The Simulation In Process Of Spatial Pattern Of Sparse Elm Woodlands

The present study use NetLogo to build an agent-based model to explore the process of the spatial pattern of sparse elm woodlands.We did a field survey in Wulanyudu Experimental Station in Inner Mongolia for model verification.This study considers some factors such as wind,vegetation cover,soil water and competition on the spatial pattern of sparse elm woodlands.The main results are as follows:(1)Wind plays an important influence on seed dispersal.The results show that the number of seeds increases first and then decrease in all wind direction.The number of seeds reached a peak at the 3-4 steps,which accounts for 21.2% of the total.The dispersal distance and number of seeds in different directions is not obvious.(2)Vegetation cover is an important factor affecting seed distribution.The nugget,partial sill,sill and nugget/sill of field survey are 127.93±118.18,241.05±216.30,368.97±334.41 and 30.26%±10.70%,the model's nugget,partial sill,sill and nugget /sill are 39.57,38.49,78.06 and 50.69%.Most of the parameters of the model fall into the range of field survey(mean±standard error).The model output for the density of seed is 74.97 per square meter.(3)The main factors on seedling are soil water content and seed germination rate.The model simulation results show that under the effect of the above factors,the average density of seedlings only reaches 1.77 per square meter,the nugget/sill is 44.40%,and the pattern of the seedling is an aggregate distributed pattern(4)Competition is the key factor for the reduction and death of elm.Considering the influence of competition factors,the average density of elm is only 119.2 trees/ha,and the pattern of elm is a random distributed pattern.This study provides important suggestions for the natural regeneration and the dynamic changes of sparse elm woodlands.