Home – Home – Jornals – Atmospheric and Oceanic Optics 2014 number 7

The article is about the pioneer atmospheric-optical space researches and conservative systems of remote sensing of the near-earth space on the basis of the Earth radiation characteristics. Space researches are such area of fundamental and applied works, which from the first steps of the formation could not develop without the use of computers. Space exploration was a significant factor in the improvement of the computers and the formation of new scientific trends connected with mathematical modeling of the Earth radiation field, the theory of image transfer, theory of vision, theory of processing and recognition of images, etc. Information and mathematical software is an indispensable part of any space project. The short history of the nuclear and space projects and formation of scientific schools on the theory of radiation transfer is presented to restore the true facts of the beginning of opening of the space era and development of the cosmonautics. Our goal is to show as the ingenious Chief theoretician-mathematician V.M. Keldysh and Chief engineer-constructor S.P. Korolev together conquered space. It is important to pay attention to the prospective scientific directions connected with the application of “computer science" with parallel supercalculations, GRID and “cloudy" technologies in varied space researches, the problems of the Earth remote sensing, the evolution of environment, and climate of the Earth.

This paper considers methods of estimating the state of various vegetation including agricultural from remote sensing data (satellite monitoring). This paper discusses the possibility of using the analysis of vegetation indices (VI) deviations from “normal” time variation based method for specific objects and regions to estimate the state of vegetation. The problems of constructing the “normal” VI time series for different regions are discussed. Special attention is paid to possibilities of using the “VEGA” satellite service created in IKI for research of Northern Eurasia vegetation to implement the proposed approach.

The application of vegetation indices for the quantitative evaluation of basic pasture parameters (grass-cover, productivity, unpalatable grass, pasture degradation) is discussed for low and moderate resolution optical sensors. Each vegetation index was correlated with ground-truth data. The algorithm of pasture condition estimation was developed using highly correlated indices.

Main stages of development of technologies for natural and anthropogenic objects recognition (cognitive technologies for optical images processing) using remote sensing data are considered together with computational procedures of atmospheric correction of multispectral and hyperspectral airspace images. Main attention is paid to recognition of forest ecosystems of various species and age, based on in-flight testing of a domestic imaging spectrometer for a selected test area, where ground-based forest inventory and other observations were carried out. High accuracies of the recognition of separate gradations of ages for the selected pure birch and pine species are revealed, using the elaborated softwave for airborne hyperspectral image processing.

The Bayes decision rule for recognition of hyperspectral aerospace data is obtained. An example of recognition of forest stand types is given.

The angular and polarization dependencies of the brightness temperature measured by the spaceborne radiometer SMOS in L-band on different parts of the earth's surface are shown. It is found that in some cases the results are not described by the existing model for the reflection coefficient of microwaves, which leads to errors in recovering the soil moisture. It is shown by calculation that the layered structure of soil moisture has a significant impact on the behavior of brightness temperature. It is proposed to expand the moisture-recovering algorithm by the method which takes into account possible layered structure of the soil moisture.

This report presents the results of a study of micro-deformation of soil and landslides on the Yamal Peninsula by the methods of differential interferometry and analysis of seasonal and long-term changes in the texture of radar images.

The technique of space infrared and microwave atmospheric sounding with ATOVS instrument is discussed. The results of the sounding are vertical profiles of atmosphere. Zenith tropospheric delay was estimated with the profiles and with the Saastamoinen and Hopfield delay models using surface measurements. Biases of the model-derived estimations are -0,0186 m for the Saastamoinen model and -0,0135 m for the Hopfield one during the whole 2012. The results are valid for Krasnoyarsk (middle latitudes of Siberia).

A software package for reconstruction of the distribution of the Earth surface reflection coefficient is described together with procedures developed to decrease considerably the computation time. On an example of a test region in the western coast of Africa, the results obtained by the suggested algorithm, algorithm of homogeneous correction, and standard MOD09 NASA algorithm are compared. The coefficient of correlation of the results obtained for the given test region using the new algorithm and the algorithm of homogeneous correction is 0.999; it is 0.984 for the given algorithm and the MOD09 algorithm.

A description is presented of software for a computing system of satellite data correction that provides calculations in automated mode with the use of distributed data. Two types of data: predictive models and satellite measurements are used as input data on the atmospheric state. The method is developed for assessment of the relevance of meteorological satellite data, which allows estimation of the reliability of calculated results. The system provides correction for the distorting effect of the atmosphere for a wide variety of satellite sensors operating under diverse atmospheric meteorological conditions and combination of various distorting factors.

Aerosol optical thickness and aerosol index variations caused by large-scale forest fires in Central Yakutia in 2002 were studied by remote sensing data (NOAA, Terra, Earth Probe). Total emissions of CO₂, CO, CH₄ etc. were calculated using modeling approach and MODIS/Terra burned area data. A comparison of total emissions from forest fires in Central Yakutia with global fire and volcanic emissions are made. Long-range transport events of aerosol particles during maximum activity of forest fires are examined.

A set of informative textural features is formed for various cirrus cloud patterns presented in MODIS satellite images with 250-m resolution. To describe the texture, the following methods are applied: Gray-Level Co-occurrences Matrix, Gray-Level Difference Vector, and Sum and Difference Histograms. Laws are determined that describe textural features’ fluctuations, and assessments of their parameters are presented as well. Results of cirrus cloud subtype classification using neural network technologies are presented and discussed.

A comparative analysis is carried out of results on changes of bioindicative tree state in the area of air pollution as a result of gas flaring at Priobskoe oil field and estimates derived from vegetation indices determined from satellite images. Acceptable matching of results of remote assessment of forest changes with data of bioindicative ground research in the taiga zone is shown.

The main directions of the ecological control of the territory from the space are the development of methods of remote sensing of the parameters of Earth surface and atmosphere and monitoring of the state of the environment. Monitoring data are necessary for investigation of natural processes and analysis of impacts of natural and anthropogenic factors on the natural environment. For receiving, processing, and analysis of data of the satellite remote monitoring, it is necessary to create appropriate software. The article considers a program-analytical complex based on MODIS satellite data allowing counting vegetation indexes and analysis of their behavior. On the basis of statistical data, the dependence of natural variations in the indexes is determined that allows detecting changes in the variations due to anthropogenic impacts.