The group lies at elevations of 2,095 to 2,535 metres and attracts more than 2,000 tourists daily during the summer season. Under the Sustainable Management of Rila National Park, Phase II project, the authority established three treated areas in the littoral zone of each lake. Each area covered 400 square metres and was divided into four 100-square-metre squares; vegetation and sediment were removed from two diagonally positioned squares to permit within-area comparisons.

Work occurred from July to September 2022 and from July to August 2023. Workers manually removed whole plants and transferred them ashore in an inflatable boat. The vegetation was dried on polyethylene sheets and deposited outside the park.

An average of 80.2 ± 13.7 g fresh weight of Ranunculus aquatilis was removed per square metre. A hand-held water lift extracted only fine, loose sediment without disrupting shore profiles or deepening lake bottoms. Dredged material was retained in dewatering filter bags onshore, with seepage returned to the lakes, and averaged 1.16 ± 0.32 kg per square metre.

Hiking trails in the watershed were also gravelled and marked to reduce erosion. The researchers sampled in August 2015, October 2021 and July 2022 to characterize pre-restoration conditions. Post-restoration surveys occurred in August 2023, approximately one week after work; October 2023, about one month afterward; and October 2024, one year after the final treatment.

Measurements included dissolved oxygen, temperature, pH, electrical conductivity, ammonium nitrogen, nitrate nitrogen, phosphate phosphorus, total nitrogen, total phosphorus and chlorophyll-a. Chlorophyll-a was used as an indicator of phytoplankton biomass. Bacterial community function was assessed with Biolog EcoPlates containing 31 carbon sources.

The analysis yielded average well colour development, community-level physiological profiles, functional richness and Shannon-Weaver diversity indices. Statistical analyses included repeated-measures ANOVA, Pearson correlation, ordination, two-way PERMANOVA and SIMPER, with significance assessed at p < 0.05. Nutrient concentrations generally decreased during the post-restoration period.

Reported average reductions were 14% for phosphate phosphorus and total phosphorus, 12% for total nitrogen and 58% for inorganic nitrogen. By October 2024, most nutrient measurements were near pre-restoration levels, except phosphate phosphorus in Bliznaka and Trilistnika. The share of inorganic nitrogen in total nitrogen declined from an average of about 82% before restoration to 33% in October 2023 and partially recovered to 52% in October 2024.

The researchers attributed part of this variation to seasonal changes in phytoplankton growth, grazing and thermal stratification, while temporary sediment resuspension was recorded in limited areas. Phytoplankton responses differed among lakes immediately after treatment. In August 2023, chlorophyll-a decreased by 32% in Bliznaka and 48% in Trilistnika but increased by 170% in Ribnoto and 26% in Dolnoto; all four changes differed significantly from pre-restoration values at p < 0.05.

By October 2024, chlorophyll-a had returned to or fallen below pre-restoration levels in all four lakes. Dolnoto shifted from mesotrophic before treatment toward the boundary between oligotrophic and mesotrophic conditions. Temperature had the most consistent association with chlorophyll-a, with correlation coefficients above 0.79, and the authors considered low autumn temperatures a constraint on phytoplankton growth.

Bacterial profiles also changed and then moved toward their pre-restoration pattern by October 2024. Average well colour development generally declined, remaining at 87 ± 1.3% of pre-restoration levels after one year, with Trilistnika approaching its earlier value. Bacterial communities initially favoured carbohydrates and polymers, whereas post-restoration profiles shifted toward carboxylic acids and, in some periods, amino acids.

The number of usable carbon sources fell by 21% before showing partial recovery, while functional diversity followed a similar pattern; richness and diversity were strongly correlated at r = 0.89 and p < 0.0001. PERMANOVA found significant effects of sampling time and lake identity but no significant interaction. The study’s bounded design limits attribution of all observed changes to restoration because pronounced seasonal variation occurred over the same period.

It covered four treated lakes and measurements through October 2024, rather than the longer-term ecological effects of the intervention. Within that scope, the monitored indicators did not show a major or persistent deterioration in water quality, while ordination and community-profile analyses indicated convergence toward pre-restoration bacterial function by the end of the first year. The authors stated that monitoring would continue to assess long-term effects.

Subject of Research: One-year monitoring of experimental plant and sediment removal in four Bulgarian high-mountain lakes

Article Title: Restoration of high-mountain lakes: effects on nutrient concentrations and biological communities

Article References: Boteva, S., Traykov, I., Angelova, B., & Kenarova, A. (2025). Restoration of high-mountain lakes: effects on nutrient concentrations and biological communities. One Ecosystem, 10, Article e153673. https://doi.org/10.3897/oneeco.10.e153673

Image Credits: AI Generated

DOI: 10.3897/oneeco.10.e153673

Keywords: high-mountain lakes, lake restoration, macrophyte harvesting, sediment removal, nutrient concentrations, phytoplankton biomass, bacterial metabolism, Rila Mountains, Bulgaria, water quality, Restoration, high-mountain