Identification of socio-economic variables to implement advanced artificial intelligence models to manage climate change risk

To develop climate change mitigation strategies, it is necessary to identify variables that facilitate the modeling of prospective scenarios. There are a large number of variables that must be analyzed in an integrated manner in order for scenarios to be proposed that include the particularities of a given area, measuring the possible effects of this phenomenon in terms of productivity. Identifying and analyzing variables and their variations over time enables fundamental predictions to understand the potential environmental impacts on ecosystems and human activity. Understanding these variables is important to support decision-making, policy development and implementing actions that help reduce greenhouse gas emissions and guarantee food security. This research study not only seeks to determine the technical variables, which are fundamental in predictive models, but also sets out to emphasize the importance of integrating social and economic aspects that can become decisive factors.

Rural areas in Colombia, with the department of Cundinamarca used as a case study, have been affected in various ways by climate change [1]. This scenario represents a challenge that needs to be addressed in a prioritized manner to ensure food security and independence, economic development, sustainability, livestock and human health, among other aspects that precisely relate to the development of a region. To propose solutions, artificial intelligence (AI) is emerging as an innovative alternative that makes it possible to process large amounts of data and find patterns, correlations and trends that can provide an understanding of the variables’ behavior, as well as develop systems to adapt to climate change. Therefore, identifying variables to apply advanced AI models to forecast the effects of climate change in a given region is a fundamental step towards generating an efficient and accurate tool to establish mitigation actions in a region that, together with the implementation of policies and actions that promote sustainability, will strengthen communities’ current capacity for action.

The variables identified include economic structure, access to technological resources, governance models, education levels, access to public services, poverty rate, demographics and crop price references. Through AI models and an in-depth analysis of available information, these types of models will become more precise for the implementation of early warning systems (EWS) and sustainable practices, as well as strengthen infrastructure. Historically in Colombia, rural areas are the most vulnerable to climate change given that they have fewer economic and technological resources that enable them to adapt to its impacts, with the most frequent phenomena being torrential rainfall, extreme flooding and forest fires; events associated with climate change.

• Peña Q, Andrés J, Arce B, Blanca A, Boshell V, J. Francisco, Paternina Q, María J, Ayarza M, Miguel A, & Rojas B, Edwin O. (2011). Trend analysis to determine hazards related to climate change in the Andean agricultural areas of Cundinamarca and Boyacá. Agronomía Colombiana, 29(2), 467-478. Retrieved January 09, 2024, from http://www.scielo.org.co/scielo.php?script=sci_arttext&pid=S0120-99652011000200014&lng=en&tlng=en.