关键词: 干旱区, 遥感, 决策树模型, 干湿季, 土壤盐渍化

Abstract: With the 16 years old Larix kaempferi plantation in eastern mountain area of Liaoning Province, China as test object, this paper studied the effects of fertilization on the nutrient concentrations of five root orders’ fine roots. Under fertilization, less difference was observed in the total C concentration of the fine roots. Among the five orders’ fine roots, the first order’s had the lowest concentration of non-structural carbohydrate (TNC) and the highest ones of N and P, while the fifth order’s was in adverse. The TNC concentration increased with increasing root order, while the N and P concentrations decreased correspondingly. Fertilization only had significant effects on the N and P concentrations of the first order’s fine roots. The C/N/P ratio in different orders’ fine roots had significant differences, being 423∶〖KG-*2〗16∶〖KG-*2〗1 and 726∶〖KG-*2〗16∶〖KG-*2〗1 in the first and the fifth order’s fine roots, respectively. With the increase of root order, the proportion of C increased significantly, while that of N varied little. N fertilization didn’t change the proportion of C, while P or P+N fertilization decreased the proportions of C and N in the first three orders’ fine roots at 0〖KG-*2〗-〖KG-*7〗10 cm soil depth or in the first two orders’ fine roots at 10〖KG-*2〗-〖KG-*7〗20 cm soil depth.

Key words: arid area, remote sensing, decision tree model, dry and wet seasons, soil salinization.