What is growing there? With smartphone apps for automatic plant identification, nature lovers can effortlessly find their way around and not only that: They also make a contribution to ecological research, according to a study. The transmitted location data therefore provide valid information about large-scale patterns of biodiversity. In the long term, the use of identification apps can also reveal ecological changes, the scientists report.
For many people, the feeling of connectedness with nature also includes knowledge – they want to know what is green, blooming, creeping and floating around them. In the meantime, those inquisitive are now able to use smartphone apps that allow easy identification of animal and plant species. Since 2017, “Flora Incognita“. Over a million nature lovers already use this app, which was developed by the Technical University of Ilmenau, to automatically identify wild plants. The almost magical recognition ability of the program is based on artificial intelligence: The system was “trained” with thousands of plant images that were determined by experts. With every successful application, the app continues to learn and improve its recognition accuracy. Flora Incognita can now automatically identify over 4800 plant species using cell phone recordings.
To what extent does usable data arise?
In addition to the benefits for laypeople and the promotion of closeness to nature, the developers also hoped for benefits for science: Against the background of climate change, the loss of habitats and changed land use, applications such as Flora Incognita could provide important information. But is their value comparable to that of long-term data sets? The researchers led by Jana Wäldchen from the Max Planck Institute for Biogeochemistry in Jena have now investigated this question. To do this, they analyzed data generated between 2018 and 2019 with the help of Flora Incognita in Germany and compared it with the “FlorKart” database of the Federal Agency for Nature Conservation. This is a long-term mapping that was created with the support of over 5000 plant experts over a period of over 70 years.
The researchers were able to show that the data, which were obtained with the help of the app in just two years, can be used to derive ecological patterns in Germany that are comparable to the long-term mapping of flora using the conventional system. Among other things, the data reflected the influence of various environmental factors on the distribution of plant species. However, there were also peculiarities in the app’s data sets: “How much data is collected in a certain region with an app depends, of course, on how many smartphone users there are,” says Wäldchen. Rural regions were therefore less covered by the app – unless they were popular tourist destinations. So it would be nice if the number of users continued to rise.
Potential for biodiversity research
In evaluations, users’ preferences for rather conspicuous plants must obviously be taken into account: “The plant observations collected with the app reflect what users see in nature and what they are interested in,” says Wäldchen. But despite these peculiarities, the sheer amount of collected plant observations helps to reconstruct known biogeographical patterns, emerges from the evaluations of the approximately 900,000 observation data. It is becoming apparent that this type of data acquisition can be ideally integrated into long-term mapping in biodiversity and environmental research, the scientists sum up.
“We are convinced that automatic species recognition has much greater potential than previously assumed and could enable changes in biodiversity to be recorded quickly,” says first author Miguel Mahecha from the University of Leipzig. With an increasing number of users of apps such as Flora Incognita, changes in ecosystems around the world could be recorded and analyzed in real time. Co-author Patrick Mäder from the Technical University of Ilmenau is also happy about the success: “For us as computer scientists, it is something special that the technologies we have developed can make an important contribution to biodiversity research,” said the scientist.
Source: German Center for Integrative Biodiversity Research (iDiv)
Halle-Jena-Leipzig, specialist article: Ecography, doi: 10.1111 / ecog.0549