Ground Network Design and Dynamic Operation for Near-Real-Time Validation of Space-Borne Soil Moisture Measurements

This project addresses the topic of “Smart Sensing.” It is motivated by a sensor-web measurement scenario including spaceborne and in-situ assets. The technology objective is to enable a guided/adaptive sampling strategy for the in-situ sensor network to meet the measurement validation objectives of the spaceborne sensors with respect to resolution and accuracy. The sensor nodes are guided to perform as a macro-instrument measuring processes at the scale of the satellite footprint, hence meeting the requirements for the difficult problem of validation of satellite measurements. The science measurement considered is the surface-to-depth profiles of soil moisture estimated from satellite radars and radiometers, with calibration/validation using in-situ sensors. Satellites allow global mapping but with coarse footprints. The total variability in soil-moisture fields comes from variability in processes on various scales. Installing an in-situ network to sample the field for all ranges of variability is impractical. The hypothesis is that a sparser but smarter network can provide the validation estimates by operating in a guided fashion with guidance from its own sparse measurements. The feedback and control take place in the context of a data assimilation system.

The task is to develop technologies for dynamic and near-real-time validation of space-borne soil moisture measurements, in particular those from the Soil Moisture Active and Passive (SMAP) mission, one of the four first-tier Earth Science missions identified by the National Research Council Decadal Survey. Soil moisture fields are functions of variables that change over time across the range of scales from a few meters to several kilometers. Therefore, an optimal spatial and temporal sampling strategy needs to be developed that will not only make the validation task feasible, but will also result in substantial improvement in science quality for soil moisture validation over conventional techniques. The task is to develop the optimal sensor placement policy based on nonstationary spatial statistics of soil moisture, and for each location, develop dynamic scheduling policies based on physical models of soil moisture temporal dynamics and microwave sensor models for heterogeneous landscapes. Tractable computational strategies are proposed. The task is then to relate the ground-based estimates of the true mean to the space-based estimates through a physics-based statistical aggregation procedure. An integrated communication and actuation platform will be used to command the sensors and transmit their data to a base station in real time. Full-scale field experiments are proposed in coordination with SMAP cal/val experiments to prototype the validation system.