Water quantity and quality in headwater catchments: Comprehensive data assessment, modeling, and simulation of scenarios

Abstract

Water availability is a major concern globally and relies highly on the quantity and quality of water. In headwater catchments of developing countries, managing said availability is additionally restricted by several factors. Among them, having scarce yet heterogeneous information regarding water quantity and quality is significant. This is because such combination impedes performing appropriate assessments, of both quantity and quality, and implementing predictive models to support decisions in these catchments. To address this issue, a framework comprising three stages and several activities is proposed. Such framework allows for: i) conducting a comprehensive assessment of water quantity and quality; ii) developing a predictive models supported on such assessment; and iii) simulating scenarios to resolve conflicts between uses and quality of water.

In the case of water quantity, the three stages focus on assessing hydrological data, preparing datasets for modeling, and developing models in daily and sub-daily resolutions. The last stage is grounded on a data-based mechanistic modeling approach and includes a novel combination of baseflow separation with digital filters and multi-objective optimization principles. The sequence of stages allows for the development of reliable yet mathematically simple models, requiring fewer inputs than data-intensive alternatives. Besides, having a hydrophysical meaning, the models are suitable for simulating alternative scenarios. In the case of quality, the three stages center in assessing water quality data, developing water quality models in headwater catchments, and simulating scenarios to resolve conflicts between uses and quality of water. These stages involve multivariate statistic techniques (i.e., Analyses of Principal Components and Clusters), and follow a modeling protocol mainly designed for mountain rivers in developing countries.

The framework is applied to the Lenguazaque River Basin, a 290 km2 headwater catchment in the Andean Fuquene Lake Watershed. Here, stakeholders are now optimizing water allocations for all users, given numerous issues such as severe pollution, and conflicts for the use of water for conservation, agriculture, and coal mining. The framework led to obtaining daily hydrological models with an acceptable performance, significantly better than a semi-distributed model. The performance of sub-daily models was however sub-optimal, yet it can be improved by enhancing the quality of sub-daily datasets and the efficiency of computational algorithms. In addition, the framework disclosed pathogens, nutrients, organic matter, and several metals, including the highly toxic Cr and Pb, among the most significant water quality constituents. It also showed that the driest season in the catchment is the one with highest pollution levels (i.e., January to March). Meanwhile, the water quality model reproduced the concentrations of pathogens, organic matter, and most nutrients, and showed a predictive capacity. This capacity was measured with an objective function to be minimized based on a normalized Root Mean Square Error. It increased only 14% when verified with a different dataset. Finally, the simulation of alternative scenarios showed that centralized treatment is not sufficient to make water safe for potabilization and agriculture in the catchment. For this reason, improving water quality in the sub-basins at the highest altitudes is required. Given these results in the case study, the proposed framework can be useful for similar purposes in other headwater catchments with similar restrictions of information, and where an improved management of water availability is needed.