Heatmap of class probabilities from the ensemble model for non-croplands classes for 2050 under the ‘business-as-usual’ trajectory scenario with moderate emissions. Credit: Climate Change Impact on Agricultural Land Suitability: An Interpretable Machine Learning-Based Eurasia Case Study.
By 2050, it is projected by scientists that the global demand for food will surge by 110%. Presently, approximately 40% of croplands and pastures face threats due to rising average temperatures, elevated greenhouse gas levels in the atmosphere, and other contributing factors impacting the planet.
A collaborative research effort involving Skoltech, the Institute of Geography of the Russian Academy of Sciences, and other prominent research institutions utilized extensive open data and artificial intelligence to assess the potential changes in agricultural land suitability over the next 25 years. The analysis indicated a projected increase in croplands within northern territories. The detailed findings of this study have been published in IEEE Access.
The research methodology encompassed three key phases: data collection and preprocessing, machine learning model training, and result evaluation through the prediction of cropland distribution based on diverse climate models and shared socioeconomic pathways scenarios. The study specifically targeted the Eastern Europe and Northern Asia regions.
Heatmap of class probabilities from the ensemble model for rainfed arable lands classes for 2050 under the ‘business-as-usual’ trajectory scenario with moderate emissions. Credit: Climate Change Impact on Agricultural Land Suitability: An Interpretable Machine Learning-Based Eurasia Case Study.
The research team relied on openly accessible data sources for their analysis. ERA5, a dataset from the European Weather Forecasting Center, integrates real measurements from weather stations and models to create a standardized grid of 30 x 30 square kilometers worldwide, spanning from 1950 to the present. Additionally, CMIP models, developed by various global institutes including those in Russia, were employed to forecast climate patterns up to 2100.
Valery Shevchenko, the primary author and a research engineer at Skoltech’s Applied AI Center, highlighted the variability in accuracy among CMIP models for different climatic parameters such as air temperature and wind speed. The researchers examined three distinct datasets corresponding to different climate change scenarios to understand the implications on irrigation conditions for arable lands, incorporating global food security-support analysis data at a 1 km x 1 km resolution.
Heatmap of class probabilities from the ensemble model for major irrigated arable lands classes for 2050 under the ‘business-as-usual’ trajectory scenario with moderate emissions. Credit: Climate Change Impact on Agricultural Land Suitability: An Interpretable Machine Learning-Based Eurasia Case Study.
Heatmap of class probabilities from the ensemble model for minor irrigated arable lands classes for 2050 under the ‘business-as-usual’ trajectory scenario with moderate emissions. Credit: Climate Change Impact on Agricultural Land Suitability: An Interpretable Machine Learning-Based Eurasia Case Study.
The study’s projections suggest a potential increase in arable land by 2050, albeit with a shift towards northern regions. This shift may necessitate enhanced irrigation in currently exploited agricultural areas, posing new challenges.
In alignment with recommendations from the Intergovernmental Panel on Climate Change, the authors stress the significance of detailed regional assessments to adapt to climate variability and ensure food security.
More information:
Valeriy Shevchenko et al, Climate Change Impact on Agricultural Land Suitability: An Interpretable Machine Learning-Based Eurasia Case Study, IEEE Access (2024). DOI: 10.1109/ACCESS.2024.3358865
Provided by
Skolkovo Institute of Science and Technology
Citation:
Researchers use AI to predict how agricultural land suitability may change by 2050 (2024, February 20)
retrieved 20 February 2024
from https://phys.org/news/2024-02-ai-agricultural-suitability.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
