Lake Srebarna is a Danube floodplain lake of exceptional conservation value protected under multiple international conservation frameworks, including the UNESCO World Heritage Convention, the Ramsar Convention, the UNESCO Biosphere Reserve Programme and the Natura 2000 network. Owing to its exceptional ecological importance as a model ecosystem for long-term ecological research, Lake Srebarna is also included in both the Bulgarian Long-Term Ecosystem Research (LTER-Bulgaria) network and the European Long-Term Ecosystem, Critical Zone and Socio-Ecological Research (eLTER) network, providing a valuable platform for long-term monitoring of ecological changes across multiple ecosystem components. Within this long-term monitoring framework, understanding spatial variability in phytoplankton communities is essential for interpreting ecosystem processes, detecting ecological change and supporting reliable ecological status assessment. The present study investigated the spatial organization of phytoplankton functional communities across the main sectors of Lake Srebarna during two consecutive years representing contrasting hydrological conditions to examine how extreme drought influenced spatial patterns in functional composition and the ecological mechanisms underlying relationships among functional composition, biomass and diversity. Phytoplankton was sampled during the vegetation seasons of 2024 and 2025 at four monitoring sites including two open-water sites: Central Lake area (CLA) and Dragayka pool (DRP) and two sites influenced by channels: South Sluice (SS) and Western Channel (WCH). Spatial comparisons were based on sampling months common to all stations within each year. Samples were processed using standard phytoplankton methods. Community composition was analyzed using the Reynolds–Padisák functional group approach. For each sample, total phytoplankton biomass and the Shannon diversity index (H′) were calculated. Spatial patterns in phytoplankton functional composition were visualized using non-metric multidimensional scaling (NMDS) based on Bray–Curtis dissimilarity and tested by one-way PERMANOVA with pairwise comparisons. Functional groups contributing most to spatial differentiation were identified using SIMPER analysis. Differences in total phytoplankton biomass and Shannon diversity among sampling sites were evaluated using the non-parametric Friedman test based on sampling months common to all stations, followed by pairwise comparisons with Bonferroni correction. All statistical analyses were performed using PAST version 5. NMDS ordination (stress = 0.14) revealed a more pronounced separation of samples among stations in 2025 than in 2024 Fig. 1. This pattern was confirmed by PERMANOVA, which indicated significant spatial differences in phytoplankton functional composition in 2025 (F = 2.47, p = 0.003), but not in 2024 (F = 1.15, p = 0.31). Following Bonferroni correction, only the CLA–SS pairwise comparison remained significant. SIMPER analysis indicated that the CLA–SS differentiation was driven mainly by higher contributions of functional groups X1 (small Chlorococcales) and X2 (small flagellates) at the CLA, whereas the SS station was characterized by greater contributions of the cyanobacterial group Sn ( Raphidiopsis raciborskii ) together with C ( mainly Cyclotella spp .) , W1 (Euglenophyceae) and P ( mainly Aulacoseira granulata , Fragilaria sp. ) , jointly explaining more than two-thirds of the average Bray–Curtis dissimilarity. Although spatial differences in phytoplankton biomass Fig. 2 were not statistically significant (Friedman test, p > 0.05), the SS station showed an opposite temporal pattern to the other stations. In contrast, Shannon diversity exhibited a significant overall spatial effect only in 2025 (Friedman test, p = 0.04). At the DRP and SS stations, high phytoplankton biomass was associated either with taxonomically and functionally diverse assemblages (July) or with extremely low diversity during near-monodominance of Raphidiopsis raciborskii (September). In contrast, the CLA station was characterized by extreme shallowing, increased conductivity and high-water transparency, where pronounced environmental stress restricted the community to a small number of tolerant taxa. Under these conditions, both phytoplankton biomass and diversity remained very low, consistent with strong environmental filtering rather than competitive dominance. Extreme drought conditions in 2025 generated contrasting phytoplankton responses within Lake Srebarna. Despite the absence of significant spatial differences in total phytoplankton biomass, functional composition and species diversity varied among sampling stations. The observed patterns demonstrate that similar reductions in diversity may arise through fundamentally different ecological mechanisms, including competitive dominance by Raphidiopsis raciborskii and environmental filtering under severe drought-induced environmental stress. These findings highlight the context-dependent relationship between phytoplankton biomass and diversity in shallow floodplain lakes exposed to extreme hydrological events.