Stable median hydrologic performance of a layered bioretention cell despite increasing variability: a five-year experimental study.

Long-term reliability of blue-green infrastructure is key issue to urban greening. This study presents the results of a long-term experiment conducted on a small bioretention cell without exfiltration, designed as a multilayered constructed soil system planted with four perennial species, located in the Czech Republic. Water balance, water retention capacity of the bioretention cell, soil water potential, and plant growth were monitored over five years. The study aimed to elucidate how perennial plant development and changes in the material properties of subsurface layers influence the retention and detention of water within the bioretention cell. A semi-quantitative method based on a vegetation-derived landscape coefficient was used to estimate the effect of inter-annual vegetation development on evapotranspiration. Rainfall-runoff episodes were individually analyzed to assess episodic runoff coefficient, peak flow reduction, peak runoff delay, and runoff delay. The results indicate stable hydrological performance over the monitoring period, with variations in runoff response associated with evapotranspiration dynamics and antecedent moisture conditions. Peak flow reduction ranged from 30% to 100%, with a median of 88%, remaining consistently high throughout the study period. Incorporation of a vegetation-based landscape coefficient, derived through semi-quantitative assessment of photographic documentation, enhanced the accuracy of evapotranspiration estimation, particularly during the initial phases of vegetation establishment. The median runoff delay was approximately 30 min, while the median peak runoff delay reached 56 min. No evidence of performance deterioration was observed, indicating stable functioning of the system over time.