The project tries to identify the origins of toxic substances in the country’s treated water by cross-referencing information on water contamination with data on polluting sources. The investigation aims to promote social control and pressure public authorities to take action against those responsible for contaminating drinking water. Detections above the Maximum Permitted Volume (VMP) set by the Ministry of Health will be used as the evaluation criterion. Data will be obtained from the Water Quality Surveillance Information System for Human Consumption (Sisagua) and complemented with data from scientific literature on contaminants in different regions of Brazil.

Doubly limited data are those that have a minimum value and a maximum value, and the most common examples are rates and proportions. When studying the Sustainable Development Goals (SDGs) and the United Nations 2030 Agenda, the indicators used are almost mainly composed of proportions of the population of interest. Furthermore, as these are objectives to be achieved by 2030, there is a need to monitor these indicators over time. Thus, this project combines doubly limited probability distributions with time series models for monitoring the SDGs. Another commitment is the sharing of computational codes. With this, the new models can overcome the barriers of mathematics and be applied by data scientists outside the academic environment.