For about three years now, smartphone users have been able to use the “Flora Incognita” app to identify plants and find out their names in no time at all. And over the course of that time, they’ve collected vast amounts of data that can now have another important use. Researchers can use it to find out how environmental conditions are changing against a backdrop of climate change, habitat loss and changes in land use.
Researchers from the German Center for Biodiversity Research, the Remote Sensing Center for Earth System Research at the University of Leipzig, the Helmholtz Center for Environmental Research, the Max Planck Institute for Biogeochemistry and the Ilmenau University of Technology have now investigated the reliability of the information collected in this way is and whether it is as accurate as long-term data sets.
To do this, the scientists examined data that users recorded with the help of Flora Incognita between 2018 and 2019 in Germany and compared it with the FlorKart database of the Federal Agency for Nature Conservation. The Federal Agency’s data had been collected using conventional long-term mapping over a period of more than 70 years with the help of more than 5,000 plant experts.
The influence of environmental factors on plant distribution
The research showed that ecological patterns comparable to those from the long-term mapping of Germany’s flora could indeed be derived from the data collected in just two years using the Flora Incognita app. Plus, the researchers were able to see from the data the extent to which the distribution of different plant species was influenced by different environmental factors.
In areas with low population density, however, a direct comparison of the two data sets showed greater variation. “How much data is collected in a particular region with an app is of course highly dependent on how many smartphone users there are,” says Jana Wäldchen, a scientist at the Max Planck Institute for Biogeochemistry and co-developer of the app. For this reason, the deviations are greater in rural regions. Popular tourist destinations, such as the area around the Zugspitze mountain or the North Sea island of Amrum, were an exception.
In addition, the individual interests of the users also had an influence on the plant species recorded, she said. “The plant observations collected with the app reflect what people see in nature and what they are interested in,” says Wäldchen. Species that occur frequently and are conspicuous by their appearance would thus also be identified more often than those that occur rarely and are less noticeable. But the sheer volume of data makes it possible to reconstruct known biogeographical patterns, the researchers explain. For their study, they had more than 900,000 observed data points available from the first two years in which the app was used.
Real-time global coverage
“We are convinced that automatic species recognition has much greater potential in the future than previously thought and could enable rapid detection of changes in biodiversity,” says Miguel Mahecha, a professor at the University of Leipzig. For example, this type of data collection for biodiversity and environmental research could soon be integrated into strategies for long-term mapping. In the process, the more people who use apps like Flora Incognita, the better changes in ecosystems could be analyzed and even recorded in real time.
Flora Incognita was developed at the Ilmenau University of Technology and is the first applied plant identification app in Germany whose deep artificial neural networks have been trained with thousands of plant images. Already, the app can recognize more than 4,800 plant species.
“In developing Flora Incognita, we found that there is a great need and interest for better biodiversity data collection technologies. For us as computer scientists, it is gratifying to see that the technologies we have developed are making an important contribution to biodiversity research,” says Patrick Mäder, professor at the Ilmenau University of Technology.
Cover photo: The Flora Incognita app can identify plants. With the help of the location data of the recorded plant species, valuable data sets on the distribution of the various species are also created. © Jana Wäldchen / MPI-BGC
More articles on the topic of smartphone apps can be found here.
