The rise of AI: How artificial intelligence will play a vital role in environmental monitoring

Increasing threats of climate change, pollution, and the loss of biodiversity highlight the importance of monitoring the health of our environment. While traditional methods may be cost-effective, they often lack accuracy and rely on difficult-to-measure parameters. The rise of artificial intelligence (AI) presents a unique opportunity to reshape environmental monitoring and make it a primary future source.

Key applications of AI

Air and water monitoring

AI-enabled sensors and Internet of Things (IoT) networks enable real-time monitoring of air and water quality. Techniques, such as Fuzzy Logic (FL), Support Vector Machines (SVMs), and Artificial Neural Networks (ANNs), are revolutionising air pollution forecasting with high precision, generalisation, and fault tolerance. These techniques handle complex, non-linear, and uncertain data, enabling accurate predictions of pollutants such as O3, SO2, and particulate matter (PM). FL manages uncertainty, SVMs recognise non-linear relationships, and ANNs are versatile for forecasting gaseous and particulate pollutants. These AI methods help to predict pollutant concentrations, analyse meteorological influences, and assess air quality indices, offering effective tools for long-term air pollution monitoring1.

AI has proven to be a valuable tool showing that fluctuations in water quality can be attributed to a range of factors including different industries, pollutant sources, and production cycles. Water quality correlation maps and frequent itemset analyses can identify abnormal fluctuations and highlight the most critical water quality issues. Additionally, long short-term memory networks can accurately predict and monitor future changes in water quality resulting from specific pollutant sources2.

Climate prediction and extreme weather monitoring

Climate modelling and forecasting is one of AI’s most obvious applications, with the increase in extreme weather events due to climate change. With AI, climate modelling tools can evaluate climate adaptations and provide future weather projections and simulations. Corrective machine learning (ML) methods have improved weather forecasts by predicting up to 10 days in advance and reducing time-mean precipitation biases and global time-mean errors by 30% compared to traditional climate models3. Moreover, it has been found that ML models can accurately predict solar radiation levels at various sites around the globe3.

Wildlife and habitat protection

Utilising AI technology is essential for surveillance and conservation of biodiversity. ML algorithms play a crucial role in identifying endangered species, monitoring their behaviour, detecting habitat destruction, and analysing the proliferation of invasive species3. These advanced tools aid in the development of species distribution models, evaluation of climate change effects on ecosystems, and optimisation of resource allocation. Through enhanced climate modelling, drones and image recognition, AI has the potential to bolster conservation initiatives, safeguard habitats, and mitigate environmental harm.

Benefits and challenges of AI in environmental monitoring

AI enables more precise measurements and reliable tracking of pollution levels, climate changes, and ecological health indicators compared to traditional laboratory-based methods. Its abilities to process vast amounts of data in real time allow for immediate responses to potential contamination and environmental threats4. For example, AI sensors are tracking dangerous substances in plants and soil. This enables environmental protection, safety of food crops, and avoids human exposure to hazardous materials4.

AI monitoring reduces the need for manual data collection and analysis, thereby decreasing workload, increasing efficiency, and offering scalability for global deployment. This addresses large-scale environmental challenges with minimal human intervention.

At the continental level, the absence of open-access databases is a main obstacle to environmental monitoring. In this sense, creating new datasets turns into a contemporary issue that encourages more accessibility to data5. Another specific challenge is that even with slight input modifications, adversarial attacks can cause severe errors in Deep Neural Networks5. The absence of standardised policies and regulations regarding AI use in environmental applications poses challenges in ensuring responsible and ethical use. There is a need to establish guidelines for ethical AI technology use for societal and environmental objectives6. Lastly, the energy consumption of AI systems, particularly in data-intensive applications, raises questions about the environmental footprint of the technology itself. Ensuring AI is used sustainably is essential to maximising its positive impact.

The future of AI in environmental monitoring

The role of AI in environmental monitoring is set to expand with new technological advancements and increased collaboration. Future AI-driven monitoring of dangerous substances has the potential to revolutionise safety measures in various industries through increased automation, advanced models, and integration with emerging technologies like IoT5. In fact, AI's use in smart factories and other contemporary industrial areas has been recognised as ‘revolutionary’7, and its efficiency and innovation in manufacturing have significantly influenced the industry 4.0 paradigm, resulting in development of advanced technologies5. Collaboration between governments, tech companies, and NGOs is crucial for responsible AI development and global accessibility, requiring robust frameworks and policies for environmental sustainability.

Conclusion

AI is becoming a reliability for protecting our planet. Predicting weather, developing sustainable products, automating processes, and implementing early-warning systems for environmental hazards all depend on advanced technology. With evolving research underway, the maximum potential of AI is yet to be realised, further enhancing its ability to tackle environmental challenges. Despite initial high costs for the investment and responsibility of deployment, AI can build a sustainable future where technology plays a vital role in preserving the environment for generations to come.

References

1. Masood, A., Ahmad, K., Journal of Cleaner Production, 2021, 322, 129072. https://doi.org/10.1016/j.jclepro.2021.129072.

2. Popescu,S., Mansoor, S., Ali Wani, O., Kumar, S., Sharma, V., Sharma, A., Arya, V., Kirkham, M., Hou, D., Bolan, N., Chung, Y.S., Frontiers in Environmental Science, 2024, 12, 1336088. https://doi.org/10.3389/fenvs.2024.1336088.

3. Konya, A., Nematzadeh, P., Science of the Total Environment, 2024, 906, 167705. https://doi.org/10.1016/.scitotenv.2023.167705.

4. Subramaniam, S., Raju, N., Ganesan, A., Rajavel, N., Chenniappan, M., Prakash, C., Pramanik, A., Basak, A.K., Dixit, S., Sustainability, 2022, 14 (16), 9951. https://doi.org/10.3390/su14169951

5. Himeur, Y., Rimal, B., Tiwary, A. and Amira, A., Information Fusion, 86-87, pp. 44-75. https://doi.org/10.1016/j.inffus.2022.06.003.

6. Wani, A., Rahayu, F., Ben Amor, I., Quadir, M., Murianingrum, M., Parnidi, P., Ayub, A., Supriyadi, S., Skiroh, S., Saefudin, S., Kumar, A., Latifah, E., Environmental Science and Pollution Research, 2024, 31 (12). https://doi.org/10.1007/s11356-024-32404-z.

7. Mao, S., Tang, Y., Qian, F., Engineering, 2019, 5 (6), pp. 995-1002. https://doi.org/10.1016/j.eng.2019.08.013.