Title: Soil Moisture Sensing Controller And optimal Estimator (SoilSCAPE): First Deployment of the Wireless Sensor Network and Latest Progress on Soil Moisture Satellite Retrieval Validation Strategies
Author: Mahta Moghaddam
Organization: University of Michigan
Co-Authors: D. Entekhabi, M. Liu, D. Teneketzis, A. Silva, M. Burgin, Q. Wang, X. Wu, A. Nayyar, Y. Goykhman, A. Kakhbod

Abstract:
In this talk we report on the latest results obtained under the SoilSCAPE project, namely, the architecture, installation, and status of the first in-situ wireless soil moisture network, and the computational strategies for validation of coarse-resolution satellite-derived soil moisture from the fine-grained in-situ data.

The SoilSCAPE project is aimed at developing technologies for dynamic and near-real-time validation of space-borne soil moisture measurements, in particular those from the NASA Soil Moisture Active and Passive (SMAP) mission. Soil moisture fields are functions of variables that change over time scales of minutes to days or weeks, and across the range of spatial scales from a few meters to several kilometers. To address this complexity, we are developing a number of sensor placement solutions based on nonstationary spatial statistics of soil moisture, and for each location, a dynamic scheduling policy based on physical models of soil moisture temporal dynamics. Furthermore, we relate the ground-based estimates of the true mean to the space-based estimates through a physics-based statistical aggregation procedure enabled by remote sensing and hydrologic landscape simulators. An integrated communication and actuation platform is developed and used to command the in-situ sensors and transmit their data to a base station in real time.

Here, we describe a nearly full-scale system prototype of this system recently installed in Oklahoma. This first SoilSCAPE multi-hop network includes 21 wireless end-nodes each with 3 in-situ soil moisture sensors, 2-3 wireless routers, and a single wireless base station. The data from the network are uploaded to a server via a 3G link. The remote sensor (radar) and hydrologic landscape simulator results are also presented, with possible aggregation strategies emerging based on the statistics of the sites considered. Full-scale field experiments are planned (and some underway) in coordination with SMAP calibration/validation experiments to prototype the validation system.