"Modelo de predicción de la calidad del agua en ríos basado en índices e indicadores del recurso hidrico y el entorno socio ambiental."

Abstract

The development model adopted by humanity in the last decades has not been effective to bring welfare to people around the world, which is evidenced by our social and ecological imbalance. Human activities as well as environmental characteristics contaminate the superficial hydric resources in basins. Sustainable development includes economic, social and cultural development, in harmony with nature, which means a superior human development now and for future generations. It is, therefore, necessary to renew the biotic and abiotic resources as a primary condition for development. Among all those resources, the water represents the best indicator of social and economic development for a country. The objective of the investigation was to develop a statistical model for predicting the surface water quality based on social and environmental variables in watersheds. Personnel that work with environmental and territorial management will have a useful tool of easy interpretation as well as reliability. The areas selected for the model development were watersheds located in the Great Metropolitan Area and the Península of Osa. Besides, it was necessary to develop a range scale related to water class classification of our regulation. That made possible to establish mathematical calculation for quality water variables to become standardized according to environmental conditions and country regulations. Several WQIs including two new ones developed during the research, were analyzed. It was also necessary to propose new mathematical calculations for standardizing the values among different water quality indicators. For the statistical model development, several water quality indicators and indexes were analyzed. Besides, the relationship among social and environmental variables was investigated. Once developed the model, the validation process was carried out to determine the degree of reliability. It was found important sensitivities in relation to contamination degrees among the different WQIs used, that agrees with the statement that there is no universal WQI. One of the WQI proposed was selected which is very sensitive to contamination of the rivers and it represents very well those values according to data field. It was obtained a prediction model based in social and environmental variables in which the determination coefficient is 80.0% and the validation stage showed it as being a good predictor model with a standard deviation of 11% as range variability and statistical acceptability.