Spatial prediction of stream physicochemical parameters for the Napo River Basin, Ecuador

Major threats to freshwater ecosystems in the Andean-Amazon region include agriculture, point and nonpoint source pollution, hydroelectric dams, oil extraction, mining and road building, yet little is known about the baseline values and current state of rivers and streams, or how expanding urban, industrial and agricultural activities could affect and change these freshwater ecosystems and their diversity. Therefore, there is a need to understand physicochemical parameters at the basin scale in many large river systems in the Andean-Amazon such as the Napo River, a world biodiversity hotspot. Establishing baselines for these parameters under less developed conditions will be impossible after development in the watersheds increases and hydroelectric projects reach completion. In addition to collecting data to fill in information gaps in geology, vegetation, disturbance, and other parameters that influence water quality, it is important to use existing data in the Napo to predict chemical parameters throughout the watershed to set a baseline from which deviations from development and climate change can be assessed and to guide data collection efforts. In this study, we provide the first basin-wide estimates of physicochemical parameters (pH, conductivity, dissolved oxygen and temperature) for the entire Napo River Basin using a recently developed geostatistical technique called top kriging. Basin-wide predictions aligned well with observed values and a model validation test showed a strong goodness-of-fit for parameter estimates. Furthermore, our predictions aligned well with observed values from independent datasets from the Napo Basin. These predictions could be useful in developing water quality reference baselines for the basin and for monitoring relative changes to parameters, assist in the identification of priority areas within the basin for management and conservation efforts, direct future research, and be applied in other data-scarce basins in remote regions worldwide.