Reduction of Predictive Uncertainty for Soil Water Aquifers of Tropical Amazon Rainforest using Sensor Placement via Goal-oriented Bayesian Experimental Design

The Amazon Basin cycles more water through streamflow and evaporation than any other contiguous forest in the world. At any instant, there is an enormous amount of water stored in the Basin in the unsaturated soil and groundwater but there is scant knowledge on their spatial distribution and seasonality – a critical knowledge gap, especially as climate changes. This project will leverage long-term historical data and ongoing fieldwork in the only experimental watershed in wet tropics in central Brazil to develop a goal-oriented Bayesian experimental design to study the variability of subsurface water and its relationship with various landscape controllers. This research will interrogate observations, theory, and numerical modeling and manipulate sensor placement to ensure optimality of spatial and temporal monitoring to reduce uncertainty in soil moisture/groundwater estimation with predictive models.

People

Project ID: 1103