Jouni Pulliainen

Detecting the onset of snow-melt using SSM/I data and the self-organizing map

In this paper, we present an algorithm to estimate the onset of seasonal snow‐melt using space‐borne microwave radiometer data. We have earlier developed a simple model called a Channel Difference Algorithm (CDA) to estimate the beginning of the snow‐melt. The new algorithm, the SOM Detection Algorithm (SDA), is based on the use of an artificial neural network system a called Self‐Organizing Map (SOM). The purpose of this research is to develop a robust and simple algorithm feasible for operative use. The algorithm is tested using SSM/I data with hydrological predictions as reference data. The reference data covers two winters, 1997 and 1998, and is for the boreal forest zone in Finland. The results are promising. The SDA is able to estimate the beginning of the final snow‐melt well, especially if the snow water equivalent exhibits large values. Using low‐pass filtering for the SDA estimated time series, the estimation can be improved.

Transmissivity of boreal forest canopies for microwave radiometry of snow

The HUTRAD airborne microwave radiometer (Helsinki University of Technology RADiometer) was used to collect data on microwave emission from snow-covered terrain, including a variety of forest types and land-cover categories in two test sites: the Tuusula test site in southern Finland, and the Syote test site in northern Finland. HUTRAD operates at 6.8, 10.65, 18.7, 23.8, 36.5, and 94 GHz, vertical and horizontal polarization, using a look angle of 50/spl deg/ off nadir. Forest information was used to determine the effect of forest canopies to the brightness temperature of snow-covered terrain. The results are expressed as the forest transmissivity vs. stem volume and observation frequency.

Monitoring of turbid coastal and inland waters by airborne imaging spectrometer AISA

Reliable monitoring of the pelagic ecosystem has proved to be problematic because of its high temporal and spatial heterogeneity. Processes like algal blooms or pollution discharges are patchy, both temporally and spatially. Consequently, they often remain unobserved using the traditional sampling methods based on temporally sparse sampling at a few fixed stations. Furthermore, the traditional programs are usually unable to rapidly report of exceptional events. Monitoring of water quality could be more effective if satellite or airborne remote sensing is used. New optical satellite sensors with high spectral resolution have recently been launched and more sensors will be available in the near future. Empirical algorithms, like spectral ratios, are widely used in the interpretation of remote sensing data. However, these algorithms seem to have local and seasonal variability and different algorithms are needed for coastal and inland waters.

Retrieval of forest parameters from multitemporal spaceborne SAR data

The aim of the study is to develop inversion methods for spaceborne SAR images of the boreal forests. The parameters of interest are stem volume (biomass), forest and land-cover types, soil and vegetation moisture. The previously developed forest stem volume/soil moisture retrieval algorithm was tested by producing soil moisture maps for large test areas. The textural information of a seasonal set of ERS-1 SAR images was studied with the first and second order statistical measures. These measures had a higher information value than intensity values (or their principal components) for the forest and land type classification. The multitemporal approach was beneficial for the application of textural measures and the textural parameters significantly improved the classification of land-use and forest types.

Estimation of snow covered area by applying apparent regional transmissivity

The detection of snow from optical instruments is often hampered by forest canopy. In this paper, an empirical reflectance model for estimating regional values for snow covered area (SCA) from optical data is presented. In the model, SCA is expressed as a function of apparent vegetation transmissivity. The estimation of SCA has been tested for NOAA/AVHRR data with drainage basins as calculation units. The same areas are used in an operative hydrological model. Comparison of estimated SCA with reference data indicates good correlation.

Modeling Snow Volume Backscatter Combining the Radiative Transfer Theory and the Discrete Dipole Approximation

A new method is developed to model the volume backscattering from dry snow. The model is a combination of the exact field approach and zeroth order vector radiative transfer theory. The field approach is used to define the scattering characteristics in a single almost indefinite small snow volume unit and the calculation is realized with discrete dipole approximation (DDA). The radiative transfer theory (RT) is utilized by defining a homogenous snow layer from the averaged scattering characteristics and combining different layers together forming a vertical structure of a snow pack. Because of the DDA the presented model takes into account all multiple reflections and all polarizations inside the snow volume. The scattering amplitude values calculated with the DDA method and according to Mie theory are compared to together. Some results for a homogeneous snow layer are presented for both cubical and needle shaped snow grains.

Classification and retrieval of dry snow parameters by means of SMM/I data and artificial neural networks

Dry snow temperature, snow water equivalent (SWE) and snow depth have been retrieved by using the 19 and 37 GHz SSM/I brightness temperatures and artificial neural networks (ANNs). The results obtained have been compared with those obtained using other approaches such as the spectral polarization difference, the HUT model-based iterative inversion, the Chang algorithm and linear regressions. In general, it has been noted that the ANN based technique gives better results than the other approaches, which tend to underestimate the unknown parameters.

Effective permittivity of wet snow by using two-phase strong fluctuation theory with non-symmetrical inclusions

The strong fluctuation theory is applied to calculate the effective permittivity of wet snow by two-phase model with non-symmetrical inclusions. In the two-phase model, wet snow is assumed to consist of dry snow (host) and liquid water (inclusions). Numerical results for the effective permittivity of wet snow are illustrated for random media with isotropic and anisotropic correlation functions. The three-phase strong fluctuation theory model with symmetrical inclusions is also presented for theoretical comparison. In the three-phase model, wet snow is assumed to consist of air (host), ice (inclusions) and water (inclusions) and the shape of the inclusions are spherical. The results are compared with the Debye-like semi-empirical model and a comparison with experimental data at 6, 18 and 37 GHz is also presented. The results indicate that (a) the shape and the size of inclusions are important, and (b) the two-phase model with non-symmetrical inclusions provides the good results to the effective permittivity of wet snow.

The HUT brightness temperature model for snow-covered terrain

The Helsinki University of Technology (HUT) semiempirical model to describe microwave emission from snow-covered terrain is discussed. The model treats the snowpack as a single layer above the ground and it takes into account the contributions from the ground, snowpack, forest canopy, and atmosphere.

Comparison of ranging scatterometer and ERS-1 SAR microwave signatures over boreal forest zone during winter season

A set of ERS-1 SAR images along with airborne non-imaging ranging scatterometer (HUTSCAT) measurements and in-situ surveys has been obtained from the Sodankyla test site in Northern Finland. A total of 5 measurement campaigns have been organized during the years 1991 to 1993. 19 test lines have been selected from the test site to represent different land-use classes. Microwave signatures representing the test lines have been extracted from ERS-1 SAR images and HUTSCAT measurements. The behaviour of these signatures has been compared to each other and with boreal forest semi-empirical backscattering model. The results indicate that the behaviour of ERS-1 SAR derived microwave signatures are similar to those of HUTSCAT even in the presence of forest canopies. By using the semi-empirical model the ERS backscattering is divided into two contributions 1) contribution from top of the canopy and 2) contribution from ground, trunks and ground-canopy reflections. This can be used for analyzing forest backscattering mechanism under various conditions.