Abstract
This study presents an assessment of below-ground carbon dynamics of green infrastructure using artificial intelligence, targeting sub-tropical bioretention basins in South East Queensland, Australia. This extended abstract describes the context for the study and the significance of the work, which was recognised and enabled through the international Microsoft Artificial Intelligence (AI) for Earth Grants (2018 Grant winner). Four different scenarios were tested with three different approaches for modelling of the regression values. The three different machine learning methods were applied to predict belowground carbon and nitrogen, based on soil physical characteristics data entry. The neural network model performed better in predicting both the carbon and nitrogen concentration in all the scenarios. The implication of this study provides a profound shift in the type of platform that can be used, wherein machine learning methods can assist decision-makers in finding low-cost proxies for measuring carbon and nitrogen capture in bioretention basins.
Highlights
Bioretention basins are one of the most frequently implemented green stormwater devices with a promising performance in heavy metal, nutrient and nitrogen removal [1]
The neural network (NN) approach shows the ability to generate the soil carbon prediction models, in which 73% and 76% of the carbon values can be explained by this prediction in scenarios I and III
The ability of a linear regression model between the target parameters and each feature individually would not exceed 25%. This evidence demonstrates the substantial ability of the trained neural network model in predicting carbon/nitrogen values based on a set of physical features
Summary
Bioretention basins are one of the most frequently implemented green stormwater devices with a promising performance in heavy metal, nutrient and nitrogen removal [1]. Monitoring of the stormwater assets is required to determine the long-term performance and effectiveness of the designed system and to predict the required maintenance activities [4]. There is little understanding of the long-term soil quality condition assessment of bioretention basins in relation to other physical characteristics of soil filter media. This current study aimed to develop an artificial neural network model that uses different soil physical characteristics to predict soil carbon and nitrogen
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IOP Conference Series: Earth and Environmental Science
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
