Research On The Mechanism Of Degraded Grassland In Western Jilin Province

The natural grassland of western Jilin Province locates in the south-west of Songnen Plain. Over the last 50 years, the impact of global climate change and human activities has dramatically deteriorated the local ecological environment. And it has brought great damage to the human living environment and sustainable development of the social economy. In addition, the ecosystem restoration and grassland environment improvement of western Jilin Province become an important issue.This thesis takes degraded grassland of western Jilin Province as the research object. Support vector machine was used to simulate the non-linear relationship between soil properties and ecological stoichiometry indexes of Leymus Chinensis. Meanwhile, matter-element extension model was used to evaluate the degeneration of grassland.With the aim of exploring the degrading succession mechanism based on the theory of ecological stoichiometry in western Jilin Province, the results indicate that:1. Building basic model of support vector machine Leymus chinensis carbon, nitrogen and phosphorus content prediction model. C, N, P content of Leymus chinensis prediction model was set by using support vector machine. The average relative error of C, N and P prediction outcomes is 0.15%, 4.97% and 3.07%, respectively. The model is of fast convergence speed and higher prediction precision, and it can be used to predict chinensis carbon, nitrogen and phosphorus content. The statistical results show that the average content of carbon, nitrogen and phosphorus of Leymus chinensis in western Jilin Province are 452.85, 20.45 and 1.40 g/kg, respectively, which are all lower than the global average contents of terrestrial plants. And the spatial variability of Leymus chinensis phosphorus content is significantly higher than carbon and nitrogen. The values of C/N, C/P and N/P are all higher than those of global terrestrial plants.2. Soil nutrients, soil alkalization parameters and Leymus chinensis ecological stoichiometry indexes were selected as the evaluation indexes. And matter-element extension method was used to a build comprehensive assessment model by taking soil properties and plant properties as a combination. The results show that grassland degradation levels mainly concentrate in three and four level, first level accounts for 1.94%, second level only accounts for 2.91%, three level accounts for 64.08%, four level accounts for 28.16%, five level accounts for 2.91%, which indicates that grassland of western Jilin Province is under severe degradation.3. With the aggravating of grassland, soil major nutrients C, N, P and main available nutrients decreased obviously, and the degree of soil salinity-alkalinity aggravate obviously, Leymus chinensis N and P content generally shows a decreasing trend. However, Leymus chinensis C has no significant trend. The process of the succession of grassland degradation in western Jilin province is along with main nutrients and soil available nutrients loss, soil salinity-alkalinity aggravation and the loss of Leymus chinensis N, P.4. With the aggravating of grassland, Leymus chinensis C/N and C/P shows an increasing trend, and N/P has no significant trend, soil C/P and N/P decreased obviously, C/N has no significant trend. N/P in first,second and five level grassland less than 14, which means Leymus chinensis growing is mainly limited by N element. N/P in three and four level grassland ranges between 14 and 16, which means Leymus chinensis growing is mainly limited by N and P.5. During the succession process of western Jilin Province grassland ecosystem, soil major nutrients and main available nutrients decreased obviously, soil properties appears to be barren andsalinized, Leymus chinensis growing is overall limited by N and P. With the photosynthesis intensity reduction and productivity declination,the Leymus chinensis community eventually succeeds to mixed community of Chloris, puccinellia tenuifolra and suaeda salsa.