This project aims to further develop and refine a method to extract near-surface soil moisture estimates from Global Positioning System (GPS) receivers as a continental-scale soil moisture network and to evaluate errors to establish the underlying uncertainty in the estimates. Specifically, the investigators will (a) quantify how various aspects of the physical environment influence soil moisture estimates, including (i) vertical profiles of soil moisture and texture; (ii) vegetation amount and structure; and (iii) topography and surface roughness, and evaluate the physical environment around each of the Earthscope Plate Boundary Observatory (PBO) sites for suitability for the GPS soil moisture technique; (b) evaluate how antenna and receiver design/performance and satellite signals influence the soil moisture content time series derived from GPS signal-to-noise ratio (SNR) data, and evaluate the equipment used by PBO because its network is large, homogeneous, and well-maintained.
This project is a collaboration between hydrologists, climate scientists, electrical engineers, and GPS geodesists. A broad range of data will be collected from seven field sites, including GPS, in situ soil moisture, meteorological, and vegetation information. The different data types will be combined using an integrated modeling system designed to understand how the physical environment and antenna/receiver design influence GPS SNR data. This will yield a retrieval algorithm for soil moisture from GPS SNR data that is guided by physical principles.
This research will benefit society by providing important new soil moisture data for hydrological and climate studies and weather forecasting. Two graduate students will be supported and trained in a interdisciplinary educational program including hydrology, remote sensing, and GPS geodesy.
The research associated with this project was focused on the development of new methodology to measure soil moisture, vegetation water content (VWC), and snow depth: three essential earth system climate variables. These newly developed techniques utilize reflected signals from the Global Positioning System (GPS), and represent an emerging observational technology called GPS interferometric reflectometry (GPS-IR). GPS-IR observations sample areas approaching 1000 square meters. These intermediate resolution observations help bridge the gap between in-situ techniques that are essentially point measurements, and may not be representative of the surrounding environment, and the coarse resolution of satellite observations that are unable to resolve sub-pixel variability in their data products. Funds from this award were used to develop these new observational methods, validate them using existing methods, and create an online data portal to access these new data products for broad utilization. Collaborators from the University of Colorado are now producing daily estimates of related to the water cycle from more than 300 stations across the western United States. Soil moisture measurements are obtained from 98 sites, snow depth from 151 sites, and vegetative index estimates from 338 sites. These data are publicly available from the PBO-H20 data portal at http://xenon.colorado.edu/portal. Examples of these data products, and their application are shown in Figures 1-3. Figure 1 is a 2013 time series of volumetric soil moisture (VSM) from station P255 in central California. Figure 2 shows almost eight years of a non-dimensional multipath reflection index (NMRI), which is related to vegetation water content. In addition to NMRI, the annual cumulative precipitation and a satellite derived parameter called NDVI (non-dimensional vegetative index) are shown. An important difference between NMRI and NDVI is that the first is related to water contained within a plant, while the latter is related to its greenness. Figure 3 displays the cumulative difference in snowfall the GPS-IR methods developed within this project and the Snow Data Assimilation System (SNODAS). These results indicate that SNODAS, an important hydrological monitoring and forecasting tool within the western US, is overestimating the snowfall in this region. These results are described in more detail in Boniface, et al. (In Review) This project has significantly leveraged existing infrastructure created by NSF. The Plate Boundary Observatory (PBO) is a NSF funded network of approximately 1200 continuously operating GPS stations in the western United States that was initially created to explore the structure and evolution of the North American Continent and understand the processes that cause earthquakes and volcanic eruptions. The research funded with this award was used to evaluate each of PBO stations for potential use within the PBO-H20 project. As mentioned previously, more than 300 of these stations are now providing water cycle observations in a distributed network across the western United States. This research produced from this project can benefit society by providing observations of various essential climate variables. Potential applications of these data include improving hydrological forecasts, improving agricultural production, and responding to natural hazards such as floods and droughts. Improvements in these areas are directly linked to the U.S. economy and public safety.
