|Other Abstract||In present study, field experiments were carried out in desert-oasis ecotone in the south edge of Taklimakan Desert. Four types of typical plots with different vegetation coverage of 30%, 15~20%, 10% and <5% were chose as study sites. Soil at depth of 0-10 cm and 10-20 cm in Tamarix ramosissima nebkhas and interdune for analysis of soil grain size, litter, water , soil organic matter, total N, total P, total K, available N, available P, available K content, soil pH, total salinity, Cl-, Na+, K+, SO42-, Ca2+, Mg2+, CO32- and HCO3- content. Results showed that :
(1) The study area was dominated by very fine sand and silt and both belong to the partical size that can be transported by wind. With the decrease of vegetation cover from 30% to 15~20% to 10% and to <5%, the average partical size in 0-10 cm soil layer gradually increased from 74.41 μm, 77.28 μm, 86.29 μm to 92.71 μm and accompanied by clay disappear and coarse sand appeared. Under the same vegetation condition, the minimum soil partical size mainly distributed under shrub, and the maximum value mainly distributed in interdune. For different position of nebkhas, the sorting of the interdune soil was the best, while under shrub was the worst. The vegetation in desert-oasis ecotone can intercept a considerable part of suspended sediment transported from far away places by wind and lead the surface material of the study area becoming more fine. The area of under shrub had much finer grains and the interdune had coarser sand apparently, which were closely related to wind-driven and redistribution of sand grain in situ. In areas with vegetation coverage less than 5% both nebkhas and interdune surface existed widespread soil erosion and ‘resource islands’ effects disappeared. For the sake of the stability of nebkhas, the total vegetation coverage should be maintained at least>10% in desert-oasis ecotone.
(2) Soil water content is extremely low in Tamarix nebkhas-interdune system. The average values are all below 0.4%. Maximum values mainly distribute under shrub position. The soil water content change trends are same from April to October, with the minimum value appeared in August.
With the decrease of vegetation coverage, Nebkhas litter, soil organic matter, total N , total K and available K content decreased. The content of available N and available P decreased firstly and then increased. The total P content had no obvious change. Litter, soil organic matter, total N, total P, available N, available P and available K content from Under shrub to Margin of nebkha to interdune to Leeward shadow decreased at first and then increased. Maximum values are happened under shrub position, and minimum values in the interdune. For each nutrient index, there was no significant difference among four study sites with vegetation coverage of <5%. Litter, organic matter, total N , total P, total K, available P and available K content was generally reduced in 10-20 cm soil layer compared to these in 0-10cm soil layer. With the decrease of vegetation coverage, the change trends of soil nutrient in two soil layers were consistent. The biological feedback of shrub made the nutrient Under shrub position obviously enriched. But with the decrease of total vegetation coverage and the increase of aeolian activities, nonbiological factors gradually became the dominant and the nutrient enrichment effect of shrub began to diminish and disappear.
(3) With the decrease of vegetation cover from plot 1 to plot 4, soil pH, total salinity, Cl-, K+and Na+ content in 0-10 cm layer increased at first and then decreased. The maximum values mainly distribute in the plot with vegetation cover of 30%, the minimum values mainly distributed in the plot with vegetation cover of 15~20%. However, the maximum values of HCO3- content mainly distributed in the plot with vegetation cover of 15~20%, and there was no obvious change of Mg2+ content. However, soil pH and HCO3- content increased at first and then decreased and slightly cor-relationship was found between soil pH and HCO3- content (r=0.644). The maximum values mainly distributed in the ground of Under shrub, while the minimum values mainly distributed in the Interdune. Total salinity, Cl-, Na+ and K+ content in 10-20 cm soil layer significantly increased compared to these in 0-10 cm soil layer. However, SO42- content, Ca2+ content and HCO3- content were reduced and soil pH without obvious change. With the decrease of vegetation coverage, the changes of soil salinity content in the two soil layers are consistent. The transpiration, recretion and selective absorption of root of Tamarix ramosissima and soil evaporation result in obvious soil salinity enrichment effects around the area of Nebkhas. However, with the decrease of total vegetation cover and the increase of desertification intensity, the non-biological factors such as wind erosion began to play a leading role, and soil salinity enrichment effects of shrubs tended to diminish and disappear gradually.
(4) Nebkhas litter content and distribution had a strong influence on the distribution of soil nutrients and salinity. Organic matter was a major source of soil nitrogen and potassium. There was a negative correlation between soil pH value and nutrient index, especially showed a significant correlation with organic matter and available potassium. So the content and distribution of soil organic matter and available potassium had a great influence on soil salinity. At the same time high soil nutrient status can reduce the degree of soil salinization. Available K had a good coordination with soil nutrients and salinity index. The distribution of soil nutrients and salinity was mainly affected by the fine material in the study area. It can be concluded that the redistribution of soil fine-grained material was accompanied along with the process of redistribution of soil nutrients and salinity. Effect of low surface soil water content on salinity was greater than the effect of nutrient in Tamarix ramosissima nebkhas and interdune in desert-oasis ecotone.|