| Poyang lake is China’s important wetland ecological system, to adjust the basin of water and sediment change plays an important role, but in recent years due to the surrounding the lake district economic development, environmental problems become more and more prominent, threatens people’s health and the sustainable development of environment and economy. So the effective of poyang lake’s ecological protection and monitoring dynamic change of lake water and sediment has important practical significance.Based on poyang lake waters as the research area, using MODIS image data of nearly 13 years, every month, with its proper NDVI threshold, the extraction of the poyang lake waters.and using HJ- CCD image of higher spatial resolution by using the decision tree model for nearly five years of poyang lake water extraction, the results at the same time different resolution remote sensing image to extract the difference, the results show that the extraction results are basically identical. And then through the poyang lake waters range of time and space dynamic change analysis showed that: no matter the poyang lake waters of the dynamic changes of the interannual or years range are big, in general, the lake water level rising from the start of April of each year, 8 ~ 9 months at a high level, as the flood season; October levels started to slow down, the water area gradually narrowed, the discontinuous. The hukou water level of poyang lake water area and the relationship model, a polynomial model had better fitting effect, R2 value of 0.8727.Using three voyage poyang lake field measured spectral data, suspended sediment concentration data and the experience of the unclassified all water inversion model is set up. Then according to the measured spectral curve features of poyang lake water body can be divided into special turbidity, medium turbidity, mild turbidity, water area, particularly clear and so on 5 kinds of water body type, and then according to this principle based on the slope of poyang lake water classification algorithm, is applied to classify remote sensing image; After the classification of all kinds of water body, respectively, set up their proper inversion model inversion, find the optimal suspended sediment inversion model of each type of water type is the ratio of each water type model of quadratic function model. Finally based on the water absorption and scattering of an analysis of the suspended sediment and half of the inherent optical inversion model, comparison results show that the three kinds of inversion model based on the experience of the inversion model of water classification effect is best, an average absolute error is 0.00217 g/L, the average relative error was 3.022%.Poyang lake water classification algorithm was applied to 20 years(1996-2012) of the remote sensing image for all kinds of time and space dynamic change of water body type, according to the results of the poyang lake water is increasingly is cloudy; Then suspended sediment inversion effect is better based on the experience of the classification model is applied to remote sensing images in 2013 and 2014, the results show that the poyang lake sediment concentration present in and out of the river channel and pine door mountains north of the suspended sediment concentration the highest near the water, up to 0.509988 g/L; Five rivers into the lake near suspended sediment is also higher; But the center of the poyang lake district and lake area in the northeast of the suspended sediment concentration is low, the lowest 0.001333 g/L; The basic trend gradually reduce from north to south, from west to east.Finally, explore the poyang lake water-sediment space-time change mechanism of poyang lake mainly affected by meteorological hydrological features and other natural factors and the lake reclamation, water loss and soil erosion caused by human factors and human factors such as construction of water project facilities. |
PDF Full Text Request