Igor V. Savin

Temperature-Dependable Microwave Dielectric Model for an Arctic Soil

Dielectric measurements of an organic-rich permafrost soil over the range from 1.0 to 16 GHz and from -30°C to +25°C are presented. The measured shrub soil contains up to 90% organic matter and is the first soil of this composition for which the soil dielectric has been characterized. The measurements were fitted to the generalized refractive mixing dielectric model (GRMDM) recently proposed by Mironov et al., which combines the complex refractive indexes for the major components of the soil. These components were found to be the solid content, bound water, transient bound water, liquid capillary water, and moistened ice water. The dielectric properties of the frequency-dispersive components are each described by their own Debye relaxation spectrum. The GRMDM has been modified to incorporate the temperature dependence of the Debye parameters. The phase transformation of the soil water components at the freezing temperature is taken into account. As a result, a temperature-dependable GRMDM (TD GRMDM) has been developed, including model parameters which have a physical interpretation. This TD GRMDM predicts the dielectric for this soil in the whole range of moistures, frequencies, and temperatures measured. The model prediction errors are on the same order as that of dielectric measurements. The model proposed is the first of its kind to provide a physical basis for radar and radiothermal remote sensing algorithms that retrieve the freeze/thaw state and the volumetric moisture in the upper layer of an Arctic soil.

A temperature-dependent multi-relaxation spectroscopic dielectric model for thawed and frozen organic soil at 0.05–15 GHz

The dielectric model for an arctic organic-rich soil collected on the Yamal peninsula (50% of organic matter) both thawed and frozen has been developed. The model is based on the soil dielectric measurements carried out in the ranges of gravimetric moisture 0.03 to 0.55 g/g, dry soil density 0.72 to 0.87 g/cm3, and temperature 25 to -30°C (cooling run), in the frequency range 0.05-15 GHz. To fit the results of measurements of the soil complex dielectric constant as a function of soil moisture and wave frequency, the refractive mixing dielectric model in conjunction with the Debye multi-relaxation equations were applied. As a result, the spectroscopic parameters of dielectric relaxations and electrical specific conductivities for the bound, transient bound, and unbound soil water components were derived, being further complimented with the thermodynamics parameters. Having these parameters, the complex dielectric constant of soil can be predicted as a function of 1) density of dry soil, 2) gravimetric moisture, 3) wave frequency, and 4) temperature1.

A Temperature-Dependent Dielectric Model for Thawed and Frozen Organic Soil at 1.4 GHz

A single-frequency dielectric model for thawed and frozen Arctic organic-rich (80%-90% organic matter) soil was developed. The model is based on soil dielectric data that were measured over the ranges of volumetric moisture from 0.007 to 0.573 cm3/cm3, dry soil density from 0.564 to 0.666 g/cm3, and temperature from 25°C to -30°C (cooling run), at the frequency of 1.4 GHz. The refractive mixing model was applied to fit the measurements of the soils complex refractive index (CRI) as a function of soil moisture, with the values of temperature being fixed. Using the results of this fitting, the parameters of the refractive mixing model were derived as a function of temperature. These parameters involve the CRIs of soil solids as well as bound, transient, and free soil water components. The error of the dielectric model was evaluated by correlating the predicted complex relative permittivity (CRP) values of the soil samples with the measured ones. The coefficient of determination (R2) and the root-mean-square error (RMSE) were estimated to be R2 = 0.999, RMSE = 0.27 and R2 = 0.993, RMSE = 0.18 for the real and imaginary parts of the CRP, respectively. These values are in the order of the dielectric measurement error itself. The proposed dielectric model can be applied in active and passive remote-sensing techniques used in the areas with organicrich soil covers, mainly for the SMOS, SMAP, and Aquarius missions.

Phase transition analysis in freezing moist soils carried out on the basis of phase transitions characteristic to the different types of soil water

The analysis of the mass exchange between the nonfreezing bound and nonfreezing transient water types, as well as that between the nonfreezing transient water and ice in frozen soils has been done. The comparison between results obtained with radiometric and calorimetric methods made it possible to to explain apparent heat capacity behavior during a freezing event using the observed phenomena of phase transition in bound, transient, and capillary soil water types.

The process of unfrozen water freezing with decreasing temperature studied by dielectric measurement in the case of an Arctic soil

Dielectric measurements of an organic-rich permafrost soil over the range from 1.0 to 16 GHz and from −30 °C to +25 °C are presented. The measured shrub soil contains up to 90% organic matter and is the first soil of this composition for which the soil dielectric has been characterized. Using the dielectric data thus obtained, the process of freezing has been analyzed of unfrozen water contained in the shrub tundra sample1.

Dielectric model in the frequency range 0.05 to 15 GHz at temperatures −30°C to 25°C for the samples of organic soils and litter collected in Alaska, Yamal, and Siberian Taiga

We prove the possibility of creating a temperature dependent multi-relaxation spectroscopic dielectric model (TD MRSDM) for a set of organic soil containing 50% (Yamal tundra), 80% (Alaskan tundra) and 90% (boreal pine litter) of organic solids. The created model is based on the previously developed TD MRSDM for the Yamal tundra soil complemented with specific temperature dependences for the hydrological parameters pertaining for each specific soil. At that, the same values for spectroscopic and thermodynamic parameters of the TD MRSDM can be applied, as those were previously obtained in the case of the Yamal tundra soil. The statistical evaluation of the errors for the predicted values of complex dielectric permittivity in the cases of both the thawed and frozen soils with respect to the measured values were carried out. The standard deviations calculated for the Alaskan soil and the Siberian boreal litter appeared to be on the same order as the ones pertaining to the Yamal tundra soil.

Temperature dependent dielectric model at 1.4 GHz for a tundra organic-rich soil thawed and frozen

A mono-frequency dielectric model for a tundra organic-rich soil both thawed and frozen has been developed. The model is based on the soil dielectric measurements carried out in the ranges of volumetric moisture from 0.007 to 0.573 cm3/cm3, dry soil density from 0.564 to 0.666 g/cm3, and temperature from 25 to -30 °C (cooling run), at the frequency of 1.4 GHz used by the SMOS instrument. To fit the results of measurements of the soil complex refractive index (CRI) as a function of soil moisture, the refractive mixing model was applied. As a result, the parameters of the refractive mixing model linked to soil solids, as well as the bound, transient, and free soil water components were derived as a function of temperature. The error of the proposed dielectric model was shown to be in the order of the dielectric measurement error itself1.

Wheat yield estimation in Russia with modis time-series data based on light use efficiency model

Wheat yield estimation in this paper is based on the above ground biomass estimation. The model of above ground biomass estimation requires satellite data, which express the vegetation status. The model was applied in Russia where winter wheat is widely grown. The results showed a high accuracy in estimating winter wheat yield. The range of fractional differences between estimated and observed yields is between -0.40 and 0.50 in 2011, and 83% of the fractional differences are between -0.30 and 0.30. When this method is used in large areas, parameters calibration is crucial. We also summarize that in future study, high resolution images, meteorological factors retrieved by remote sensing data and more field observed data should be used to improve the method.