Recent applications of AI to environmental disciplines: A review.

Artificial Intelligence Carbon footprint Environmental data Interdisciplinary collaborations Real-world data

Journal

The Science of the total environment
ISSN: 1879-1026
Titre abrégé: Sci Total Environ
Pays: Netherlands
ID NLM: 0330500

Informations de publication

Date de publication:
01 Jan 2024
Historique:
received: 13 08 2023
revised: 06 10 2023
accepted: 07 10 2023
medline: 12 10 2023
pubmed: 12 10 2023
entrez: 11 10 2023
Statut: ppublish

Résumé

The rapid development and efficiency of Artificial Intelligence (AI) tools have made them increasingly popular in various fields and research domains. The environmental discipline is now experiencing an exponential interest in harnessing the potential of AI over the past decade. We have reviewed the latest applications of AI tools in the environmental disciplines, highlighting the opportunities they present and discussing their advantages and disadvantages in this field. After the emergence of deep learning algorithms in 2010, interest in using AI tools for environmental tasks has grown exponentially. Among the studied articles, over 65 % of environmental tasks that demonstrate interest in using AI tools initially relied on conventional statistical and mathematical models. Using AI tools can greatly benefit the areas of environmental science and engineering. One of the main advantages of utilizing AI tools is their ability to analyze and process large amounts of data efficiently. Recently, the European Union established a European supercomputing ecosystem program to advance science and enhance the quality of life for its citizens. Nine of these projects prioritize environmental and sustainable goals. Despite the benefits of AI, it is still in its early stages of development, which comes with environmental concerns. The amount of power consumed and the time required to train an AI model can greatly affect the carbon emissions it produces, exacerbating the challenges posed by climate change. Efforts are currently underway to develop AI technology that is environmentally sustainable, minimizes energy consumption, and has a low carbon footprint. Selecting the appropriate AI model architecture can reduce energy consumption by almost 90 %. The main finding suggests that collaboration between environmental and AI professionals becomes crucial in leveraging the full potential of AI in addressing pressing environmental challenges.

Identifiants

pubmed: 37820816
pii: S0048-9697(23)06332-5
doi: 10.1016/j.scitotenv.2023.167705
pii:
doi:

Types de publication

Journal Article Review

Langues

eng

Sous-ensembles de citation

IM

Pagination

167705

Informations de copyright

Copyright © 2023 Elsevier B.V. All rights reserved.

Déclaration de conflit d'intérêts

Declaration of competing interest The authors have no financial or non-financial competing interests related to this work.

Auteurs

Aniko Konya (A)

University of Illinois, Chicago, IL 60637, USA. Electronic address: akonya@uic.edu.

Peyman Nematzadeh (P)

Independent Researcher, Greater Chicago Area, IL 60527, USA.

Classifications MeSH