International Journal of Applied Earth Observation and Geoinformation
Estimation of snow depth and snow water equivalent distribution using airborne microwave radiometry in the Binggou Watershed, the upper reaches of the Heihe River basin
Highlights
► We estimate SWE by airborne microwave radiometers in a cold region watershed. ► SWE was retrieved by differences of brightness temperature (TB) at K- and Ka-bands. ► Blowing snow and sun radiation are two factors for the distribution of snow. ► Change of angles of incidence from terrain is important to research of TB. ► Change of angles of incidence from terrain is not important to retrieval of snow.
Introduction
Snow cover plays an important role in the water and energy cycles at both a global and a regional scale. An accurate estimation of the snow water equivalent (SWE) can improve the understanding of the climate, land surface processes and, in particular, the cold regional hydrological model.
Spaceborne passive microwave (PM) radiometers have provided an opportunity to estimate SWE at both a regional and global scale (Chang et al., 1987, Foster et al., 1997, Che et al., 2003, Kelly et al., 2003). The core of these SWE algorithms is generally the brightness temperature differences (TBD) on two frequencies (generally using K- and Ka-bands), which are based on the volume scattering variations of snowpacks between different wavelengths, and the rule that a larger snow mass results in a larger volume scattering.
However, the spatial resolution from PM satellites is too coarse to introduce or assimilate the SWE into the hydrological process model at the basin scale (Li and Cheng, 2008). The observational data for snow in mountainous stations cannot represent the snow conditions within the PM footprint due to the spatial heterogeneity of the snow distribution. For example, snow data located in mountainous terrain have been removed for the development of the global AMSR-E SWE algorithm (Kelly et al., 2003). The relationship between TBD and in situ SWE in high elevation regions (>2000 m) is quite different from the low elevation regions (Che et al., 2003). It is not surprising to find large errors in the global SWE product algorithm in mountainous regions (Andreadis and Lettenmaier, 2006). Therefore, the estimation of SWE in mountain regions requires further improvements because of (1) the coarse spatial resolution of PM on the satellite and (2) the complexity of the topography. Airborne microwave systems can provide data at higher resolutions, and thus, they were considered as a bridge between the ground and spaceborne observations.
Two airborne passive microwave remote sensing experiments were implemented on the BOREAS campaign to estimate the SWE in February of 1994 and 2003, respectively (Chang et al., 1997, Parde et al., 2007). The results from the experiment in 1994 revealed that the boreal forest was one of the most important factors in estimating SWE (Chang et al., 1997), and thus, a forest modification algorithm was proposed (Foster et al., 1997). Spatial consistency between ground snow measurements and airborne pixel locations was difficult, so the mean measurement on each airborne segment was used to validate the results (Parde et al., 2007). The SWE retrieval algorithm was based on a snow emission model (HUT) that was initialized using standard meteorological measurements (snow depth and air temperature), and this algorithm can estimate the SWE and snow grain size simultaneously.
The Cold Land Processes Experiment (CLPX) in 2003 and 2004 used the airborne polarimetric scanning radiometer to obtain multi-frequency brightness temperature data of different snow conditions (dry, wet, and refrozen). A series of empirical SWE retrieval algorithms, which were similar to the Chang algorithm but with more frequencies and polarizations, were analyzed according to the relationships between emissivity and snow properties (Stankov et al., 2008). Their results also suggested that a new algorithm should be developed to account for local conditions, such as snow properties and land cover.
Recently, two experiments between Canada and Finland were carried out to compare airborne microwave brightness temperature and snow properties in these two regions in 2005 and 2006, respectively (Lemmetyinen et al., 2009). The comparison showed that larger snow grain sizes lead to a lower brightness temperature, which indicates that grain size is an important factor in the volume scatter of snowpack. These studies suggested that future studies should measure the snow stratigraphic condition in the field.
In this study, to obtain the SWE at a high spatial resolution, the airborne microwave radiometers were used to observe the TB at a K-band (18.7 GHz) and Ka-band (36.0 GHz). Furthermore, we adopted a nadir observation to reduce the influences of topography, particularly the local incidence angle changes, on TB. A simple snow sampling scheme was implemented to obtain the snow's stratigraphic condition by simultaneous ground observations. The microwave emission model of layered snowpacks (MEMLS) was used to simulate the relationship between TB and SD/SWE. This study is one of many airborne remote sensing experiments within the watershed allied telemetry experimental research (WATER) (Li et al., 2009). The aim of this airborne mission was to retrieve SD/SWE at a watershed scale in cold and mountain regions.
Section snippets
Study area
The Binggou Watershed, with an area of 30.48 km2, is located in the boundary regions of the Qinghai–Tibet plateau (see Fig. 1). In hydrology, this watershed belongs to the upstream of the Heihe River basin and its elevation varies from 3450 m to 4400 m (Li et al., 2009). Melted snow water is one of the most important water resources in the Heihe River basin, which supplies agriculture in the middle reaches and the ecosystem in the lower reaches.
Airborne microwave radiometry observations
In the Binggou Watershed, the airborne microwave
TB comparison of simulation and observation
The brightness temperature difference (TBD) at 18/19 and 36 GHz has been successfully used to derive volume scattering signatures of snowpacks (Chang et al., 1987, Foster et al., 1997, Kelly et al., 2003, Che et al., 2008), which is the kernel of retrieval algorithms of snow depth or SWE. In this study, we simulated the TB at 18.7 and 36 GHz by MEMLS and analyzed the relationships between TBD and SD/SWE.
The AIEM was used to calculate the reflectivity (s0) between the snow and ground. First, we
MEMLS model and retrieval algorithm
The MEMLS model was developed according to many field measurements that determines several basic relationships between snow physical variables and microwave properties, such as the permittivity (Mätzler, 1996), scattering and absorption coefficients (Wiesmann et al., 1998). Several snow-pit data with ground-based microwave radiometer measurements have proved the MEMLS can simulate brightness temperature of snow (Wiesmann and Mätzler, 1999, Mätzler and Wiesmann, 1999, Tedesco and Kim, 2006,
Acknowledgments
We wish to thank two anonymous reviewers for their helpful comments and suggestions which significantly improved this paper. The work was funded by the Chinese Academy of Sciences Action Plan for West Development Project ’Watershed Allied Telemetry Experimental Research (WATER)’ (KZCX2-XB2–09), the China State Key Basic Research Project (2007CB714400), and the National Natural Science Foundation of China (40971188&40701030). Authors thank every participant in cold region experiment of WATER.
References (28)
- et al.
Assimilating remotely sensed snow observations into a macroscale hydrology model
Advances in Water Resources
(2006) - et al.
Comparison of snow mass estimates from prototype passive microwave snow algorithm, a revised algorithm and a snow depth climatology
Remote Sensing of Environment
(1997) - et al.
Quantifying the uncertainty in passive microwave snow water equivalent observations
Remote Sensing of Environment
(2005) - et al.
Extension of the microwave emission model of layered snowpacks to coarse-grained snow
Remote Sensing of Environment
(1999) - et al.
Inversion of a passive microwave snow emission model for water equivalent estimation using airborne and satellite data
Remote Sensing of Environment
(2007) - et al.
Microwave emission model of layered snowpacks
Remote Sensing of Environment
(1999) - et al.
Nimbus 7 SMMR derived global snow cover parameters
Annuals of Glaciology
(1987) - et al.
Snow parameters derived from microwave measurements during the BOREAS winter field campaign
Journal of Geophysical Research-Atmospheres
(1997) - et al.
Estimation of snow water equivalent from passive microwave remote sensing data (SSM/I) in Tibetan Plateau
- et al.
Snow depth derived from passive microwave remote-sensing data in China
Annals of Glaciology
(2008)
Emission of rough surfaces calculated by the integral equation method with comparison to three-dimensional moment method Simulations
IEEE Transactions on Geoscience and Remote Sensing
Quantifying uncertainty in modeling snow microwave radiance for a mountain snowpack at the point-scale, including stratigraphic effects
IEEE Transactions on Geoscience and Remote Sensing
A distributed model of blowing snow over complex terrain
Hydrological Processes
Microwave Scattering and Emission Models and their Applications
Cited by (24)
Improving snow depth estimation by coupling HUT-optimized effective snow grain size parameters with the random forest approach
2021, Remote Sensing of EnvironmentCitation Excerpt :From a theoretical perspective, spatial and temporal variations in snow microstructure have a large effect on observed microwave emissions, complicating the relation of TB to snow depth (Tedesco and Jeyaratnam, 2016; Brown et al., 2018; Kim et al., 2018; Pulliainen et al., 2020). Many airborne PMW remote sensing experiments, e.g., implemented in Canada (Derksen, 2008), Finland (Lemmetyinen et al., 2009), China (Che et al., 2012) and North America (the SnowEx campaign, see Brucker et al., 2017), also showed that snow metamorphism tends to lead to a lower TB at 36.5 GHz even though the snow depth is stable, which highlights the importance of snow GS in the volume scattering of snowpack. In microwave snow emission models, snow metamorphism has typically been parametrized as a function of microstructure metrics related to GS (described in Section 3.1).
Multi-sensor land data assimilation: Toward a robust global soil moisture and snow estimation
2018, Remote Sensing of EnvironmentCitation Excerpt :Some widely used global soil moisture retrievals are from AMSR-E (Advanced Microwave Scanning Radiometer for Earth Observing System) (Koike et al., 2004; Njoku et al., 2003; Owe et al., 2008), ASCAT (METOP-A Advanced Scatterometer) (Bartalis et al., 2007), SMOS (Soil Moisture and Ocean salinity) (Kerr et al., 2016), and SMAP (Soil Moisture Active and Passive) (Entekhabi et al., 2014). In terms of snow, while passive and active microwave are also used to estimate snow mass (Chang et al., 1982; Che et al., 2012, 2016; Dai et al., 2012), visible and near-infrared signals are often used to detect snow extent over most of land surfaces (Hall et al., 2002). However, uncertainties and biases are commonly observed in satellite retrieved soil moisture (Al-Yaari et al., 2014; Jackson et al., 2010), snow cover fraction (SCF) or snow water equivalent under certain circumstances (Foster et al., 2005; Hall and Riggs, 2007).
Estimation of snow depth from passive microwave brightness temperature data in forest regions of northeast China
2016, Remote Sensing of EnvironmentCitation Excerpt :The Scanning Multichannel Microwave Radiometer (SMMR), Special Sensor Microwave Imager (SSM/I), and Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) have been used to retrieve the snow depth or snow water equivalent since the 1970s (Chang, Foster, & Hall, 1987; Foster, Hall, Kelly, & Chiu, 2009; Frei et al., 2012; Hancock et al., 2013; Tedesco & Miller, 2007). Studies have shown that knowledge of snow properties, including stratigraphy, snow grain size and snow density, is important in estimating snow depth from passive microwave data (Che, Dai, Wang, Zhao, & Liu, 2012; Dai, Che, Wang, & Zhang, 2012; Davenport, Sandells, & Gurney, 2012; Durand & Liu, 2012). All remote sensing methods face a problem in deriving snow cover information in vegetation-covered regions, particularly in forest regions (Rees, 2006).
Assimilating passive microwave remote sensing data into a land surface model to improve the estimation of snow depth
2014, Remote Sensing of EnvironmentCitation Excerpt :Passive microwave (PM) remote sensing has the capability of providing snow depth (SD) and snow water equivalent (SWE) information independent of weather or light conditions. PM remote sensing data, such as data from the scanning multichannel microwave radiometer (SMMR), special sensor microwave/imager (SSM/I), and Advanced Microwave Scanning Radiometer for Earth Observation System (AMSR-E), enable the retrieval of SDs and SWEs on large regional and global scales (Chang, Foster, & Hall, 1987; Che, Dai, Wang, Liu & Zhao, 2012; Dai, Che, Wang, & Zhang, 2012; Foster, Chang, & Hall, 1997; Kelly & Chang, 2003; Tedesco & Narvekar, 2010). Such retrieval algorithms have a common kernel of brightness temperature difference (TBD) at 18 and 37 GHz (or similar frequencies) that is based on the fact that larger amounts of snow crystals can lead to greater volume scattering, which corresponds to a larger TBD.
Snow depth and snow water equivalent estimation from AMSR-E data based on a priori snow characteristics in Xinjiang, China
2012, Remote Sensing of EnvironmentCitation Excerpt :Therefore, when snow depth is more than this threshold, the lower frequencies (such as 10 GHz and 18 GHz) can be used to retrieve snow depth/SWE. In fact, given stable snow depth, the TBD will increase with increasing grain size and decrease with increasing snow density (Che et al., 2011; Foster et al., 2005; Tsang et al., 2000). Therefore, it is necessary to develop an algorithm that accounts for local snow properties (Che et al., 2008; Stankov et al., 2008).
Retrieval of key eco-hydrological parameters for cold and arid regions
2012, International Journal of Applied Earth Observation and Geoinformation