Abstract

The increasing demand for timely, comparable, and policy-relevant climate change indicators poses significant challenges for official statistics, particularly in regions with heterogeneous institutional capacities, data availability, and methodological practices. This paper presents an applied study developed within a Regional Public Goods initiative financed by the Inter-American Development Bank applied to eleven countries in Latin America and the Caribbean: Argentina, Brazil, Chile, Colombia, Costa Rica, Ecuador, Mexico, Paraguay, Peru, the Dominican Republic, and Uruguay.

The initiative is technically coordinated by the UN Big Data Hub Brazil, within a line of work focused on strengthening cooperation among producers of official statistics. A central contribution of the project is the promotion of a common and harmonized approach to climate change indicator production, aimed at improving cross-country comparability, reducing methodological fragmentation, and supporting consistent regional analyses.

Results from the first diagnostic phase reveal a highly heterogeneous regional scenario. While some countries have reached advanced stages in compiling indicators from the Global Set of Climate Change Statistics and Indicators, others are still in the process of consolidating basic methodological and institutional practices. This diversity of development stages and institutional maturity across Latin America and the Caribbean underscores the need for a shared methodological framework capable of accommodating national specificities while ensuring regional consistency.

Current phase focus on identify, develop, implement and test methodologies and Artificial Intelligence–based solutions using alternative data sources, particularly Earth Observation data, to automate key processes required for indicator production, including data exploration, analysis, modeling, and visualization. The team is evaluationg AI-driven methods for selected Global Set indicators, including Mitigation Indicator 125 (increase in forest area relative to total forest area) and Adaptation Indicators 143 (nature-based solutions related to ecosystem services for storm mitigation, coastal protection, and flood mitigation) and 145 (proportion of urban green areas in total city areas).

A key output is an open-source computational tool designed to ensure transparency, reproducibility, and adaptability for national statistical offices, while remaining extensible for future indicators. The design, development, and deployment of a scalable, AI-enabled digital solution aim at automating the production, processing, and regular updating of selected climate change mitigation and adaptation indicators. This ongoing study can demonstrate how a shared AI-based framework can strengthen official climate statistics through regional cooperation.