Using Machine Learning to Explore Plant-Climate Connections
Introduction:
Scientists from UNSW and Botanic Gardens of Sydney have trained AI to unlock data from millions of plant specimens kept in herbaria around the world to study and combat the impacts of climate change on flora. Using a new machine learning algorithm, the team discovered that leaf size doesn’t increase in warmer climates within a single species, contrary to frequently observed patterns. Published in the American Journal of Botany, this research not only reveals the factors affecting leaf size within a plant species but also demonstrates how AI can be used to document climate change effects effectively. The digitization of herbarium collections has made it possible to quickly process and analyze large datasets, providing valuable insights into plant evolution and adaptations.
Full Article: Using Machine Learning to Explore Plant-Climate Connections
Scientists from UNSW and Botanic Gardens of Sydney have developed an AI system that can unlock data from millions of plant specimens stored in herbaria worldwide. This advancement will help researchers study and combat the impacts of climate change on flora. Herbarium collections, which are time capsules of plant specimens, are constantly growing and it is no longer possible to manually go through them all. The new machine learning algorithm developed by the scientists can process thousands of leaf samples to understand the effects of climate on leaf size within a single species.
Herbarium collections move to the digital world
Herbaria have existed since the 16th century and contain valuable scientific information about plants. To facilitate scientific collaboration and make the information more accessible to researchers, efforts have been made to transfer these collections online. The Botanic Gardens of Sydney undertook the largest herbarium imaging project, transforming over 1 million plant specimens into high-resolution digital images. This digitalization project was followed by the development of an algorithm that could automate the detection and measurement of leaf sizes in scanned herbarium samples.
Computer vision measures leaf sizes
The algorithm created by the research team uses computer vision, specifically a convolutional neural network, to identify and measure the size of leaves in herbarium samples. By teaching the computer to locate and measure leaves, the team was able to process specimens and log their individual characteristics at a much faster speed than before. This advancement allows for the analysis of the relationship between leaf size and climate within and among plant species.
A break in frequently observed patterns
Traditionally, it has been observed that plants in wetter climates tend to have larger leaves compared to those in drier climates. However, the research team discovered that this correlation is not consistently seen within a single species across the globe. The machine learning algorithm revealed that different factors, such as gene flow, are operating within species and may contribute to variations in leaf size-climate relationships. This finding challenges the commonly observed patterns in leaf size and highlights the importance of studying within-species dynamics.
Using AI to predict future climate change responses
The machine learning approach used in this study provides a level of accuracy suitable for examining the links between leaf traits and climate. By effectively documenting climate change effects, these methods can help researchers predict how plants will respond to future climate change. Additionally, AI algorithms have the potential to identify trends and insights that may not be immediately apparent to human researchers, leading to new discoveries in plant evolution and adaptation.
In conclusion, the collaboration between UNSW and Botanic Gardens of Sydney has resulted in the development of an AI system that can unlock and analyze data from herbarium collections. By studying the relationship between leaf size and climate within and among plant species, researchers can gain a better understanding of the impacts of climate change on flora. The use of AI algorithms not only speeds up the research process but also provides new insights into plant evolution and predictions about future climate change responses.
Summary: Using Machine Learning to Explore Plant-Climate Connections
Scientists from UNSW and the Botanic Gardens of Sydney have used AI to analyze millions of plant specimens stored in herbaria worldwide in order to study the effects of climate change on flora. The researchers developed a machine learning algorithm to process leaf samples and found that leaf size does not increase in warmer climates within a single species, contrary to previous expectations. The study, published in the American Journal of Botany, demonstrates how AI can transform static collections and quickly document the impacts of climate change. By digitizing herbarium collections, scientists can facilitate collaboration and access valuable plant specimen data.
