This paper selected various types of slope soil under different restoration modes of Xiangjiaba hydropower construction to study the fractionation of inorganic phosphorus. The results showed that the content of inorganic phosphorus was lower.In seven different types of slop soil,the content of total inorganic phosphorus was between 254.43 mg·kg-1 and 992.98 mg·kg-1,and accounted for between 22.60% and 47.12%.The content of soil total inorganic phosphorus and the raito of Water-soluble phosphorus,Al-P to total phosphorus increased as the content of soil total P increase,but the ratio of soil O-P,Ca-P to total phosphorus decreased.The Water-soluble phosphorus and Al-P in Vegetation-growing concrete gunning were the highest,the O-P in Natural secondary forests were the highest and Ca-P in External-soil spray seeding were the highest among all the types of slope soil.

The roots can significantly increase the soil reinforcement of vegetation, and the fractal theory provides a new perspective for vegetation roots studies. This article applied the Fractal Fox software to calculate the fractal dimension of medicago sativa and cynodon dactylon roots in different growth periods and proved that the two species show fractal characteristics. The conclusions from the analysis are as follows: ①The fractal dimensions of the two plant roots tend to be stable with the increase of growth period; ②The fractal characteristic value of cynodon dactylon root is more significant than medicago sativa root; ③Compared with medicago sativa root, cynodon dactylon root is more effective in increasing the shear strength of soil.

Prediction of water demand is a basic link in water resources plan and management. Reasonable and accurate prediction of storage helps to develop the plan of water resources the next year, which is very favorable to improve the utilization ratio of water resources and reduce the waste of water resources. This paper uses BP neural network to simulate and predict the water content based on the data of water in recent ten years in Hubei province and evaluates the forecast results. The results show that BP neural network for water demand prediction is feasible.

Two vegetation-growing concrete slopes completed are chosen as research objects and the Analytical Hierarchy Process method is adopted to establish the evaluation index system, aiming at quantifying the physical and chemical properties of the base material and analyzing the results contrastively. The study shows that the physical and chemical properties of vegetation-growing concrete base material keep improving with the growth of time to better adapt to the growth of vegetation. This evaluation results tally with the actual situation of the sample sites, which suggests that this base material soil quality index system has great practical value. The differences occurring in the evaluation value of two pieces of sample sites suggest that regional environment is the main factor to affect the quality of vegetation-growing concrete base material.

Now, with the large-scale of various projects constructed, a lot of the original vegetation were destroyed and formed a large area of bare slope. The existence of slopes increased the occurrent intensity of soil erosion, landslides and debris flow, and also caused local ecological disasters of the deterioration of the microclimate and the destruction of the biological chain. Based on this situation, this article discussed the reinforced soil reinforcement of shallow fine roots and deep rough roots, and generalized comprehensively the root factors on the impact of the reinforced soil, and explored the lack of research and development trends of vegetation slope protection.