| Abstract: |
The ground permittivity ε_G and snow density ρ_S retrieval scheme, proposed in [1] and experimentally validated in [2], assumes a dry snowpack with uniform density distribution and a homogeneous underlying ground. However, different types of “geophysical noise” can cause serious digressions from the aforementioned assumptions. Snow liquid water content W_S, spatial heterogeneities of ground permittivity, imperfect parametrization of ground roughness, and digressions from the assumption of a uniform snowpack density profile are the four types of “geophysical noise” whose impact on the (ε_G,ρ_S) retrievals is investigated using experimental and synthetic L-band radiometry data. The experimental L-band radiometry data together with in-situ measurements of ground permittivity and temperature, snow density profile, and meteorological data were collected during the 2016/17 winter campaign at the Davos-Laret Remote Sensing Field Laboratory (Switzerland). Furthermore, synthetic analyses are conducted using the emission model developed from parts of the “microwave emission model of layered snowpacks” (MEMLS) coupled with components adopted from the “L-band microwave emission of the biosphere” (L-MEB) model.
Both experimental and synthetic analyses identify major disturbing impacts of snow liquid water W_S on (ε_G,ρ_S) retrievals. Nevertheless, the impact of the other three “geophysical noise” sources is non-negligible but less pronounced. Given the fact that ground permittivity and snow mass-density are two naturally uncorrelated parameters, it is suggested that the coefficient of determination (R^2) between the ε_G and ρ_S retrievals can be used as a “quality indicator” to flag unrealistic retrievals caused by the overall impact of “geophysical noise”. It is further shown that impacts of “geophysical noise” sources are polarization dependent, suggesting to use different retrieval modes RM = “HV”, “H”, or “V” where brightness temperatures T_B^p from both p = H and V, or exclusively p = H or p = V are used, respectively, in the retrieval.
Additionally, as part of ELBARA-II’s raw data processing to achieve calibrated brightness temperatures, a refined RFI mitigation approach is suggested based on fitting a Gaussian model to the distribution of measured raw-data voltage samples. The better performance of the proposed RFI mitigation and detection approach compared to conventional “normality” tests (such as kurtosis and skewness) is demonstrated, and its ability to quantify non-thermal disturbances ∆T_B^p in individual measurements and their use in retrievals is highlighted. It is considered prudent to employ ∆T_B^p in future retrieval schemes using data measured with radiometers that output raw data in time-domain, such as ELBARA-II and the SMAP radiometer.
References
[1] M. Schwank, C. Mätzler, A. Wiesmann, U. Wegmüller, J. Pulliainen, J. Lemmetyinen, K. Rautiainen, C. Derksen, P. Toose, and M. Drusch, "Snow Density and Ground Permittivity Retrieved from L-Band Radiometry: A Synthetic Analysis," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 8, pp. 3833-3845, 2015.
[2] J. Lemmetyinen, M. Schwank, K. Rautiainen, A. Kontu, T. Parkkinen, C. Mätzler, A. Wiesmann, U. Wegmüller, C. Derksen, P. Toose, A. Roy, and J. Pulliainen, "Snow density and ground permittivity retrieved from L-band radiometry: Application to experimental data," Remote Sensing of Environment, vol. 180, pp. 377-391, 2016.
|
