Correlation Analysis On Soil Nutrient And Plant Community During Ecological Reclamation In Opencast Coal Mine Of Semiarid Areas

In this paper, a0.8hm2permanent plot (Robinia pseudoacacia＋Pinus tabuliformis) was established in Pingshuo opencast coal mine dump. Further, we took the restored vegetation and soil factors as our research objects, applied the R programming language, and centered on the spacial-temporal pattern changes of the species composition during the process of restored vegetation, the inter-specific association of the plant, the relationship between fundamental characteristics of the population and the soil factors to explore the correlation between soil nutrient and plant community during the process of vegetation restoration of opencast coal mine dump in semiarid areas. The main results were as follows:There were2501survival individuals in the plot, among which Robinia pseudoacacia and Pinus tabuliformis accounted for63.7%of the total number of species, Ulmus pumila35.5%. The number of regeneration individuals of Robinia pseudoacacia was less than that of Ulmus pumila, and the regeneration of Pinus tabuliformis was poor. The statistics of breast-height basal area and the important value showed that Robinia pseudoacacia and Pinus tabuliformis were obviously dominant species. The size-class distribution of all individuals showed an invert J-shape, which indicated that the community was in a stable status. Based on their size-class frequencies, the size-class distribution of the dominant species was classified into three types:1) invert J-shape represented Robinia pseudoacacia;2) unimodal shape, Pinus tabuliformis;3) and, L-shape, Ulmus pumila. The analysis of spacial pattern showed that the individuals with small size-class of Robinia pseudoacacia and Ulmus pumila were obviously gathered in the lower slope position of the plot. The big and small size class of Robinia pseudoacacia was complementary in spatial pattern. Comparing with the initial reclamation, the spacial pattern of Pinus tabuliformis almost had no changes, and it was evenly distributed. Populations mainly showed aggregated distributions at relatively smaller scales and random distributions at larger scales, and the aggregation degree decreased with increasing spatial scales. The understory layer tended to be aggregated, while the emergent layer tended to be randomly or regularly distributed. The aggregation degree decreased with canopy height increasing. Spacial associations among tree species were mainly found at smaller scales and gradually disappeared with increasing scales. From the analysis of the inter-specific association of species, Pinus tabuliformis mainly showed negative correlation with Robinia pseudoacacia and Ulmus pumila. And, there was no stable special association between Robinia pseudoacacia and Ulmus pumila. Robinia pseudoacacia stems in emergent layer were negatively associated to the stems in understory layer. There was generally a negative correlation between Pinus tabuliformis and Robinia pseudoacacia in understory layer at smaller scales, the special association between Robinia pseudoacacia in emergent layer and Ulmus pumila in understory layer had the similar correlation. There was a significantly positive correlation among all spicies in understory layer at smaller scales.Through17a vegetation restoration, the contents of organic matter, total N and available K were close to or in excess of the original landscape level, meanwhile, pH and the contents of available N and available P were far lower than that of the original landscape. With the increasing of soil sampling depth, the content of soil nutrient decreased. The relationship between mean density of Robinia pseudoacacia and standing tree mean density was significantly positive, and standing tree mean density had a significantly negative with mean DBH and mean hight of Robinia pseudoacacia. There was a negative correlation between quantitative characteristics of Pinus tabuliformis population and standing tree mean density. The correlation could be expressed by simple linear equations. Robinia pseudoacacia population was closely associated with pH, the average content of available N and available P which could be expressed respectively by the multiple step linear regression between population characteristics parameters and soil factors in the surface soil. The average content of available N had the greatest impacts on Robinia pseudoacacia population, and the least did the average content of available P. Based on these models, cultivation and reproduction of Robinia pseudoacacia population in ecological reclamation could be designed.