- Tytuł:
-
Klasyfikacja gleb słonych doliny Czuj w Kirgistanie na podstawie wielospektralnych obrazów satelitarnych Landsat TM, Landsat ETM+, TERRA ASTER oraz danych naziemnych
Classification of salt-affected soils of the Chuy Valley in Kyrgyzstan using multispectral satellite Landsat TM, Landsat ETM+, TERRA ASTER images and ground-collected data - Autorzy:
- Kokoeva, G.
- Powiązania:
- https://bibliotekanauki.pl/articles/132209.pdf
- Data publikacji:
- 2007
- Wydawca:
- Polskie Towarzystwo Geograficzne
- Tematy:
-
gleby słone
klasyfikacja
obraz satelitarny
satelita Landsat
satelita TERRA ASTER
dane naziemne
Kirgistan
salt-affected soils
classification
satellite Landsat
satellite TERRA ASTER
satellite image
ground-collected data
Kyrgyzstan - Opis:
- The natural conditions of Kyrgyzstan and consequences of human-induced processes, such as inappropriate methods of irrigation, have led to the extension of salt-affected soils. Extensive areas of irrigated land have been increasingly degraded by salinization from over-irrigation and other forms of inadequate agricultural practices. Between 1985 and 1990, the area of salt-affected soils increased from 666 300 ha to 1170 300 ha (Mamytov, 1995). In recent years salinity processes have been described as one of the problems of agriculture in that area. For the last ten years many none-affected soils of the Chuy Valley have become salinized. According to Mamytov et al. (1991) the total area of salt-affected soils in the Chuy Valley exceeds 259.5 thousands ha, which is more than 42% of the research area. In this research, an attempt has been made to estimate soil salinity quantitatively and also spatially by applying remote sensing techniques. Conventional methods of mapping salt-affected soils consume a lot of energy, time and money. Remote sensing enables us to detect and to map salt-affected soils by using relatively cheap multispectral satellite data such as Landsat TM, Landsat ETM+ and TERRA ASTER. The objective of this study is to identify salt-affected soils by integrating satellite images with ground-collected data. In order to achieve this goal the best algorithms of an unsupervised and a supervised classifi cation have been chosen using TNTmips software. The Normalized Difference Vegetation Index (NDVI) and the Transformed Vegetation Index (TVI) have been applied to distinguish densely and partly vegetation- covered soils, which are not salt-affected. To distinguish areas covered with stone and sands from saline soils the Salinity Index (SI) has been applied. For the differentiation of arable land which is not covered with vegetation the brightness parameter of Tasseled Cap transformation has been used. All these indices were calculated from satellite images. Finding an appropriate interpretation scheme for identifying the saltaffected soils of the Chuy Valley becomes a very important factor infl uencing the accuracy of the supervised classifi cation. The temporal change of salinity accumulation is demonstrated by comparing the classifi cation’s results of the multispectral satellite images from 1994 to those of 2001. This study also includes measurements of spectral properties of collected soil samples for better understanding the difference in classifi cation accuracy of various types of salt-affected soils. Spectral refl ectance was registered from the surfaces of saline and saline-sodic soils using fi eld luminancemeter CIMEL CE 313-21 in the following wavelength bands: 450 nm, 550 nm, 650 nm and 850 nm.
- Źródło:
-
Teledetekcja Środowiska; 2007, 37; 3-50
1644-6380 - Pojawia się w:
- Teledetekcja Środowiska
- Dostawca treści:
- Biblioteka Nauki
Artykuł