| With the rapid development of social economy, the contradiction between the growing population and limited resources is increasingly deepened.As a developing country with a great amount of population, the big problem is showed between people and limited lands.The strategic decision, the protection of cultivated land resource, has always been strict enforced by the Party Central Committee and the State Council. The explanation of the cultivated land protection should be interpreted from two aspects:On the one hand, from quantity, we should protect the arable land minimum, On the other hand, we should keep the steady quality of cultivated land. Nowadays, a series of strong national and local implementation for cultivated land protection policies, have made to make sure a dynamic balance for the number of farmland in our country. Furthermore, the department of lands and resources attaches a great importance on protecting the quality of cultivated land at present.Knowing clearly about the quality of the cultivated lands, can effectively avoid the situation, "compensating poor land for the rich one", and achieve the requisition-compensation balance in the quantity and quality for the cultivated land. Meanwhile, knowing well about the quality of the cultivated lands can benefit on making the rules about the land and on providing a scientific basis for demarcating the farmland. Nowadays, many scholars at home and abroad have taken the focus from the dynamic change in the quantity of cultivated land to the quality monitoring research.The black soil area of Songliao plain,one of the world three big black soil zone, lies in the northeast of China, mainly is distributed in Heilongjiang and Jilin provinces, and a small amount of which is distributed in the Inner Mongolia Autonomous Region and Northern part of Liaoning province. The quality of cultivated lands in black soil area, the major grain production area in China, directly affects the food security of our country.However, It is urgent to have the protection of black soil for the following reasons. There is a serious soil degradation, after hundreds of years reclamation.Under the overexploitation and the natural factors, a lot of typical black soil region has lost their own characteristics. Organic matter content of some black soil of this region has dropped to below 2%.With the fast development of remote sensing and computer technology, space information technology has spread to all trades of life. Soil remote sensing has become the new direction of Soil Science.The increasingly mature technologies of multi-spectral imaging and hyperspectral imaging take a great number of soil science researchers from the traditional soil investigation with the limit of time and space to the indoor, rapid, and efficient study for the extraction and analysis of soil physical and chemical information.Soil remote sensing is a new technology based on physical basis about the soil spectral reflection characteristic.A large number of studies have shown that there is a significant relationship between physical and chemical properties of the soil and soil spectral reflectance characteristicsSoil organic matter content is an important index of soil fertility, and also a standard for the quality of black soil.This paper shows that we established the relationship model between the typical black soil organic matter content and the spectral reflectance in remote sensing image, so as to achieve the inversion estimation for the black soil organic matter content on regional scale, and achieve the monitoring about the dynamic change of soil organic matter content on time in nearly a decade in this research area. After collecting the soil sample on field, on the basis of test results related to the soil organic matter content and related soil composition, with multiple source for remote sensing image data.This research attains the following achievements:1.The study summarized the spectral features for black soil in the data of Landsat 8 OLI, starting with soil spectral characteristics in the sampling point, comparing spectral information for soil sample in the field with it in the spectral characteristics in remote sensing image data. Through comparative analysis, it is believed that the band for Landsat-8 OLI being sensitive to soil organic matter content of are 2 band (0.525-0.600 microns),3 band (0.630-0.630 microns) and 4 (0.845-0.885 microns).The control points of the spectral characteristics play a great role for analyzing soil organic matter identification.2.The study concluded that the influence of water content on soil reflectance in the image, aiming at the shortage of samplings in the study in 2006, through the experimental analysis of soil water content in the soil sampling points. Black soil spectral reflectance values are carried through mathematical transformation. We tried to use the mathematical change of the reflection spectrum to remove the influence of soil water content. It is found that the commonly used soil reflectance differential technology in the study of hyperspectral is applied to the spectral values in the Landsat 8 OLI. We combined the moisture content and moisture of the single band correlation analysis results, in the sampling point Around it is dealt in first order differential treatment in the soil spectral reflectance, soil moisture content and the correlation of each band is changed from highly related to low related. It is considered that some parts of soil moisture affect is removed3. It is not a simple linear problems to use the spectral reflectance of soil organic matter in soil in the image inversion, since its high spatial resolution of the multispectral image, and the spectral characteristics are the results of comprehensive functions in many aspects. After comparing support vector machine (SVM) regression algorithm, with the traditional linear regression algorithm, we proposed the inversion model for the soil organic matter content based on support vector machine (SVM), and improved the kernel function of the model. On the evaluation of the model, the improved regression model raised the inversion precision of the content of soil organic matter. It was also improved the efficiency of the inversion.4. Soil organic matter content for the whole typical black soil is inverted after the model is applied to Landsat 8 OLI in the remote sensing image. Remote sensing interpretation is performed to the study area combining interpretation results and the inversion, before the inversion results are classified. The soil organic matter content is divided into five types in the study of "pure" black soil in the area and we displayed the typical spatial distribution of the farmland soil organic matter content in the black soil zone status visually. Through the comparison, black soil organic matter content distribution presented that northern organic matter content is higher than the trend of the organic matter content in the southern part. The distribution is like this for the reason that on the one hand, low dimension region inhibits the activity of microorganisms at low temperature for a long time, which declines its ability to decomposition of soil organic matter. On the other hand, the different for land reclamation and the degree of development on each region, and the cultivation of crops type.5.Based on field sampling data in 2006, taking the same research methods in Landsat 8 OLI images, we analyzed the spectral characteristics of sample in the Landsat TM data, and compared it with data of OLI, and used TM data to interpret 2006 typical black soil zone, and to make the inversion classification for soil organic matter content, and the classified result was compared with the one in 2014.Through the comparison, we found that typical black soil organic matter content is changed little in recent 10 years generally, and only a slightly change in local area.From the value of organic matter content in the inversion, most of the content of soil organic matter tends to slow down, which proves that the typical black soil region is in a relatively stable state.6. This paper tried to have a quantitative research of black soil composition, and took the results to the land consolidation project of replenish arable land quality evaluation and replenish arable land in the area is carried with a class assessment.The assessment and the surrounding farmland showed no discrepancy with the old class assessment. It is showed that the Landsat-8 OLI data is available to cultivated land quality assessment in the typical black soil region in northeast China and the research results have certain application. |
PDF Full Text Request