Multi-sensor UAV-platform including SAR for high-fidelity measurement of vertically resolved soil moisture distribution and its coupling with distributed sensor network data

Period of Performance: 02/21/2017 - 11/20/2017

$200K

Phase 1 SBIR

Recipient Firm

Mirage Systems, Inc.
2188 Bering Drive Array
San Jose, CA 95131
Firm POC, Principal Investigator

Abstract

Robust reactive transport models of subsurface hydrobiogeochemical processes are crucial to understanding complex subsurface systems. However, predictive capabilities of models that simulate coupled interactions within these systems are limited by accuracy and fidelity of properties populating or constraining the models. To improve these models’ predictive capabilities, advanced sensing systems are needed to accurately capture subsurface properties at adequate spatiotemporal resolution over relevant spatial scales. Real‐time in‐situ monitoring is increasingly done using distributed sensor networks relying on point‐ scale measurements at various locations distributed over a watershed. The approach is promising while integration of this data with spatially continuous measurements is critical to upscaling results. Unmanned Aerial Vehicle (UAV)‐based sensors offer potential to complement distributed point sensor networks and bridge the spatial gap, while providing increased temporal resolution once automated. Soil moisture is an important hydrologic state variable critical to successful hydrological and biogeochemical model predictions. Importantly, it varies spatiotemporally due to variations in precipitation, soil properties, topographic features, and vegetation characteristics. It is used for indirect estimation of soil hydraulic parameters (through inverse methods), calibration of models for improved estimation of hydrological processes, and improved estimation of evapotranspiration and hydraulic redistribution around roots. An understanding of root zone hydrology is crucial in accurately quantifying water use of plants, root‐soil biogeochemical interactions, nutrient fluxes, and potential for groundwater recharge. The project has two goals: (1) Develop a UAV‐based multi‐sensor platform, including a novel Synthetic Aperture Radar (SAR) sensor. This sensor will collect ultrawideband frequency data (P‐Band through L‐Band) needed to develop a high‐fidelity, vertically resolved soil moisture distribution from surface into root zone; (2) Develop a data assimilation framework integrating vertically resolved soil moisture data with data from distributed sensor networks to improve calibration of complex hydrobiogeochemical models. For this project: (1) Mirage Systems will develop a novel SAR sensor, mountable on a UAV platform, with capability to collect data useful to extract high‐fidelity, vertically resolved soil moisture distribution into the root zone, (2) USC will enhance and further validate algorithms for estimation of vertically resolved profiles of soil moisture into the root zone from the SAR data, (3) LBNL will develop a data assimilation framework such that the above data can be merged with additional UAV‐based data (surface temperature & reflectance, microtopogrpahy, snow distribution) and distributed sensor network data into a physically‐based hydrological model (coupling surface & subsurface processes) to improve estimation of soil hydraulic parameters and evapotranspiration. In phase 1, we’ll perform local experiments testing preliminary developments. In phase 2, we’ll further develop and refine our approach at the East River site near Crested Butte, CO, part of LBNL Watershed Functions Scientific Focus Areas (SFA). The developed capability will be highly valuable to improve ecosystem understanding and will have strong potential for commercial applications in ecosystem management, precision agriculture and water resource management.