Italian National Agency for New Technologies, Energy and Sustainable Economic Development
Agriculture: ENEA applies AI in soil health monitoring
Studying the impact of climate change on soil health using artificial intelligence (AI) is the purpose of a study conducted by ENEA and the University of Bari whose first outcomes - which open up new perspectives for sustainable soil management - have been published in the journal “Machine Learning and Knowledge Extraction”.
“The machine learning techniques employed enabled to identify one of the main soil health indicators[1], the microbiome, i.e. the vast variety of bacteria, fungi and protists residing in the soil, which play a pivoltal role in soil carbon dynamics in response to climate change,” explained Claudia Zoani, researcher at the ENEA Sustainable Agrifood Systems Division and coauthor of the study with the University of Bari research team[2] coordinated by Professor Sabina Tangaro. “This discovery,” the ENEA researcher continued, ”could play an important role in climate change mitigation and sustainable soil management.
Climate change alters temperature regimes and precipitation patterns, directly affecting soil temperature and soil water availability. These changes modify the composition of soil microbial communities and consequently organic matter decomposition processes. “The rise in temperatures accelerates the decomposition of the microbiome, increasing the amount of greenhouse gases emitted into the atmosphere, like carbon dioxide and methane, which lead to a degradation of soil quality, putting agricultural production and food security at risk for millions of people around the world,” Zoani pointed out.
An increase in the activity of microorganisms in the soil when the temperature rises by 10°C is expressed as the Q10 value and specifically describes the sensitivity of microbial respiration to temperature variations. Warming of soil can enhance microorganisms activity, resulting in increased carbon dioxide (CO2) release in the atmosphere. Knowledge of this value of the microbiome activity becomes important to predict how the carbon cycle[3] in the soil will respond to global warming. And, in this context, "artificial intelligence can play a crucial role because it offers very effective tools to analyse complex data, make predictions and develop innovative solutions to mitigate the effects of climate change and adopt sustainable agricultural practices to ensure long-term food security", said Professor Tangaro.
Notes
[1] The data refer to 332 soil samples collected in 29 countries worldwide, representing a wide range of climate conditions and soil properties worldwide. The dataset includes a variety of environmental factors, like mean annual temperature (MAT), ranging from -7 °C to 30 °C, and different types of vegetation and soil properties. For each soil sample, 27 factors of different nature were identified, covering environmental aspects, soil microbiota, biochemical recalcitrance, amount of substrate and mineral protection factors.
[2] Together with Pierfrancesco Novielli, Michele Magarelli, Donato Romano, Lorenzo de Trizio, Pierpaolo Di Bitonto, Alfonso Monaco, Nicola Amoroso and Anna Maria Stellacci guided by Prof. Sabina Tangaro of the University of Bari.
[3] The soil carbon cycle is the process that moves carbon between soil, plants, animals and the atmosphere. Plants absorb carbon from the atmosphere in the form of carbon dioxide (CO2) through photosynthesis and use it to build plant structures. When plants and animals die or leave behind debris (such as leaves or feces), this organic matter falls on the ground and the carbon content enters the soil. Subsequently, soil microorganisms (bacteria, fungi, etc.) break down the organic matter. During this stage, some of the carbon is converted into useful nutrients for other plants, while some is released back into the atmosphere in the form of CO2 through microbial respiration. The soil carbon cycle is a ‘natural sink’ for carbon, which helps regulate the climate and support plant and animal life.