Hydrological performance and design of bioretention systems for heavy rainfall management: A laboratory study

Bioretention systems are widely used for urban stormwater management, yet their performance under intense rainfall—especially in tropical regions—remains underexplored. This study evaluated the hydrological performance of three full-scale bioretention cells (100 × 50 × 70 cm) with varying soil–sand...

Full description

Saved in:
Bibliographic Details
Main Authors: Muhammad Baitullah Al Amin, Joko Sujono, Radianta Triatmadja
Format: Article
Language:English
Published: Elsevier 2025-12-01
Series:Nature-Based Solutions
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772411525000540
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Bioretention systems are widely used for urban stormwater management, yet their performance under intense rainfall—especially in tropical regions—remains underexplored. This study evaluated the hydrological performance of three full-scale bioretention cells (100 × 50 × 70 cm) with varying soil–sand compositions, tested using a custom rainfall simulator. Saturated hydraulic conductivities ranged from 63.3 to 325.6 mm/hr. The Storm Water Management Model (SWMM) was used to simulate the bioretention cells and was calibrated and validated against experimental data, showing strong agreement (r > 0.9; NSE > 0.8). Results indicated that standard designs (100–300 mm/hr conductivity; 5% area coverage) were insufficient for mitigating peak flows under heavy rainfall events (P > 100 mm; iave > 10 mm/hr). Increasing area coverage from 5% to 30% reduced peak discharge by up to 50%, delayed peak runoff by 82 min, and extended detention time. Runoff volume reductions ranged from 2.1–11.2% for 2-year design storms and 1.4–6.8% for 50-year events. An area coverage of 10–20% is recommended for effective mitigation. Dimensionless empirical equations were developed for design applications, and refinements to SWMM’s percolation modeling are suggested to improve model accuracy.
ISSN:2772-4115