Oz Kira

Urban vegetation contributes to climate regulation and environmental quality, but its seasonal activity is difficult to compare across heterogeneous city landscapes. This study assessed the city-region monthly correspondence among TROPOMI solar-induced chlorophyll fluorescence (SIF), Sentinel-2 NDVI and NIRv, the radiation-scaled index NIRvP, and model-derived MODIS gross primary production (GPP) in Tel Aviv, Rome, Washington DC, Johannesburg, Beijing, and Buenos Aires from January 2019 to October 2020. ESA WorldCover 2020 was used to compare full-region and vegetation-masked Sentinel-2 indices. Raw SIF–GPP relationships were strong across the six cities (R² = 0.772–0.950), and NIRvP also showed strong raw correspondence in several cities. However, harmonic residualization and PAR-controlled analyses showed that much of the raw correspondence reflected shared annual seasonality and radiation forcing. Adjusted SIF–GPP relationships remained statistically significant only in Rome and Johannesburg. NIRvP had higher adjusted correlations than NIRv in Tel Aviv, Rome, and Johannesburg, but lower correlations in Washington DC, Beijing, and Buenos Aires, indicating no consistent cross-city advantage. Direct comparison of full-region and vegetation-masked results produced a median absolute ΔR² of 0.011 across 18 city-by-index comparisons, although larger changes occurred for NDVI in Buenos Aires and NIRv in Tel Aviv and Buenos Aires. TROPOMI footprint-selection effects were generally small where many footprints were available but were larger in Tel Aviv and Buenos Aires under strict vegetation thresholds. The findings therefore describe period-specific seasonal consistency and processing sensitivity among satellite-derived indicators at a coarse city-region scale, rather than validation of true urban GPP or fine-scale urban vegetation productivity.