integrity. Multispectral water index analysis of Sentinel-2 and Landsat imagery documents a declining cooling pond surface-water extent following the destruction of the upstream Kakhovka Dam in June 2023, approaching a critical water threshold. High-resolution thermal infrared imagery from HotSat-1 resolves unit-level heat signatures, independently corroborating shutdown configurations. Dynamic Mode Decomposition of decadal Landsat 8/9 thermal time series isolates reactor-attributable anomalies from seasonal forcings, detecting a statistically significant shift following reactor shutdown. Sentinel-1 SAR change detection further identifies sustained infrastructure modifications at a key transmission substation consistent with grid integration activity. Our study demonstrates that integration of various sensors can provide systematic and independently verifiable monitoring of nuclear facility safety functions when ground-based verification is limited or denied, with direct implications for satellite-based verification within international nuclear safety frameworks.

Read the full article at sciencedirect.com