A new study shows that the 40-year and ongoing airborne gamma Snow Water Equivalent (SWE) dataset from the National Oceanic and Atmospheric Administration (NOAA), which is a key component in water management, has substantial potential to be used as a long-term, reliable reference SWE across the United States and Southern Canada.
Credit: Airborne Snow Survey Program https://www.nohrsc.noaa.gov/snowsurvey/
Snowpack is a crucial component of the hydrological cycle, impacting our lives as a water supply source (e.g., in the western U.S.) and potential source of flooding (e.g., in north-central and the northeastern U.S.). Accurate, timely estimates of the snowpack are required over the U.S. to manage water under a changing climate and a growing population that demands more water. The most hydrologically-relevant measure of the snowpack is Snow Water Equivalent (SWE)—the depth of water that would result if the entire snowpack were to melt. Reliable real-time observations and long-term records of SWE are essential for identifying climate variability and for developing effective water management and flood risk assessments. This is especially important as changes in seasonal snow have accelerated across the U.S. in the last few decades.
Historically, point-scale long-term SWE records from snow station networks have been used to provide high-quality measurements. However, they do not necessarily represent the snowpack distribution, especially in regions with complex terrain and land cover. To overcome the limitations of point-scale measurements, remote sensing techniques have offered an opportunity to develop observation-based gridded SWE products using satellite remote sensing and/or in situ snow station networks with assimilation techniques.
In the 1970-1980s, snow remote sensing pioneers had estimated SWE from airborne measurements using gamma-ray surveying to measure areal SWE accurately. The NOAA airborne gamma snow survey was designed to help hydrologists and flood forecasters improve operational spring flood predictions and water supply outlooks. The naturally occurring gamma emission is attenuated by the mass of overlying water—whether ice or snow—and can be used to calculate the SWE within 35 cm depth accuracy (Rees, 2006). Terrestrial gamma radiation is detected from an aircraft flying 150 m above the snow-covered ground (Figure 1). Since the water mass in the snow cover attenuated the terrestrial radiation signal, the difference between the terrestrial measurements over the bare ground and snow-covered ground can be used to estimate the mean SWE.
Figure 1. The ground is always emitting gamma radiation. On the left: Flight at about 150 m (or 500 feet) above the ground when there is no snow on the ground to impact the radiation signature, creating a reference signature along that flight line; On the right: Snow will accumulate through the season and flight is repeated along the same flight line at 150 m elevation. There is a direct relationship between how much radiation is absorbed and how much water there is. So, more water causes less terrestrial radiation detected by the sensor on the aircraft. [Depicted from Carrie Olheiser presentation on https://www.youtube.com/watch?v=NZU1yw5N8qU&feature=youtu.be]
Despite the accuracy of the airborne gamma SWE studies, there have been few gamma SWE studies in the snow remote sensing community over the last 30 years. This year, however, Cho et al. (2020) published a study evaluating three daily, long‐term (>30 years) SWE products in relation to the NOAA historical airborne gamma SWE observations. The datasets compared were the empirically-based SWE from passive microwave (SSMI/S), GlobSnow‐2 SWE, and the University of Arizona (UA) SWE from 1982 to 2017 over the conterminous United States. SSMI/S uses microwave frequencies (in both horizontal and vertical polarizations) to estimate SWE. GlobSnow-2 utilizes a data assimilation approach—combining ground‐based snow depth with passive microwave satellite estimates. UA assimilates in-situ snow measurements with modeled, gridded temperature and precipitation data to estimate SWE.
In their evaluation, Cho et al. (2020) determined that UA SWE has the strongest agreement with gamma SWE across all regions, while there are relatively weak agreements in the evergreen needleleaf forest and grasslands land-cover and tundra snow-class, most likely due to potential limitations of UA products as well as gamma SWE in the mountainous regions. GlobsSnow-2 generally agrees better with gamma SWE in forested areas (mixed forest, deciduous broadleaf forest) and maritime snow-class in the northeastern U.S. SSMI/S has the poorest agreements in croplands/grasslands land-cover and prairie snow-class over the northern and northeastern U.S. It also has an extremely low correlation in evergreen needleleaf forest and grasslands in mountainous regions over the western U.S.—corresponding to weaknesses of microwave satellite‐driven signals (Figure 2).
Figure 2. Correlation (R value) maps of daily SSMI/S, GlobSnow‐2, and UA snow water equivalent with daily NOAA airborne gamma radiation snow water equivalent for each gamma flight line from 1982 to 2017 (black color represents that the R value is a negative value) [from Cho et al., 2020]
In their research, Dr. Cho and co-authors provide evidence for the value of the 40-year historical airborne gamma SWE. They found that the NOAA’s long-term airborne SWE record is of great importance, especially in forested regions, where SWE estimation remains challenging. Their findings show that the NOAA gamma SWE has substantial potential to be used as a long-term reliable independent dataset to evaluate the gridded satellite and assimilated SWE products from the evolving land surface models and regional climate models.
Biography
Dr. Eundang Cho received his Ph.D. in Civil and Environmental Engineering from University of New Hampshire in 2020. Dr. Cho is now a postdoctoral scholar at the University of New Hampshire since May 2020 working with Dr. Jennifer Jacobs. He is focusing on accurate and timely estimations and predictions of critical hydrologic fluxes, storage, and processes for enhanced water resources management, flood prediction, and infrastructure design. For this, he uses field observations, multiple remote sensing techniques (active and passive microwave, infrared, thermal, and gamma radiation) via UAS, aircraft, and satellite platforms, and climate and hydrological model simulations along with big-data analytics. One of his achievements during his Ph.D. program was a visit to the NASA Goddard Space Flight Center funded through the 2020 CUAHSI Pathfinder Fellowship program. His research interests includes several aspects of hydrology and flood forecasting, remote sensing, land use and land cover change, and climate change impacts.
Reference
Cho, E., Jacobs, J. M., & Vuyovich, C. M. ( 2020). The value of long‐term (40 years) airborne gamma radiation SWE record for evaluating three observation‐based gridded SWE data sets by seasonal snow and land cover classifications. Water Resources Research, 56, e2019WR025813. https://doi.org/10.1029/2019WR025813
Rees, W.G. 2006. Remote sensing of snow and ice. Boca Raton, FL, CRC Press.
By Irene Garousi-Nejad Utah State University