| Deposit body is the accumulation of waste soil and residue sourcing from variousconstruction projects, which is related to transportation, mining engineering, hydraulicengineering and urbanization. Its quantity will increases with the advancing economy. Anddeposit body has much more serious soil erosion problem than initial landform cause its steepside slope, looseness surface and compound ground substance. Soil and water erosion ofdeposit body has become a part of newly increased soil and water loss. Soil erodibility, anindicator of soil sensitivity to soil loss, is a key part in quantitative evaluation of soil loss. It isof great meaningful to the soil loss prediction, supervision and evaluation in constructionprojects for researching deposit body soil erodibility.Considering the antecedent fieldinvestigation, we generalize the deposit body surface as mixture of soil and specific contentartificial selected rock. This paper were aim to evaluate the soil erodibility of deposit body,and to analyze the overflow erosion process using simulated rainfall experiment. Onanalyzing the experiment data, we have the following mainly conclusions.(1) Runoff characteristics of deposit overflow erosion in sandy loess and red earth hadbeen discussed. We found correlations were significant and positive for time to initial runoffand rainfall intensity, and negative for time to initial runoff and fragment percent. Time toinitial runoff increases with rainfall intensity. An increasing rainfall intensity always cause anincreasing runoff, there exist power relationship between them. Runoffs rise then go downwhile the fragment percent increasing. The runoffs have valley points at10%fragmentcontent. On the contrary, infiltration rates fall down first then upswing along with anincreasing fragment percent. Infiltration rates have peek values at10%fragment content.Infiltration rates increase with rainfall intensity. The average runoff and runoff coefficient insandy loess were1~3and1~2times to that in the red earth.(2) Sediment characteristics of deposit overflow erosion in sandy loess and red earthrespectively and their differences. The average sediment rates increased with rainfall intensityin power law. Sediment rates rise first then fell down with fragment content for there to bepeek values at10%in1.0mm/min intensity treatment. When the intensity larger than1.0mm/min, sediment rates decreased with fragment content. Sandy loess have lager peek sediment rates, average sediment rates and15min maximum sediment rates compared withthe red earth. The three parameters in sandy loess were1~8.7,1.2~4.5,1.6~8.3times asbigger as in red earth, respectively.(3) The average overflow velocity power increased with rainfall intensity in both soiltypes. The velocities of soil with fragment in were smaller than that of pure soil. The Froudenumber varied within0.72~1.41in sandy loess and within0.53~0.97in red earth. It impliedthat the overflow in sandy loess were torrent but subcritical flow in red earth. The Froudenumbers were getting bigger while rainfall intensity and fragment percent increasing. TheReynolds numbers in both soil types were smaller than500except one simulation of2.5mm/min in rainfall intensity and0%in fragment percent. Those suggest the overflow werelaminar flow. The Reynolds numbers increased with rainfall intensity in power correlationshipand fragment content. The average overflow velocity and the Froude numbers of sandy loesswere significantly larger than than of red earth. However, none of any difference had beenfound between their Reynolds numbers.(4) Hydrodynamic shear stress, stream power and cross-section specific energy allincreased with rainfall intensity. The values at soil with fragment in were larger than thevalues at soil without fragment. The parameters were no differences between the two soils.There were significantly linear relationships between hydrodynamic shear stress, streampower, cross-section specific energy and average sediment rates, instantaneous sediment rates,soil loss. The responsive relationships of erosion parameters to hydrodynamic parameterscould be ranged in the order of cross-section specific energy>stream power> hydrodynamicshear stress. And the relationships of hydrodynamic parameters to erosion parameters couldbe ranged in the order of instantaneous sediment rates> average sediment rates> soil loss.(5) The critical hydrodynamic shear stress in sandy loess were1.598,1.971,2.106and3.104Pa at0%,10%,20%and30%fragment content. The stream power were0.090,0.122,0.169and0.252N/m·s, respectively. The cross-section specific energy were0.060,0.061,0.063and0.104cm. the critical hydrodynamic shear stress in red earth were2.422,2.445,2.542and2.579Pa at0%,10%,20%and30%fragment content. The stream power were0.098,0.129,0.141and0.142N/m·s, respectively. The cross-section specific energy were0.079,0.082,0.088and0.090cm. Their values in red earth at0%fragment content were1.56,1.08and1.32times bigger than that in sandy loess. Those three parameters increased withfragment content.(6) The K values of0%,10%,20%and30%rock fragment content were0.1775,0.0912,00498and0.0440t·hm2·h/(hm2·MJ·mm) in sandy loess, and0.0534,0.0490,0.0365and0.0215t·hm2·h/(hm2·MJ·mm) in red earth, respectively. Regression analysis reveals the K value decrease with rock fragment percent. Then a linear relationship among Kvalues of different rock fragment content, pure soil erodibility and rock fragment content wasestablished to provide an easy way of acquiring soil K value with rock fragment mixed inresearch region.(7) The comparison results between K value observed in natural rainfall plots and in thisstudies suggested the soil erodibility in deposit body is approximately equal to that in farmland. This conclusion may provide a gist for the following deposit body soil erodibility studyon the basis of results achieved in farmland.(8) Based on the application analyze of three soil erodibilily estimation model, thenomograph model had the best result for the soil estimation in deposit body. |
PDF Full Text Request