Artificial Intelligence in Environmental Impact Assessment

It's a reality, we environmental consultants still evaluate environmental impacts using basic qualitative methods. The time has come to gradually introduce an innovative approach based on Artificial Intelligence (hereinafter AI) thus generating environmental impact assessment indices that statistically reflect the relationships between different environmental factors and the environment, providing patterns between the data, thus minimizing the influence of human bias, but without forgetting the experience of the environmental consultant as a crucial component in the evaluation process.

In the field of environmental impact assessment, AI can use machine learning algorithms to analyze large and complex data sets. They can also be trained to identify patterns and trends in data that may be difficult or impossible for humans to detect, making it possible to predict potential environmental impacts more rigorously.

Another way in which AI can help in environmental impact assessment is through natural language processing techniques, analyzing and extracting relevant information from large volumes of text, such as scientific reports, regulatory documents and public comments, thus streamlining the review process and ensuring that all relevant information (own or not) is considered in the evaluation.

To date, the main artificial intelligence works related to the environment are very diverse: those related to the assessment of biodiversity, the analysis of the life cycle, the modeling of environmental systems (air, groundwater,...) or the management of natural resources stand out, but above all, works related to the analysis of environmental data from satellites, drones, sensors, cameras,... as well as the optimization of sources and systems such as renewable energy or the design and application of sustainable solutions for waste management, among others.

Despite the large number of publications, the use cases of prediction in environmental assessment using artificial intelligence are smaller. Some representative and recent examples are, for example, in the field of mining, where a process has been applied. AutoML developed with a Bayesian network unsupervised capable of dynamically evaluating the potential environmental impacts of a shale mine in Galicia (Spain) surrounded by the Natura 2000 Network. This study made it possible to identify the factors of surface and subsurface runoff, shapes and volumes, species or individual concentration and degradation of grief as the possible nodes with the greatest environmental impact in the study area.

In the field of environmental acoustics, AI has been used to predict the environmental impact of tunnel blasting using ordinary artificial neural networks, particle swarms and optimized artificial neural networks.

Despite the potential benefits of AI to improve environmental impact assessment, we must recognize that these technologies are not a panacea and it is essential to ensure that the use of AI in environmental assessment does not undermine the importance of the judgment and experience of the environmental consultant since it is a crucial component in the evaluation process.

We are at the dawn of the use of AI in environmental assessment due, among other aspects, to the problem of data quality, algorithmic bias and the need for interdisciplinary collaboration between AI experts and environmental professionals, and that is why Ideas MedioAmbiental has already found a great travel companion with whom it has already begun to follow the path of artificial intelligence applied to environmental impact assessment.

Bibliography:

• Curmally, A., Sandwidi, B.W., & Jagtiani, A. (2022). Artificial intelligence solutions for environmental and social impact assessments. Handbook of Environmental Impact Assessment, 163.
• Gerassis, S., Giráldez, E., Pazo-Rodríguez, M., Saavedra, Á., & Taboada, J. (2021). AI approaches to environmental impact assessments (EIAs) in the mining and metals sector using AutoML and Bayesian modeling. Applied Sciences, 11 (17), 7914.
• Lawal, A.I., Kwon, S., & Kim, G.Y. (2021). Prediction of an environmental impact of tunnel blasting using ordinary artificial neural network, particle swarm and Dragonfly optimized artificial neural networks. Applied Acoustics, 181, 108122.
• Yassine Himeur, Bhagawat Rimal, Abhishek Tiwary, Abbes Amira, Using artificial intelligence and data fusion for environmental monitoring: A review and future perspectives, Information Fusion, Volumes 86—87, 2022, Pages 44-75, ISSN 1566-2535

Luis A. Monteagudo, Environmental Assessment