Study On The Particle Characteristics Of Wind Erosion Surfaces Based On Digital Image Processings

The module of surface particle information extraction which integrated with ERDASwas developed based on the previous studies; the wind tunnel was used to simulate theerosion characteristics of coarse graining surface; the characteristics of several typicalerosion surface particles of the agro-pastoral transition area in the middle west of InnerMongolia was studied through quadrat survey and statistics analysis, the geostatistics wasused to study the spatial distribution features of coarse graining surface in Jilantai. Thepotential risk index of soil erosion was raised and used to analysis the potential erosionrisk of the research areas. The main conclusions are as follows:（1） First, portable digital camera is used to obtain surface image. Then,decision-making tree classification is conducted based on the characteristics of RGB grayvalue of particle image. In the next step, the classified and extracted result is revised byarea compensation equation. Finally, technology system to estimate particle content ofdifferent diameter class on surface is established, and the final result error is less than5%.（2） The function module to extract particle information on earth surface is integrateddeveloped. Meanwhile, multi-step batching template is also built to be appropriate fordifferent area, and operating steps of particle image processing on earth surface aresimplified.（3） Compared with quality content, cover characeristics of particle can accuratelyreflect the state of soil particle on wind erosion surface. With the increase of cover degreeof coarse particle, wind erosion rate on surface presents gradually depressed tendency inaccordrance with index regular pattern; with the increase of wind speed, protecting effectof coarse particle can be strengthened, which shows that relatively wind erosion rate onsurface reduces. But, with the increase of coarse particle class on surface, wind erosionrate presents enlargement tendency. Especially when the particle size exceeds3mm andrandom cover degree is10%around, suface wind erosion can be obviously promoted,which resutls in higher erosion rate in initial stage of wind erosion when gravel coverdegree is lesser. The suggest is that gravel sand barrier should reach the big enough coverdegree, and it would be better to be used to sediment transport in prevent sandaccumulated area.（4） Semmivariance funtion models of three soil particle （>0.84mm,0.42-0.84mmand<0.42mm） contents are all linear models in bareness Gobi sample in Jilantai, which shows random distribution stage; coarse graning degree is totally lighter in scrub forestland, that three particles show aggregated distribution.（5） Single Ammopiptanthus mongolicus （Maxim.） Cheng f. and NitrariaTangutorum Bobr. are all present obvious prevent soil wind erosion effect. Blocking sandis obvious on windward side around Nitraria tangutorum Bobr.. On the contratry, thatappeared on leeward side around Ammopiptanthus mongolicus （Maxim.） Cheng f., and itshorizon protection distance is greater than that of Nitraria Tangutorum Bobr.. The spatialheterogeneity scale of highly erodible soil particles are respectively8.41m and14.91m,which both reach or exceed the average space of nebkhas（8.77m which includedsemidiameter of nebkhas）. Given preventing soil wind erosion, that illustrated mutualstimulative united relation existed between nebkhases of constructive species in researcharea, which is benefit for the stabilization and development of vegetations of this area.（6） The indicial equation of soil potential wind erosion risk isQ=(1-P₁-P_0.5×0.5)×D/3.65, which is based on the particle content information on winderiosin surface and the data of the days of high wind in district. The equation is used toevaluate potential wind erosion risks in different research area. The results shows thatpotential wind erosion risk is maximum in Wuchuan farming area, which suppose that weshould increase the force of plant protect and protective farming implementation in thisarea.