Helsinki University Of Technology Model Research Articles

Brightness temperature simulation of snow cover based on snow grain size evolution usingin situdata

Snow depth parameter inversion from passive microwave remote sensing is of great significance to hydrological process and climate systems. The Helsinki University of Technology (HUT) model is a commonly used snow emission model. Snow grain size (SGS) is one of the important input parameters, but SGS is difficult to obtain in broad areas. The time series of SGS are first evolved by an SGS evolution model (Jordan 91) using in situ data. A good linear relationship between the effective SGS in HUT and the evolution SGS was found. Then brightness temperature simulations are performed based on the effective SGS and evolution SGS. The results showed that the biases of the simulated brightness temperatures based on the effective SGS and evolution SGS were −6.5 and −3.6 K , respectively, for 18.7 GHz and −4.2 and −4.0 K for 36.5 GHz. Furthermore, the model is performed in six pixels with different land use/cover type in other areas. The results showed that the simulated brightness temperatures based on the evolution SGS were consistent with those from the satellite. Consequently, evolution SGS appears to be a simple method to obtain an appropriate SGS for the HUT model.

Arctic Snow Microstructure Experiment for the development of snow emission modelling

Abstract. The Arctic Snow Microstructure Experiment (ASMEx) took place in Sodankylä, Finland in the winters of 2013–2014 and 2014–2015. Radiometric, macro-, and microstructure measurements were made under different experimental conditions of homogenous snow slabs, extracted from the natural seasonal taiga snowpack. Traditional and modern measurement techniques were used for snow macro- and microstructure observations. Radiometric measurements of the microwave emission of snow on reflector and absorber bases were made at frequencies 18.7, 21.0, 36.5, 89.0, and 150.0 GHz, for both horizontal and vertical polarizations. Two measurement configurations were used for radiometric measurements: a reflecting surface and an absorbing base beneath the snow slabs. Simulations of brightness temperatures using two microwave emission models, the Helsinki University of Technology (HUT) snow emission model and Microwave Emission Model of Layered Snowpacks (MEMLS), were compared to observed brightness temperatures. RMSE and bias were calculated; with the RMSE and bias values being smallest upon an absorbing base at vertical polarization. Simulations overestimated the brightness temperatures on absorbing base cases at horizontal polarization. With the other experimental conditions, the biases were small, with the exception of the HUT model 36.5 GHz simulation, which produced an underestimation for the reflector base cases. This experiment provides a solid framework for future research on the extinction of microwave radiation inside snow.

Snow Depth Inversion Using the Localized HUT Model Based on FY-3B MWRI Data in the Farmland of Heilongjiang Province, China

Snow depth parameter inversion in the farmland using passive microwave remote sensing is of great significance to the agricultural production in Northeast China. Firstly, the Helsinki University of Technology (HUT) snow emission model was validated in the farmland based on microwave radiation imager (MWRI) onboard FengYun-3B satellite (FY-3B). The results showed that there was a big difference between the brightness temperature of HUT model simulation and MWRI for 18.7 GHz horizontal polarization (18.7 H) and 36.5 GHz horizontal polarization (36.5 H). To improve HUT model, the empirical parameter in the model was localized. Then the localized HUT (LHUT) model was built, where the extinction coefficient was calculated by the new extinction coefficient formula. Next, LHUT model was validated based on MWRI data and compared with HUT model. The results showed that LHUT underestimates slightly the brightness temperature with 0.91 and 4.19 K for 18.7 and 36.5 H respectively, and LHUT is superior to HUT model. Finally, the genetic algorithm (GA) was used to invert snow depth based on LHUT. The results showed that snow depth was underestimated with 6.79 cm based on LHUT. The inverted snow depth based on LHUT model is in better agreement with the measured snow depth.

Differences Between the HUT Snow Emission Model and MEMLS and Their Effects on Brightness Temperature Simulation

Microwave emission models are a critical component of snow water equivalent retrieval algorithms applied to passive microwave measurements. Several such emission models exist, but their differences need to be systematically compared. This paper compares the basic theories of two models: the multiple-layer Helsinki University of Technology (HUT) model and the microwave emission model of layered snowpacks (MEMLS). By comparing the mathematical formulation side by side, three major differences were identified: 1) by assuming that the scattered intensity is mostly (96%) in the forward direction, the HUT model simplifies the radiative transfer equation in $4\pi$ space into two one-flux equations, whereas MEMLS uses a two-flux theory; 2) the HUT scattering coefficient is much larger than the one of MEMLS; and 3) MEMLS considers the trapped radiation inside snow due to internal reflection by a six-flux model, which is not included in HUT. Simulation experiments indicate that the large scattering coefficient of the HUT model compensates for its large forward scattering ratio to some extent, but the effects of one-flux simplification and the trapped radiation still result in different $T_{B}$ simulations between the HUT model and MEMLS. The models were compared with observations of natural snow cover at Sodankyla, Finland; Churchill, Canada; and Colorado, USA. No optimization of the snow grain size was performed. It shows that the HUT model tends to underestimate $T_{B}$ for deep snow. MEMLS with the physically based improved Born approximation performed best among the models, with a bias of −1.4 K and a root-mean-square error of 11.0 K.

Plot Scale Passive Microwave Measurements and Modeling of Layered Snow Using the Multi-layered HUT Model

. The excavation of snow layers was used to compare observed and modeled plot-scale brightness temperature (TB) values for 19 and 37 GHz, with regard to snow water equivalent (SWE), snow type, grain size, and layered structure, for 3 land cover types acquired near Churchill, Manitoba, Canada in March of 2010. In situ snow geophysical measurements were input to the Helsinki University of Technology (HUT) multi-layer snow emission model, and performance was characterized by RMSE and MBE. Emission scattering from depth hoar was disproportionate to its SWE contribution when compared to other snow types, and it also masked observed scattering contributions from upper snow layers. The simulated and observed TB diverged above 130 mm SWE, as the model did not capture scattering extinction or snowpack emission. These may impact the initial effective grain size optimization process if applied to the GlobSnow SWE assimilation technique. Grain size is optimized to fit emission simulations to TB from satellite observations; the model's continued sensitivity to SWE above observed signal saturation may be attributed to the existing snowpack. These inaccuracies may then be carried forward in the assimilation process, through variance calculations used to weight the contributions of assimilation modules, leading to less accurate SWE retrievals.Résumé. L’excavation des couches de neige a été utilisée pour comparer les valeurs observées et modélisées de la température de brillance (TB) à 19 et 37 GHz à l’échelle de la parcelle, en lien avec l’équivalent en eau de la neige (SWE), le type de neige, la taille des grains, et la structure en couches, pour trois types de couverture terrestre acquises près de Churchill au Manitoba (Canada) en mars 2010. Les mesures géophysiques in situ de neige ont été utilisées en entrées du modèle HUT (Helsinki University of Technology), un modèle multicouche d’émission de la neige, et la performance a été caractérisée par l’écart moyen quadratique (RMSE) et l’erreur de justesse moyenne (MBE). La diffusion de l’émission du givre de profondeur était disproportionnée par rapport à sa contribution au SWE en comparaison avec d’autres types de neige et a aussi masqué les contributions observées de la diffusion des couches de neige supérieures. Les valeurs de TB simulées et observées ont divergé au-dessus de 130 mm de SWE, parce que le modèle n’a pas représenté l’extinction de la diffusion ou l’émission du manteau neigeux. Cela peut affecter le processus d’optimisation initiale de la taille effective des grains, s’il est appliqué à la technique d’assimilation GlobSnow SWE. La taille des grains est optimisée pour ajuster les simulations d’émission aux TB provenant des observations par satellite; la sensibilité du modèle au SWE au-dessus de la saturation du signal observé peut être attribuée au manteau neigeux existant. Ces inexactitudes peuvent ensuite être reportées dans le processus d’assimilation, par des calculs de la variance utilisés pour pondérer les contributions des modules d’assimilation ce qui conduit à des résultats moins précis d’estimation du SWE.

Simulating seasonally and spatially varying snow cover brightness temperature using HUT snow emission model and retrieval of a microwave effective grain size

The Helsinki University of Technology (HUT) snow emission model forms the basis of the European Space Agency's GlobSnow snow water equivalent (SWE) product (Takala et al., 2011). The model applies a semi-empirical radiative transfer calculation to account for the interaction of the snow medium with microwaves; separate components are applied to account for vegetation, the atmosphere, and emission from the ground surface. For the retrieval of SWE, an innovative method is used to account for spatial and temporal variability in snow conditions by retrieving an effective parameter describing the scattering behavior of microwaves in the snow (a proxy indicator of the microwave effective snow grain size). In this study, the influence of differing snow conditions, as well as varying land cover, on the retrieved microwave effective snow grain size was analyzed. Passive microwave measurements were acquired using tower-based, mobile sled-based and airborne radiometers in a mixed forest environment near Sodankylä, Finland. Forward simulations at the tower site over an entire winter period showed that the use of an empirical relation to modify the classical in situ measured grain size produced HUT model bias errors less than 6K on average at 19 and 37GHz from a ~20–80cm deep boreal snowpack. Model simulations for airborne and sled-based observations showed that using a simplified 2-layer representation of the snowpack improves simulation biases and RMSE, although modification of the measured grain size was again necessary to achieve these results, regardless of the layering configuration. The microwave effective grain size retrieved from HUT predictions was closely related to a simple average grain size measured in situ, both in terms of magnitude and temporal trend. This is an important finding as the retrieval scheme of Takala et al. (2011) relies on the microwave effective grain size to retain a degree of physical basis for values to be generally applicable over larger areas, which is challenging because this free parameter also accounts for errors unrelated to the effective snow grain size (i.e. vegetation, soil conditions, scene heterogeneity, etc.).

Snow stratigraphic heterogeneity within ground‐based passive microwave radiometer footprints: Implications for emission modeling

Two-dimensional measurements of snowpack properties (stratigraphic layering, density, grain size, and temperature) were used as inputs to the multilayer Helsinki University of Technology (HUT) microwave emission model at a centimeter-scale horizontal resolution, across a 4.5 m transect of ground-based passive microwave radiometer footprints near Churchill, Manitoba, Canada. Snowpack stratigraphy was complex (between six and eight layers) with only three layers extending continuously throughout the length of the transect. Distributions of one-dimensional simulations, accurately representing complex stratigraphic layering, were evaluated using measured brightness temperatures. Large biases (36 to 68 K) between simulated and measured brightness temperatures were minimized (−0.5 to 0.6 K), within measurement accuracy, through application of grain scaling factors (2.6 to 5.3) at different combinations of frequencies, polarizations, and model extinction coefficients. Grain scaling factors compensated for uncertainty relating optical specific surface area to HUT effective grain size inputs and quantified relative differences in scattering and absorption properties of various extinction coefficients. The HUT model required accurate representation of ice lenses, particularly at horizontal polarization, and large grain scaling factors highlighted the need to consider microstructure beyond the size of individual grains. As variability of extinction coefficients was strongly influenced by the proportion of large (hoar) grains in a vertical profile, it is important to consider simulations from distributions of one-dimensional profiles rather than single profiles, especially in sub-Arctic snowpacks where stratigraphic variability can be high. Model sensitivity experiments suggested that the level of error in field measurements and the new methodological framework used to apply them in a snow emission model were satisfactory. Layer amalgamation showed that a three-layer representation of snowpack stratigraphy reduced the bias of a one-layer representation by about 50%.

Brightness Temperature Simulations of the Canadian Seasonal Snowpack Driven by Measurements of the Snow Specific Surface Area

Snow grain size is the snowpack parameter that most affects the microwave snow emission. The specific surface area (SSA) of snow is a metric that allows rapid and reproducible field measurements and that well represents the grain size. However, this metric cannot be used directly in microwave snow emission models (MSEMs). The aim of this paper is to evaluate the suitability and the adaptations required for using the SSA in two MSEMs, i.e., the Dense Media Radiative Theory-Multilayer model (DMRT-ML) and the Helsinki University of Technology model (HUT n-layer), based on in situ radiometric measurements. Measurements of the SSA, using snow reflectance in the short-wave infrared, were taken at 20 snowpits in various environments (e.g., grass, tundra, and dry fen). The results show that both models required a scaling factor for the SSA values to minimize the root-mean-square error between the measured and simulated brightness temperatures. For DMRT-ML, the need for a scaling factor is likely due to the oversimplified representation of snow as spheres of ice with a uniform radius. We hypothesize that the need for a scaling factor is related to the grain size distribution of snow and the stickiness between grains. For HUT n-layer, using the SSA underestimates the attenuation by snow, particularly for snowpacks with a significant amount of depth hoar. This paper provides a reliable description of the grain size for DMRT-ML, which is of particular interest for the assimilation of satellite passive microwave data in snow models.

CREST-Snow Field Experiment: analysis of snowpack properties using multi-frequency microwave remote sensing data

Abstract. The CREST-Snow Analysis and Field Experiment (CREST-SAFE) was carried out during January–March 2011 at the research site of the National Weather Service office, Caribou, ME, USA. In this experiment dual-polarized microwave (37 and 89 GHz) observations were accompanied by detailed synchronous observations of meteorology and snowpack physical properties. The objective of this long-term field experiment was to improve understanding of the effect of changing snow characteristics (grain size, density, temperature) under various meteorological conditions on the microwave emission of snow and hence to improve retrievals of snow cover properties from satellite observations. In this paper we present an overview of the field experiment and comparative preliminary analysis of the continuous microwave and snowpack observations and simulations. The observations revealed a large difference between the brightness temperature of fresh and aged snowpack even when the snow depth was the same. This is indicative of a substantial impact of evolution of snowpack properties such as snow grain size, density and wetness on microwave observations. In the early spring we frequently observed a large diurnal variation in the 37 and 89 GHz brightness temperature with small depolarization corresponding to daytime snowmelt and nighttime refreeze events. SNTHERM (SNow THERmal Model) and the HUT (Helsinki University of Technology) snow emission model were used to simulate snowpack properties and microwave brightness temperatures, respectively. Simulated snow depth and snowpack temperature using SNTHERM were compared to in situ observations. Similarly, simulated microwave brightness temperatures using the HUT model were compared with the observed brightness temperatures under different snow conditions to identify different states of the snowpack that developed during the winter season.

Comparison of the multi-layer HUT snow emission model with observations of wet snowpacks

Information on snow properties plays an important role in hydrological, meteorological and climatological applications. Passive microwave remote sensing is an effective method to retrieve snowpack parameters; however, the observations can be obscured if there is wet snow in the satellite footprint. To study the emission properties of wet snow and check its response to snow wetness, this paper applies the multi-layer Helsinki University of Technology (HUT) snow emission model coupled with the Advanced Integral Equation Model to simulate the low-wetness snowpack observed at Luancheng in November 2009, and the high wetness snowpack observed at Weissfluhjoch in June 1995. Input parameters are acquired by the in-situ snow pit measurements, while the snow grain size is fitted by comparing model predictions with the observed passive microwave signals at a range of observing angles. Results show that the application of a multi-layer model is capable to consider the distribution pattern of the snow wetness along the snow profile and the refrozen ice crust of the snow surface. The multi-layer HUT model is able to reproduce the wet snow emission properties, with an rms error of 4.4 K (at Luancheng) and 5.7 K (at Weissfluhjoch) at vertical polarization, and an rms error of 7.9 K (at Luancheng) and 11.4 K (at Weissfluhjoch) at horizontal polarization. Copyright © 2012 John Wiley & Sons, Ltd.

Determination of snow water equivalent over the eastern part of Turkey using passive microwave data

AbstractInformation on regional snow water equivalent (SWE) is required for the management of water generated from snowmelt. Modeling of SWE in the mountainous regions of eastern Turkey, one of the major headwaters of Euphrates–Tigris basin, has significant importance in forecasting snowmelt discharge, especially for optimum water usage. An assimilation process to produce daily SWE maps is developed based on Helsinki University of Technology (HUT) model and AMSR‐E passive microwave data. The characteristics of the HUT emission model are analyzed in depth and discussed with respect to the extinction coefficient function. A new extinction coefficient function for the HUT model is proposed to suit models for snow over mountainous areas. Performance of the modified model is checked against the original, other modified cases and ground truth data covering the 2003–2007 winter periods. A new approach to calculate grain size and density is integrated inside the developed data assimilation process. An extensive validation was successfully performed by means of snow data measured at ground stations during the 2008–2010 winter periods. The root mean square error of the data set for snow depth and SWE between January and March of the 2008–2010 periods compared with the respective AMSR‐E footprints indicated that errors for estimated snow depth and predicted SWE values were 16.92 cm and 40.91 mm, respectively, for the 3‐year period. Validation results were less satisfactory for SWE less than 75.0 mm and greater than 150.0 mm. An underestimation for SWE greater than 150 mm could not be resolved owing to the microwave signal saturation that is observed for dense snowpack. Copyright © 2012 John Wiley & Sons, Ltd.

Evaluation of passive microwave brightness temperature simulations and snow water equivalent retrievals through a winter season

Plot-scale brightness temperature (T B) measurements at 6.9, 19, 37, and 89 GHz were acquired in forest, open, and lake environments near Churchill, Manitoba, Canada with mobile sled-based microwave radiometers during the 2009–2010 winter season. Detailed physical snow measurements within the radiometer footprints were made to relate the microwave signatures to the seasonal evolution of the snowpack, and provide inputs for model simulations with the Helsinki University of Technology (HUT) snow emission model. Large differences in depth, density, and grain size were observed between the three land cover types. Plot-scale simulations with the HUT model showed a wide range in simulation accuracy between sites and frequencies. In general, model performance degraded when the effective grain size exceeded 2 mm and/or there was an ice lens present in the pack. HUT model performance improved when simulations were run regionally at the satellite scale (using three proportional land cover tiles: open, forest, and lake) and compared to Advanced Microwave Scanning Radiometer (AMSR-E) measurements. Root mean square error (RMSE) values ranged from approximately 4 to 16 K depending on the frequency, polarization, and land cover composition of the grid cell. Snow water equivalent (SWE) retrievals produced using forward T B simulations with the HUT model in combination with AMSR-E measurements produced RMSE values below 25 mm for the intensive study area. Retrieval errors exceeded 50 mm when the scheme was applied regionally.

Investigation of the snow‐cover dynamics in the Upper Euphrates Basin of Turkey using remotely sensed snow‐cover products and hydrometeorological data

AbstractThe Euphrates and Tigris rivers serve as the most important water resources in the Middle East. Precipitation in this region falls mostly in the form of snow over the higher elevations of the Euphrates Basin and remains on the ground for nearly half of the year. This snow‐covered area (SCA) is a key element of the hydrological cycle, and monitoring the SCA is crucial for making accurate forecasts of snowmelt discharge, especially for energy production, flood control, irrigation, and reservoir‐operation optimization in the Upper Euphrates (Karasu) Basin. Remote sensing allows the detection of the spatio‐temporal patterns of snow cover across large areas in inaccessible terrain, such as the eastern part of Turkey, which is highly mountainous. In this study, a seasonal evaluation of the snow cover from 2000 to 2009 was performed using 8‐day snow‐cover products (MOD10C2) and the daily snow‐water equivalent (SWE) product. The values of SWE products were obtained using an assimilation process based on the Helsinki University of Technology model using equal area Special Sensor Microwave Imager (SSM/I) Earth‐gridded advanced microwave scanning radiometer—EOS daily brightness‐temperature values. In the Karasu Basin, the SCA percentage for the winter period is 80–90%. The relationship between the SCA and the runoff during the spring period is analysed for the period from 2004 to 2009. An inverse linear relationship between the normalized SCA and the normalized runoff values was obtained (r = 0·74). On the basis of the monthly mean temperature, total precipitation and snow depth observed at meteorological stations in the basin, the decrease in the peak discharges, and early occurrences of the peak discharges in 2008 and 2009 are due to the increase in the mean temperature and the decrease in the precipitation in April. Copyright © 2011 John Wiley & Sons, Ltd.

Evaluation of the HUT modified snow emission model over lake ice using airborne passive microwave measurements

The algorithms designed to estimate snow water equivalent (SWE) using passive microwave measurements falter in lake-rich high-latitude environments due to the emission properties of ice covered lakes on low frequency measurements. Microwave emission models have been used to simulate brightness temperatures (Tbs) for snowpack characteristics in terrestrial environments but cannot be applied to snow on lakes because of the differing subsurface emissivities and scattering matrices present in ice. This paper examines the performance of a modified version of the Helsinki University of Technology (HUT) snow emission model that incorporates microwave emission from lake ice and sub-ice water. Inputs to the HUT model include measurements collected over brackish and freshwater lakes north of Inuvik, Northwest Territories, Canada in April 2008, consisting of snowpack (depth, density, and snow water equivalent) and lake ice (thickness and ice type). Coincident airborne radiometer measurements at a resolution of 80 × 100 m were used as ground-truth to evaluate the simulations. The results indicate that subsurface media are simulated best when utilizing a modeled effective grain size and a 1 mm RMS surface roughness at the ice/water interface compared to using measured grain size and a flat Fresnel reflective surface as input. Simulations at 37 GHz (vertical polarization) produce the best results compared to airborne Tbs, with a Root Mean Square Error (RMSE) of 6.2 K and 7.9 K, as well as Mean Bias Errors (MBEs) of −8.4 K and −8.8 K for brackish and freshwater sites respectively. Freshwater simulations at 6.9 and 19 GHz H exhibited low RMSE (10.53 and 6.15 K respectively) and MBE (−5.37 and 8.36 K respectively) but did not accurately simulate Tb variability (R = −0.15 and 0.01 respectively). Over brackish water, 6.9 GHz simulations had poor agreement with airborne Tbs, while 19 GHz V exhibited a low RMSE (6.15 K), MBE (−4.52 K) and improved relative agreement to airborne measurements (R = 0.47). Salinity considerations reduced 6.9 GHz errors substantially, with a drop in RMSE from 51.48 K and 57.18 K for H and V polarizations respectively, to 26.2 K and 31.6 K, although Tb variability was not well simulated. With best results at 37 GHz, HUT simulations exhibit the potential to track Tb evolution, and therefore SWE through the winter season.

Multiple-Layer Adaptation of HUT Snow Emission Model: Comparison With Experimental Data

Modeling of snow emission at microwave frequencies is necessary in order to understand the complex relations between the emitted brightness temperature and snowpack characteristics such as density, grain size, moisture content, and vertical structure. Several empirical, semiempirical, and purely theoretical models for the prediction of snow emission properties have been developed in recent years. In this paper, we investigate the capability of one such model to simulate snow emission during the peak snow season-a new multilayer version of the Helsinki University of Technology (HUT) snow model. Developed with a single layer, the original HUT model was easily applied over large geographic areas for the estimation of snow cover characteristics by model inversion. A single homogenous layer, however, may not accurately allow the simulation of vertically structured natural snowpacks. The new modification to the model allows the simulation of emission from a snowpack with several snow or ice layers, with the individual component layers treated as in the original HUT model. The results of modeled snowpack emission, using both the original model and the new multilayer modification, are compared with reference measurements made using ground-based radiometers deployed in Finland and Canada. Detailed <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> measurements of the snowpack are used to set the model inputs. We show that, in most cases, use of the multiple-layer model improves estimates for the higher frequencies tested, with up to 38% improvement in rms error. In some cases, however, the use of the multiple-layer model weakens model performance particularly at lower frequencies.

Inversion of a Snow Emission Model Calibrated WithIn SituData for Snow Water Equivalent Monitoring

Snow water equivalent (SWE) is critical information for hydrological studies over northern latitudes. This information is needed to adequately simulate spring runoff in order to optimize hydropower generation. In this paper, a new geophysical inversion methodology for monitoring SWE evolution on a daily basis is presented. The methodology is based on a modified version of the Helsinki University of Technology (HUT) snow emission model and assimilates <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> data as ground truth. Previous studies used the mean snow grain size (?) as an optimization parameter to retrieve SWE with the HUT model. Therefore, <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a priori</i> information on ? was needed to estimate SWE, but this information is rarely measured in the field. In the HUT model, the mean snow grain size is used to compute the total extinction coefficient (? <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</i> ) of the snowpack. We modified the model to eliminate the required <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a priori</i> knowledge on mean snow grain size and used the total extinction coefficient as an optimization parameter instead. The methodology was applied in a subarctic territory in northern QuE?bec (Canada) using snow measurements acquired during the 2001 to 2006 winters (January 1 to March 31). A metric based on the brightness temperature difference between the special sensor microwave/imager 19- and 37-GHz (vertical polarization) was used for SWE retrieval. In the monitoring process, the result obtained from the previous day inversion was used to initialize the current day SWE retrieval. When <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> measurements are available, they are assimilated and used to calibrate the model. The results obtained with this methodology show an improvement in the experimental context considered (high SWE values). For all winters considered in this paper, the results described in this paper show that SWE was estimated with a root-mean-square error (rmse) of less than 19% after the first process of <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> data assimilation and with an rmse of 14% for <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nondense</i> forested areas and for SWE values below 300 mm.

A comparative study of Chang and HUT models for UK snow depth retrieval

For the development of passive microwave remote sensing techniques, brightness temperature information on the medium covering the Earth's surface under different conditions is required. An emission model is a useful tool for the estimation of the brightness temperature of the medium. If the medium is a snow pack, the microwave radiation emitted will depend on the physical temperature, crystal characteristics, stratification and density of the snow. The parameters of microwave emission models available for the retrieval of snow characteristics are highly dependent on local environmental and climatological conditions. The aim of this study was to compare the empirical Chang model with the semiempirical radiative transfer model of snow developed at Helsinki University of Technology (HUT) for snow depth (SD) retrieval for UK snow packs. In the first step we used the HUT model. The root mean square error (RMSE) was used to validate the accuracy of model estimates. The snow events from different days in 1995, 1996 and 1997 were used in this study. In the second step a revised form of the Chang model, which was originally calibrated for global snow monitoring, was applied to estimate the SD. It is evident from this study that the Chang model underestimates the SD whereas the HUT model both underestimates and overestimates the SD for UK snow cover. This study also demonstrates that the application of algorithms for snow pack monitoring requires local calibration for effective and reasonable results.

Observed and modelled effects of ice lens formation on passive microwave brightness temperatures over snow covered tundra

Immediately before an April 2007 snow survey and passive microwave radiometer field campaign in the Northwest Territories, Canada, a rain-on-snow event deposited a thin (~ 3 mm) continuous layer of ice on the surface of the snowpack. At eight sites the brightness temperature ( T b) of the undisturbed snow pack was measured with a multi-frequency dual polarization (6.9, 19, 37, and 89 GHz) ground based radiometer system. The ice lens was then carefully removed and the T bs were measured again. The individual V-pol channels and the 37 V − 19 V difference were largely unaffected by the presence of the ice lens, exhibiting a systematic shift of about 3 K. In comparison, the ice lens had a considerable effect on the H-pol T b at all frequencies, with a mean difference (ice lens present − ice lens removed) of − 9 K (± 5.3 K) at 6.9 GHz, − 40 K (± 11.3 K) at 19 GHz, − 33 K (± 7.6 K) at 37 GHz, and − 19 K (± 8.0 K) at 89 GHz. The effect of the ice lens on H-pol measurements was also observed with spaceborne data from the Advanced Microwave Scanning Radiometer (AMSR-E) satellite data. Simulations of T b were produced for each site using a new two layer formulation of the Helsinki University of Technology (HUT) snow emission model. The ice lens was used as the top layer and the underlying snowpack considered as a homogenous second layer. The agreement between observations and simulations was variable, with agreement strongest at 19 GHz. A comparison with simulations produced using the Microwave Emission Model of Layered Snowpacks (MEMLS) suggests HUT model uncertainty is related not to the ice lens, but to difficulties in simulating emission from deep snow. Overall, the observations and simulations suggest H-pol measurements are capable of detecting new ice layers across the tundra snowpack, while V-pol measurements are more appropriate for snow water equivalent (SWE) retrievals due to their relative insensitivity to ice layers.

Estimation of snow depth in the UK using the HUT snow emission model

Snow cover characteristics have significant effects on upwelling naturally emitted microwave radiation through processes of forward scattering. This study simulates numerically the electromagnetic responses from snow in the UK using the radiative transfer‐based semiempirical model developed at the Helsinki University of Technology (HUT), which takes into account the influence of soil surface, forest canopy and atmosphere on space‐borne observed brightness temperature by using empirical and semiempirical formulas. A sensitivity analysis of the HUT model was conducted to determine the most sensitive parameter affecting upwelling radiation from snow in the UK. The model‐based results were compared with observed Special Sensor Microwave Imager (SSM/I) brightness temperatures to better understand the SSM/I response to snow. The available ensemble of data required for input to the HUT model comprise surface physical temperature, ground level pressure and water vapour content, forest stem volume and land cover water fraction. Based on the sensitivity analyses, numerical parameters representing physical snow pack quantities (e.g. snow grain size, snow moisture and snow depth (SD)) were varied and the method of root mean square error (RMSE) minimization was used to invert the SD. The HUT model was applied to different days in 3 months (23–31 January, 1–5 and 26–27 February and 1–7 March 1995) of records of daily SD and SSM/I observations. The results show that the HUT model both underestimates and overestimates SD prediction. For the month of January 1995, the HUT model underestimated SD with a bias of −0.59 cm, whereas for February and March 1995 the HUT model overestimated the SD with a bias of 1.89 cm and 1.64 cm, respectively. This study demonstrates that microwave remote sensing of snow can be used successfully in the UK, where most research on snow cover is conducted by using a visible and infrared radiometer. It is also evident from this work that application of algorithms to snow pack monitoring needs local calibration for effective and reasonable results.

Snow water equivalent retrieval in a Canadian boreal environment from microwave measurements using the HUT snow emission model

Snow water equivalent (SWE) is a critical parameter for climatological and hydrological studies over northern high-latitude areas. In this paper, we study the usability of the Helsinki University of Technology (HUT) snow emission model for the estimation of SWE in a Canadian boreal forest environment. The experimental data (airborne passive microwave and ground-based data) were acquired during the Boreal Ecosystem-Atmosphere Study winter field campaign held in February 1994 in Central Canada. Using the experimental dataset, surface brightness temperatures at 18 and 37 GHz (vertical polarization) were simulated with the HUT snow emission model and compared to those acquired by the airborne sensors. The results showed an important underestimation at 37 GHz (-27 K) and an overestimation at 18 GHz (10 K). In this paper, we demonstrate that the errors in the model simulations are due mainly to the extinction coefficient modeling, which is a function of snow grain size. Therefore, we propose a new semiempirical function for the extinction coefficient, based on an empirical correction to the Rayleigh scattering expression. Results presented in this paper show that the proposed function improves the HUT model accuracy to predict brightness temperature in the experimental context considered, with a mean error of /spl plusmn/5 K and /spl plusmn/9 K, respectively, at 18 and 37 GHz, and a negligible bias (less than 4 K) in both cases. These errors are comparable in magnitude to the accuracy of the radiometers used during the airborne flights. SWE was retrieved using the modified HUT snow emission model based on an iterative inversion technique. SWE was estimated with a mean error of /spl plusmn/10 mm and a negligible bias. Only a rough knowledge of mean snow grain size /spl phi/~ was required in the inversion procedure. The effects of possible errors on mean snow grain size /spl phi/~ are presented and discussed.