|
Getting your Trinity Audio player ready...
|
In a significant effort to monitor and protect biodiversity, researchers at Nanyang Technological University (NTU) are leveraging advanced technologies, including artificial intelligence (AI) and remote sensors, to study moth populations and assess ecosystem health. This innovative approach is part of the AMBER (AI-assisted Monitoring of Biodiversity using Edge-processing and Remote sensors) project, which aims to revolutionise biodiversity monitoring with cutting-edge technology.
The project involves the deployment of ten remote monitoring stations across Singapore. Each station is equipped with ultraviolet (UV) lights to attract moths and cameras to automatically photograph these insects. AI algorithms then analyse the images in real-time, identifying moth species by referencing a comprehensive local database. This database is continuously enriched with images from the citizen science platform iNaturalist, ensuring a robust and dynamic identification system.
“Moths play a crucial role in pollinating plants and flowers at night, complementing the daytime activities of butterflies and bees,” explained Associate Professor Eleanor Slade of NTU’s Asian School of the Environment. “Their presence is also an important indicator of ecosystem health, as they are a key food source for bats and birds.”
Singapore is estimated to host between 400 and 1,000 moth species, but many remain understudied due to their nocturnal nature. Nicole Su-Yin Dorville, a PhD student at NTU, is developing novel methods to monitor how biodiversity responds to environmental changes. Her research aims to fill the knowledge gap regarding nocturnal pollinators and their ecological significance.
Evidence of global wildlife declines highlights the urgent need for more insect data, given their vital roles in ecosystems, such as food for birds and mammals, nutrient recycling, and crop pollination. These insects also serve as key indicators of climate change impacts. The AMBER project addresses this need by developing and testing automated sensors, deep learning, bioacoustics, and computer vision to standardise the monitoring of insects, bats, and birds worldwide.
Research has shown alarming trends, such as a 75% decline in insect biomass in Germany, signalling an “Insect Armageddon.” Understanding species trends and the factors driving these changes is crucial for addressing these challenges. The Automated Monitoring of Insects (AMI) trap offers a practical solution, providing non-biased, non-invasive monitoring on wide spatial and temporal scales. This system uses machine learning models for insect detection and classification, ensuring efficient data analysis.
AMBER’s global network includes 40 monitoring systems deployed in biodiversity hotspots such as Singapore, Kenya, Costa Rica, Japan, and Thailand. Funded by the Abrdn Charitable Foundation, the project seeks to trial and refine these advanced monitoring technologies, ultimately providing a more accurate and comprehensive picture of global biodiversity.
“The integration of AI and remote sensing technology in biodiversity monitoring represents a significant leap forward,” noted Associate Professor Slade. “It allows us to gather and analyse data on a scale and with a precision that was previously unattainable.”
This technological advancement not only enhances our understanding of moth populations but also offers a powerful tool for conservation efforts worldwide. By identifying and tracking changes in biodiversity, researchers can better inform and implement strategies to protect and preserve vital ecosystems.
The fusion of AI and remote sensors in the AMBER project exemplifies how modern technology can be harnessed to address critical environmental challenges. By focusing on moths as indicators of ecosystem health, NTU researchers are paving the way for more effective and informed conservation practices, ensuring the sustainability of our planet’s rich biodiversity.
Nanyang Technological University (NTU) is exploring the use of artificial intelligence across various fields, including environmental science, healthcare, and smart cities. Recognising the need for energy-efficient AI, researchers from the National University of Singapore (NUS) and industry leaders have developed a groundbreaking suite of silicon systems to enhance the efficiency of AI-connected devices.
