Answers are grounded in this paper and cite their sources.
Hydrological drought in Serbia has been studied only in a few basins. We compiled monthly discharge for 27 gauges (84–210,250 km2) for 1991–2024 and transcribed the 1991–2003 values from scanned yearbooks. Using a non-parametric Standardised Streamflow Index, we tested three hypotheses on drought scale dependence and synchrony, accounting for spatial autocorrelation and multiple testing. We hypothesised that drought duration would increase with catchment size; instead, mean drought duration decreased with drainage area (Spearman ρ = −0.72; 95% CI −0.85 to −0.51; pFDR = 0.012), and event frequency increased (ρ = 0.71; pFDR = 0.013). The association weakened with 3-month accumulation and was not detected with 12-month accumulation, for severe droughts or without the rescued 1991–2003 data. Drought synchrony decreased with distance (Mantel r = −0.60; pFDR < 0.001; excess above independence halving at ≈190 km) and was higher between nested gauges (Jaccard index +0.13 at equal distance). After accounting for distance, droughts on the Danube, Tisa and Sava were not significantly less synchronous with domestic rivers than domestic rivers were with each other. These results suggest that, at the monthly scale, drought at large-river gauges is fragmented into shorter events, so monitoring there may under-represent drought persistence in small catchments.
No takes yet. Share an insight, caveat, or question.
Diaconu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: