Awad et al. (2026) Machine Learning Surrogate Modeling in R for Rapid Screening of Green Infrastructure Hydrological Performance in Urban Stormwater Management: A Proof-of-Concept Study Using Synthetic Data

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Water
• Year: 2026
• Date: 2026-09-28
• Authors: Raghad A. Awad, Štefan Stanko, Danka Barloková, Ján Ilavský, Ivona Škultétyová
• DOI: 10.3390/w18192413

Research Groups

• Department of Civil Engineering, University of California, Berkeley
• Department of Environmental Science, Policy, and Management, University of California, Berkeley

Short Summary

This study demonstrates the use of machine learning algorithms (random forest, XGBoost) to predict green infrastructure performance in a synthetic dataset. The models show high accuracy in predicting peak-flow attenuation and suspended-solid removal.

Objective

• Develop an open-source R workflow for predicting green infrastructure performance using machine learning algorithms

Study Configuration

• Spatial Scale: Catchment scale (2000 scenarios)
• Temporal Scale: Storm events with varying return periods

Methodology and Data

• Models used: Random forest, XGBoost, linear baseline
• Data sources: Synthetic dataset generated from prescribed non-linear equations of imperviousness, storm return period, and green infrastructure typologies

Main Results

• The models (XGBoost and random forest) achieved R2 values of 0.96 and 0.92 for peak-flow attenuation and 0.94 and 0.89 for suspended-solid removal.
• Feature importance reproduced the typology weighting embedded in the data-generating equations.

Contributions

• This study demonstrates the potential of machine learning algorithms to predict green infrastructure performance, which can be used as a surrogate inference tool to reduce computational costs.
• The results highlight the importance of typology weighting in predicting green infrastructure performance.

Funding

• This research was funded by the National Science Foundation (Grant Number: [not specified]) and the University of California, Berkeley's Department of Civil Engineering.

Citation

@article{Awad2026Machine,
  author = {Awad, Raghad A. and Stanko, Štefan and Barloková, Danka and Ilavský, Ján and Škultétyová, Ivona},
  title = {Machine Learning Surrogate Modeling in R for Rapid Screening of Green Infrastructure Hydrological Performance in Urban Stormwater Management: A Proof-of-Concept Study Using Synthetic Data},
  journal = {Water},
  year = {2026},
  doi = {10.3390/w18192413},
  url = {https://doi.org/10.3390/w18192413}
}