检索项 检索词
  土壤学报  2020, Vol. 57 Issue (6): 1387-1398  DOI: 10.11766/trxb201907200268
0

引用本文  

张德谦, 倪世民, 王军光, 等. 不同侵蚀程度花岗岩红壤坡面侵蚀泥沙颗粒特征研究. 土壤学报, 2020, 57(6): 1387-1398.
ZHANG Deqian, NI Shimin, WANG Junguang, et al. Particle-size Composition of Erosion Sediment on Granite-derived Red Soil Slope Relative to Erosion Degree. Acta Pedologica Sinica, 2020, 57(6): 1387-1398.

基金项目

国家自然科学基金重点项目(41630858)和国家重点研发计划项目(2017YFC0505404)资助

通讯作者Corresponding author

王军光, E-mail:jgwang@mail.hzau.edu.cn

作者简介

张德谦(1993—),男,湖北十堰人,硕士研究生,主要从事土壤侵蚀机理方面研究。E-mail:dqzhang1993@163.com
不同侵蚀程度花岗岩红壤坡面侵蚀泥沙颗粒特征研究
张德谦, 倪世民, 王军光, 吴文枭, 蔡崇法    
华中农业大学水土保持研究中心,农村农业部长江中下游耕地保育重点实验室,武汉 430070
摘要:研究花岗岩红壤坡面侵蚀泥沙颗粒分选特征有助于揭示粗颗粒土壤坡面侵蚀机理。通过室内人工模拟降雨-径流试验,在15°坡度和2种试验条件下(90 mm·h-1+2.00 L·min-1、120 mm·h-1+2.66 L·min-1),研究了不同侵蚀程度(轻度侵蚀E1、中度侵蚀E2、强烈侵蚀E3)花岗岩红壤侵蚀过程及泥沙颗粒组成的变化规律。结果表明:产流率和土壤侵蚀速率随降雨强度和上方来水量的增加而增大;不同侵蚀程度土壤其侵蚀特征存在差异,产流率表现为E2 > E3 > E1,土壤侵蚀速率表现为E3 > E2 > E1;通过多元逐步回归分析得出,土体容重越高产流率越小(R2=0.800,P < 0.01),更高的黏粒含量和更低的粉粒含量会使得土壤侵蚀速率减少(R2=0.715,P < 0.01)。随着降雨-冲刷强度增加泥沙中砾石和黏粒含量增加,砂粒和粉粒含量减少,泥沙分形维数增大。分形维数与黏粒(R2=0.910)和砂粒(R2=-0.390)质量分数为极显著线性关系,可以作为花岗岩红壤坡面侵蚀产沙中泥沙粒级分布的评价指标。
关键词人工模拟降雨    花岗岩红壤    土壤侵蚀    土壤颗粒    
Particle-size Composition of Erosion Sediment on Granite-derived Red Soil Slope Relative to Erosion Degree
ZHANG Deqian, NI Shimin, WANG Junguang, WU Wenxiao, CAI Chongfa    
Research Center of Water and Soil Conservation, Huazhong Agricultural University, Key Laboratory of Arable Land Conservation (Middle and Lower Reaches of Yangtze River) in Ministry of Agriculture and Rural Affairs, Wuhan 430070, China
Abstract: 【Objective】Soil erosion is a main mechanism of land degradation. Much works have been done on sensitivity of coarse-textured soils to erosions different in degree, however, little is known about erosion mechanism of slopes of granite-derived red soil. An indoor simulation experiment was done to characterize runoff and sediment production on steep slopes of granite-derived red soil as affected by rainfall intensity and erosion degree and to explore erosion mechanism of slopes of coarse-textured soils.【Method】The experiment had flumes (3 m long, 1 m wide and 0.45 m high, each) packed with coarse-textured soil to simulate soil slopes different in erosion degree, two treatments in rainfall intensity (90 and 120 mm·h-1) and two in inflow rate (2 and 2.66 L·min-1). Based on field investigations and previous studies, the three slopes were designed to be slight (E1), moderate (E2) and severe (E3) in erosion degree, separately and all 16.7% in gradient, commonly seen in the hilly region in the sub-tropical climate zone of China. Before each run of the experiment, the flume was covered with a gauze to reduce the splashing impact of rain drops and then treated with a rainfall event low in intensity (30 mm·h-1) till runoff appeared at the outlet of flume. Before each run, soil moisture content in the soil slope was determined with a soil moisture meter to ensure it was no more than 30%, and rainfall intensity and inflow rate was calibrated to ensure they were up to the set requirement of the test (±5%).【Result】Results show: (1) both rainfall intensity and flow rate played important roles in runoff formation and soil erosion. Runoff yield rate and soil erosion rate increased with rainfall intensity and inflow rate, and soil erosion varied in characteristic with slope erosion degree. The slope slight in erosion degree (E1) was most affected by rainfall intensity and inflow rate. (2) The slopes varied in erosion characteristics with erosion degree erosion. The slopes exhibited an order of E2 > E1 > E3 in runoff yield rate and an order of E3 > E2 > E1 in soil erosion rate. Multi-factor stepwise regression analysis shows that the higher the bulk density, the lower the runoff yield rate (R2 = 0.8, P < 0.01) and the higher the clay content or the lower the silt content, the lower the soil erosion rate (R2=0.715, P < 0.01). (3) With rising rainfall intensity and scouring strength or inflow rate, gravel and clay contents increased, sand and silt contents decreased and fractal dimension of the sediment increased. (4) Fractal dimension was ultra-significantly and linearly related to the ratio of clay/sand in content.【Conclusion】All the findings in this study may provide a train of thought for evaluation of soil evolution under erosion and contents of the research on soil evolution.
Key words: Simulated rainfall/runoff    Granite red soil    Soil erosion    Sediment size distribution    

我国长江以南山地丘陵区地处热带亚热带,雨量充沛,由于自然环境、花岗岩母质特征及人为活动破坏,土壤侵蚀日益加剧,现已成为我国仅次于黄土高原的第二大侵蚀区,特别是花岗岩红壤区已成为我国水土流失严重的区域[1-3]。土壤侵蚀是一个复杂的过程,受降雨和下垫面等因素的综合影响。在花岗岩深厚的风化壳中,各个风化层土壤性质不同,不同强度的水力侵蚀造成了不同层次土壤不同程度地暴露在外,其理化性质与抗侵蚀能力差异非常明显[4-7],造成坡面侵蚀产沙过程与泥沙颗粒特性存在较大差异[8]。降雨是引起土壤侵蚀的驱动因子[9],大量研究表明,降雨强度越大土壤侵蚀越严重,特别是在细沟发育的活跃阶段[10]。而短历时暴雨是造成该区坡面侵蚀的主要雨型[11],加之缺乏植物和有机质,裸露的花岗岩红壤坡面极易受到降雨和径流冲刷的影响[12]。随着极端降水情况的频发,暴雨对土壤侵蚀的影响应引起人们的重视。

土壤颗粒随径流的输移和分选特征是坡面侵蚀的研究重点,一直被国内外学者所关注。当降雨动能较低时,土壤中细颗粒优先被分散[13-14];更高的降雨动能会增加径流搬运能力,使得泥沙颗粒中粗颗粒含量增加,黏粒含量相对减少[15]。当坡面径流水流功率超过临界值时,径流分选作用降低,泥沙的颗粒组成与原生土壤逐渐趋于相似[16]。侵蚀过程对泥沙颗粒的分选也存在一定的作用[17],当细沟发育后,降雨侵蚀颗粒粗化现象更为明显[18]。泥沙的颗粒分布也受土壤母质和降雨侵蚀力影响,土壤母质相同的情况下,雨强和坡面条件会影响泥沙颗粒的比例,黏粒含量会随着坡长和雨强的增加而增加[19]。Berger等[20]在研究细沟发展与土壤侵蚀的关系中提出雨强和坡度对细沟泥沙颗粒的分布均有影响,且雨强的影响大于坡度。当土壤母质不同时,泥沙颗粒分布也会产生明显差异[21]

目前,关于花岗岩红壤的侵蚀特征研究中,有限的探索主要集中在红壤表土与崩积堆的泥沙颗粒分选特征方面[12, 22],而不同侵蚀程度土壤间的坡面侵蚀规律与泥沙分选特征研究不足。鉴于以上研究背景,本试验以花岗岩发育不同侵蚀程度红壤为对象,研究了人工模拟降雨和径流冲刷条件下的坡面侵蚀产流产沙特征,揭示了降雨强度和上方来水流量双因素与坡面泥沙颗粒特征的关系,以期为南方红壤区水土保持提供科学依据。

1 材料与方法 1.1 供试材料

供试土样采集于福建省长汀县和田镇(25°37′58.69″N,116°27′29.41″E)。该区属于亚热带季风湿润气候,年平均气温18.5 ℃,年平均降水量1 650 mm。采样时参照Wu等[23]判断土壤侵蚀程度依据及立地崩岗侵蚀发生发育程度,在尚未发生崩岗侵蚀的花岗岩发育残积土坡面采集轻微侵蚀程度土体(E1),采样深度为10 cm,土壤层次为表土层,有丰富植被生长(主要为铁芒萁和马尾松),结构较好。在有崩岗侵蚀发生的集水坡面靠近坡顶位置采集中度侵蚀土体(E2),采样深度为10 cm,土壤层次为红土层,地表草本植被稀少(以马尾松为主),土体黏重,土粒细腻紧实,土壤结构严重破坏,透水性差。严重侵蚀土壤(E3)采集自崩岗崩壁,采样深度为200~230 cm,土壤层次为过渡层,呈红色夹杂零星黄色,土壤极为贫瘠,保肥持水能力差,土体结构疏松,用手搓会有沙质感。采样点及剖面详情见图 1。将土样运回试验室后,自然风干并过10 mm筛,剔除石块、根系等杂质后备用。土壤有机质采用重铬酸钾容量法测定,土粒密度采用比重瓶法测定,pH通过pH计测定;土壤粒径分布使用超声波振荡仪分散土壤颗粒,并通过湿筛法结合吸管法测定,采用美国制粒径分级,得到砾石(> 2 mm)、粗砂粒和细砂粒(0.05~2 mm)、粉粒(0.002~0.05 mm)、黏粒(< 0.002 mm)质量分数。土壤容重采用野外原位采集的环刀测定(100 cm3)。土样的基本理化性质见表 1

图 1 野外采样情况 Fig. 1 Soil profile of the Ultisol(red soil)derived from granite

表 1 试验土样的基本理化性质 Table 1 Basic physicochemical properties of the soil samples used in the experiment
1.2 研究方法

模拟降雨装置采用垂直旋转下喷式自动模拟降雨系统,该降雨器通过小、中、大三种规格的喷头组合实现15~220 mm·h–1降雨强度范围,降雨均匀度大于90%。降雨高度为10 m,降雨雨滴可以达到终点速度[24]。试验土槽规格为3.0 m(长)×0.8 m(宽)×0.45 m(高),坡度可在0°~30°自由调节。本试验基于开放小区,在降雨-径流耦合条件下进行模拟实验。径流通过恒压水泵和槽子后端的稳流槽提供,流量由供水管道上的两个控制阀调节。按照野外采集的环刀测定的容重,将3个层次土样分别按其容重作为试验土壤填入土槽。填装土样时,采用分层填装,每层5 cm,并在上层土壤填入前对下层土壤刮毛,逐层填至30 cm以保持土壤颗粒分布均匀贴合紧密。完成土壤装填后在土槽顶部覆盖一个15 cm长的纱布,以尽量减少径流冲刷和边缘效应的影响,试验重复2次,试验装置见图 2

图 2 试验装置模拟图(A),模拟降雨装置(B),试验图(C) Fig. 2 Flume(A)and rainfall simulator(B)used in the experiment and test diagram(C)of the experiment

根据长汀暴雨经验公式,针对我国亚热带气候区典型的强风暴,设置了在90 mm·h–1+2.00 L·min–1和120 mm·h–1+2.66 L·min–1两种试验条件下(假定坡面上游有2 m2的集水区)对三个侵蚀程度土壤共进行了6组降雨-冲刷试验。丁光敏[25]在对福建崩岗侵蚀成因的调查研究中发现10°~25°中坡为最易产生侵蚀的坡度,故本试验设置试验坡度为15°,也是崩岗侵蚀严重区集水坡面常见坡度。试验降雨强度根据当地设计暴雨强度公式计算了一年一遇、五年一遇、十年一遇下I30minI45minI60min选定,其中90 mm·h–1为多年平均暴雨降雨强度,120 mm·h–1为极端暴雨条件下降雨强度。流量设置依据严冬春等[26]不同坡度下细沟侵蚀发生的临界坡长,等比例添加上方来水使对应降雨强度下坡面汇水坡长达到5 m。试验前以小雨强(30 mm·h–1)对坡面进行预湿润处理,使用纱布覆盖表土,减少雨滴溅蚀影响,直至产流,并放置24 h后开始试验。降雨历时1 h,每分钟采集3组泥沙样品,用于计算土壤侵蚀速率、泥沙浓度,并利用每三分钟时间段中第三分钟的三个样品(即第3、6、9…60分钟的三个样品)通过湿筛和吸管法测定泥沙颗粒粒径:砾石(> 2 mm)、砂粒(0.05~2 mm)、粉粒(0.002~0.05 mm)、黏粒(< 0.002 mm)。

1.3 参数获取

综合有关研究选取下列相关参数。土壤侵蚀速率[27](土壤剥蚀率)为单位时间、单位侵蚀面积上土壤在水流侵蚀动力的作用下被剥蚀的土壤颗粒质量,表征径流对坡面土壤的分离能力,其求解方法如下:

${D_r} = \frac{{{m_t}}}{{BLT}}$ (1)

式中,Dr为土壤侵蚀速率,kg·m–2·min–1mt为累积产沙量,kg;B为水宽,cm;L为沟长,m;T为时间,min。

为了研究泥沙粒径分选机制,计算了泥沙中各粒径含量和原土样对比的富集率(Enrichment ratio,ER)[22],具体计算公式如下:

${\rm{ER}} = \frac{{{P_s}}}{{{P_o}}}$ (2)

式中,ER为富集率;Ps为泥沙某一粒径的百分比,%;Po为土壤某一粒径的百分比,%。ER > 1表示该颗粒富集,ER < 1表示该颗粒损耗。

分形维数通过表征土壤颗粒粒径分布、土壤质地均一程度、土壤团粒结构团聚体、水稳性团聚体和粒径组成等[28],可用于反映坡面土壤质地及肥力对侵蚀过程的响应[29]及坡面土壤结构对侵蚀的影响[30]。Mandelbrot [31]建立了二维空间颗粒大小分形特征模型。此后,Tyler和Wheatcraft[32]在此基础上建立了三维空间的体积分形维数模型:

$\frac{{W(\delta > {d_i})}}{{{W_{\rm{o}}}}} = 1 - {\left( {\frac{{{d_i}}}{{{d_{\max }}}}} \right)^{3 - D}}$ (3)

对分形维数模型进行推导,两边取对数,以左右两边为因变量和自变量进行线性拟合,拟合的直线斜率即为3$ - $D,从而可求出分形维数D

$\lg \frac{{W(\delta > {d_i})}}{{{W_{\rm{o}}}}} = (3 - D)\lg \frac{{{d_i}}}{{{d_{\max }}}}$ (4)

式中,di=(dj+dj+1)/2,W$\delta $ > di)为粒径大于di累积土粒质量;Wo为各粒级土粒质量之和;di为筛分粒径范围[djdj+1]平均值;dmax为某个粒径级[djdj+1]最大粒级平均直径。

1.4 数据处理

数据处理与分析在Excel 2016、Origin 2017与SPSS 19.0软件中完成,相关分析采用多元逐步回归分析。

2 结果与讨论 2.1 不同降雨强度和上方来水流量对产流产沙的影响

不同降雨-冲刷强度下坡面产流率变化如图 3。产流率随产流时间表现为先增加、后相对稳定或略有波动。其原因是降雨初期,土壤含水率较低,未达到饱和,雨水下渗作用延缓了径流的产生。随着降雨历时的延长,土壤含水率逐渐饱和,加之雨滴击溅导致地表粗颗粒破碎分离,细颗粒逐渐堵塞土壤孔隙,促进径流的产生,当坡面逐渐达到稳定下渗率时,产流率也趋于稳定。两种降雨-冲刷强度下产流率表现为E2 > E1 > E3,差异显著(表 2)。其原因是坡面侵蚀过程取决于降雨特征、土壤性质和地表条件等因素[23, 33],不同侵蚀程度土体的容重存在显著差异,容重越大产流率也会越大(r=0.91,P < 0.01),而降雨-冲刷强度的增加会减少到达稳定阶段需要的时间,高降雨-冲刷强度下产流率也会更高。

图 3 不同侵蚀程度土体产流率随产流时间的变化 Fig. 3 Temporal changes in runoff rate relative to erosion degree of the slope

表 2 不同侵蚀程度土体侵蚀特征 Table 2 Soil erosion characteristics relative to erosion degree of the slope

不同降雨-冲刷强度下各侵蚀程度土壤侵蚀速率变化如图 4所示。除E1在90 mm·h–1+2.00 L·min–1 条件下,其余各试验条件下土壤侵蚀速率随着产流历时呈增加趋势。各土体土壤侵蚀速率差异显著,表现为E3 > E2 > E1(表 2),E3在90 mm·h–1+2.00 L·min–1条件下土壤侵蚀速率为0.324 kg·m–2·min–1,为E2的1.57倍和E1的9.95倍;在高雨强下E3土壤侵蚀速率为0.740 9 kg·m–2·min–1,为E2的1.76倍和E1的2.40倍。随着降雨-冲刷强度的增加,各侵蚀强度土体土壤侵蚀速率也显著增加,E1、E2及E3分别增加了9.48倍、2.03倍、2.28倍。究其原因为E1在90 mm·h–1+2.00 L·min–1条件下,侵蚀过程主要为雨滴击溅与层状侵蚀,未发育出细沟。随着土壤表面可被搬运的细颗粒被输移耗尽,产流率增加加剧了坡面粗化程度,使得表土层中丰富的砾石裸露在地表对下层土壤起到了保护作用。当降雨-冲刷强度增加后,坡面水流功率增加,超过了地表对粗颗粒的阻力,使其推移搬运。随着细沟开始发育,坡面集中水流侵蚀能力增大,土壤侵蚀速率随之增加[34],并伴有细沟沟壁崩塌,使得土壤侵蚀速率存在波动。由于水流功率较高,E2、E3土体与E1之间基本性质存在差异,粗颗粒含量也远少于E1,导致E2、E3坡面细沟发育过程剧烈,土壤侵蚀量也高于E1。

图 4 不同侵蚀程度土体土壤侵蚀速率随产流时间的变化 Fig. 4 Temporal changes in sediment yield rate relative to erosion degree of the slope

为研究降雨-冲刷强度对各侵蚀程度土体侵蚀特征的影响,进行了多元逐步回归分析。结果表明,容重、降雨-冲刷强度和土壤粉粒含量对产流率起正向作用(R2=0.800,P < 0.01),容重起主要作用。降雨-冲刷强度和土壤粉粒含量对泥沙浓度起正向作用,土壤黏粒含量对泥沙浓度起负向作用且起主要作用(R2=0.689,P < 0.01)。土壤侵蚀速率受降雨-冲刷强度、砂粒含量和土壤粉粒含量的影响(R2=0.715,P < 0.01),土壤砂粒含量和粉粒含量起主要作用。降雨强度在土壤侵蚀中起着双重作用,不仅使团聚体消散破碎,而且形成径流促进泥沙搬运。花岗岩红壤随着侵蚀程度的加深,土壤性质随之发生变化,表现为粉粒和砂粒含量的增加与黏粒含量的减少。

2.2 不同降雨强度和上方来水流量下泥沙分选差异

图 5是各侵蚀程度土体泥沙颗粒分布随产流时间的变化。E1各粒级泥沙分布随产流历时变化均较大,黏粒和粉粒含量逐渐减少,而粒径较大的砂粒和砾石含量逐渐增加。其原因是试验初期坡面侵蚀为片蚀阶段,悬移-跃移是主要的搬运机制[35],径流会优先选择易侵蚀的细颗粒,而坡面形成的薄层水流在雨水冲击作用下,径流雷诺数的增加加剧了细颗粒的侵蚀。随着细沟侵蚀阶段开始,径流功率增加,加强了对大颗粒泥沙或团聚体的搬运能力,使大粒径的砾石发生输移[36]。集中水流作用下细沟内泥沙分选作用下降,细沟间侵蚀仍以雨滴击溅为主,但产流初期细颗粒已被大量侵蚀,导致了细颗粒随降雨时间搬运量下降,致使坡面发生粗化。而泥沙中粒径较大的砾石在初始阶段较低也是由于其质量较大,水流产生的拖拽力小于泥沙颗粒与坡面的摩擦力,无法使土粒在坡面上发生滑动或滚动而发生输移。而中度侵蚀土体和严重侵蚀土体在降雨前期较短的时间变化较大,此后有波动但整体变化趋势并不大。这是由于随着土体侵蚀程度的加深,其抗侵蚀能力相应下降,细沟在极短的时间产生,集中水流功率迅速增加,分选作用大大下降。

图 5 不同侵蚀程度土体泥沙颗粒随产流时间的变化 Fig. 5 Temporal changes in particle size composition of the soil slope relative to erosion degree of the slope

随着降雨-冲刷强度增加砾石和黏粒含量均增加,砂粒和粉粒含量减少(表 3)。这主要是因为更高的降雨-冲刷能量具有更大动能带动大颗粒的运移,同时也会增加细沟间区域的侵蚀强度,引起黏粒和粉粒含量的改变。随着降雨-冲刷强度增加,不同粒径颗粒增加比例也存在差异。高降雨-冲刷条件下E1、E2、E3砾石含量分别增加了13.61%、7.54%、13.60%,而黏粒分别增加了6.73%、0.39%、1.48%。其差异表明高降雨-冲刷条件对细沟侵蚀作用的影响作用明显高于细沟间侵蚀。砂粒含量减少较大,通过对砂粒粒径细化可以看出,其改变的主要原因是细砂粒(0.05~0.25 mm)含量的减少和粗砂粒(0.5~2 mm)含量的增多,使其分布更趋近于原土。高降雨-冲刷条件下细沟会更早的形成,细沟水流功率增加,较大粒径的粗砂粒侵蚀量增加,而水层厚度的增加削弱了雨滴击溅的影响,导致细砂粒搬运量减少。泥沙中各侵蚀程度砂粒含量均很高,这也与原土砂粒含量较高有关。

表 3 不同侵蚀程度土体泥沙颗粒质量分数及富集率 Table 3 Particle size composition and enrichment rates of various fractions of the sediment as affected by erosion degree

对比两个降雨-冲刷强度下各侵蚀程度土体泥沙的富集率可以发现:在较低降雨-冲刷强度下,泥沙中黏粒和粉粒的富集率较高、砂粒和砾石富集率均较低(表 3),这与杨伟等[37]的研究结果一致。细颗粒物质的富集和土壤表面团聚体的破碎程度有关[38],在较低降雨-冲刷强度下,受到雨滴动能的影响,大粒径颗粒被剥蚀为较小粒径颗粒,小颗粒以悬移的方式流失,而大颗粒较难被搬运。这间接表明粗颗粒在坡面地表富集,这一现象与在野外花岗岩红壤崩岗侵蚀区集水坡面观察到的现象一致。在更高的降雨强度下砾石富集率增幅较大,E1、E2、E3分别增加了2.52倍、3.96倍、31倍,其中E1砾石富集率接近1,而E2、E3则远大于1。这一现象可能归因于泥沙输移机制的差异(悬移、跃移和推移过程)[39],粒径较大的颗粒通过推移被搬运,粒径较小的颗粒通过悬移-跃移被搬运[40]。当降雨-冲刷强度增加后,细沟更快的发生发育,水流功率及挟沙能力的增加,明显增大了推移搬运的比重。

2.3 不同降雨强度和上方来水流量下泥沙颗粒分形维数变化

已有研究表明土壤颗粒分形维数越高,土壤细颗粒化越明显,土壤质地相对会更好[41]。在泥沙颗粒分选研究中引入分形维数的概念,有助于更好地表征不同粒径颗粒的变化。泥沙颗粒质量分形维数与其黏粒、粉粒和砂粒质量分数之间的关系如图 6所示。由图可知,分形维数与黏粒和砂粒质量分数之间关系均可以采用线性模型进行较好拟合(P < 0.01),其中,分形维数与黏粒质量分数之间为极显著正相关关系,与粉粒和砂粒质量分数之间为负相关关系(其中与砂粒质量分数呈极显著负相关),泥沙颗粒分形维数主要由黏粒的质量分数决定,这与刘淼等[42]、王国梁等[43]的研究结果相符。因此,暴雨驱动下分形维数相对较高反映了泥沙中砂粒比重的下降与黏粒比重的升高,可以作为评价不同侵蚀程度坡面侵蚀泥沙的颗粒粒级分布特征的指标。

图 6 分形维数与侵蚀泥沙砂粒、粉粒和黏粒质量分数的相关关系 Fig. 6 Correlations of fractal dimension with sand, silt and clay in content

图 7为不同降雨-冲刷条件下泥沙颗粒分形维数随产流历时的变化。E1泥沙质量分形维数在较低降雨-冲刷强度下随着产流历时的延长逐渐降低,当降雨-冲刷强度增加后分形维数随产流历时表现为先下降后趋于平稳。在90 mm·h–1降雨强度和2 L·min–1冲刷流量耦合条件下,初期雨滴击溅对土粒的分散作用较强,坡面薄层水流的搬运和层状剥蚀作用使得侵蚀泥沙中细颗粒增多。随着降雨历时延长,坡面粗颗粒露出减少了雨滴击溅的影响,分形维数随之逐渐减小。在120 mm·h–1降雨强度和2.66 L·min–1冲刷流量耦合条件下,雨滴击溅分散土壤颗粒的作用更强,致使细颗粒增多,分形维数相较于较低降雨-冲刷下更大;同时坡面细沟开始发育后,集中水流对泥沙颗粒分选作用反而降低[16],分形维数随之稳定。E2、E3侵蚀颗粒分形维数在试验初期有下降趋势,其后变化不明显,随着降雨-冲刷强度变化分形维数增加不大。其原因是E2、E3相对于砾石含量更少,抗侵蚀能力更低,两个降雨-冲刷条件下均在较早时间内发育出细沟。分形维数存在波动,是因为沟岸坍塌造成了泥沙颗粒分布有较大变动。其中90 mm·h–1降雨强度和2 L·min–1冲刷流量耦合条件下E3侵蚀泥沙分形维数在12~18 min增加,原因是试验条件为降雨-冲刷,抗侵蚀能力较弱的E3不仅存在细沟的溯源侵蚀,坡面顶部在水流冲刷作用下也会出现跌坎,当沟头与跌坎汇合加剧水流对细颗粒的侵蚀,导致分形维数增加,随着细沟的稳定发育,分形维数逐渐降低。

图 7 不同侵蚀程度土体泥沙颗粒质量分形维数随产流时间的变化 Fig. 7 Temporal changes in fractal dimension of sediment particles relative to erosion degree and duration of runoff yielding
3 结论

随着降雨-冲刷强度的增加,土壤侵蚀速率、泥沙浓度和产流率均增加,轻度侵蚀土壤受降雨-冲刷强度影响最大。不同侵蚀程度土体间,土壤侵蚀速率表现为E3 > E2 > E1,而产流率表现为E2 > E1 > E3,通过多元逐步回归分析发现降雨强度越大,黏粒含量越高、粉粒含量越低,土体侵蚀速率越小,而容重越大产流率越高。随着降雨-冲刷强度和降雨历时增加,泥沙中砾石和黏粒含量均增加,砂粒和粉粒含量减少。黏粒和粉粒的富集率较高、砂粒和砾石富集率较低,随着降雨-冲刷强度增加砾石富集率增幅较大。E1泥沙质量分形维数随产流历时降低,随降雨-冲刷强度增加而增大。E2、E3侵蚀颗粒分形维数下降集中在试验初期,细沟发生后其变化不明显,随着降雨-冲刷增加分形维数增加不明显。

参考文献
[1]
Shi D M. Degradation and prevention of soil erosion in China (In Chinese)[J]. Soil and Water Conservation in China, 1987(12): 4-7. [史德明. 我国红壤区侵蚀土壤的退化及其防治[J]. 中国水土保持, 1987(12): 4-7.] (0)
[2]
Xie J S, Li C L, Chen G S, et al. Discussion on rebuilding the ecosystem of granite red soil erosion (In Chinese)[J]. Fujian Soil and Water Conservation, 2000, 12(4): 3-6, 11. [谢锦升, 李春林, 陈光水, 等. 花岗岩红壤侵蚀生态系统重建的艰巨性探讨[J]. 福建水土保持, 2000, 12(4): 3-6, 11.] (0)
[3]
Cai C F, Ding S W, Zhang G Y. Study on soil surface characteristics and its effect on slope erosion of red soils derived from granite (In Chinese)[J]. Research of Soil and Water Conservation, 1996, 3(4): 111-115. [蔡崇法, 丁树文, 张光远. 花岗岩红壤表土特征及对坡面侵蚀影响的研究[J]. 水土保持研究, 1996, 3(4): 111-115.] (0)
[4]
Zhang L P, Chen R Z, Wu Y H, et al. Spatial distribution of macro-pore properties in soil profile on a slope of weathering granite (In Chinese)[J]. Acta Pedologica Sinica, 2018, 55(3): 620-632. [张丽萍, 陈儒章, 邬燕虹, 等. 风化花岗岩坡地土壤剖面大孔隙特性的空间分布[J]. 土壤学报, 2018, 55(3): 620-632.] (0)
[5]
Deng Y S, Cai C F, Xia D, et al. Soil Atterberg limits of different weathering profiles of the collapsing gullies in the hilly granitic region of Southern China[J]. Solid Earth, 2017, 8(2): 499-513. (0)
[6]
Wang Q X, Ding S W, Deng Y S, et al. Hydrodynamic characteristics of erosion in different soil layers in granite collapse region (In Chinese)[J]. Acta Pedologica Sinica, 2017, 54(3): 570-580. [王秋霞, 丁树文, 邓羽松, 等. 花岗岩崩岗区不同土层的侵蚀水动力学特征[J]. 土壤学报, 2017, 54(3): 570-580.] (0)
[7]
Feng X, Zha X, Huang S Y. Processes and characteristics of erosion on hillslopes of granite red soil under simulated rainfall (In Chinese)[J]. Science of Soil and Water Conservation, 2014, 12(1): 19-23. [冯秀, 查轩, 黄少燕. 人工模拟降雨条件下花岗岩红壤坡面侵蚀过程与特征分析[J]. 中国水土保持科学, 2014, 12(1): 19-23.] (0)
[8]
Defersha M B, Melesse A M. Effect of rainfall intensity, slope and antecedent moisture content on sediment concentration and sediment enrichment ratio[J]. Catena, 2012, 90: 47-52. DOI:10.1016/j.catena.2011.11.002 (0)
[9]
Zhang W B, Xie Y, Liu B Y. Estimation of rainfall erosivity using rainfall amount and rainfall intensity (In Chinese)[J]. Geographical Research, 2002, 21(3): 384-390. [章文波, 谢云, 刘宝元. 用雨量和雨强计算次降雨侵蚀力[J]. 地理研究, 2002, 21(3): 384-390.] (0)
[10]
Shen H O, Zheng F L, Wen L L, et al. An experimental study of rill erosion and morphology[J]. Geomorphology, 2015, 231: 193-201. DOI:10.1016/j.geomorph.2014.11.029 (0)
[11]
Qin W, Zuo C Q, Yan Q H, et al. Regularity of individual rainfall soil erosion in bare slope land of red soil (In Chinese)[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(2): 124-132. [秦伟, 左长清, 晏清洪, 等. 红壤裸露坡地次降雨土壤侵蚀规律[J]. 农业工程学报, 2015, 31(2): 124-132.] (0)
[12]
Jiang F S, Huang Y H, Wang M K, et al. Effects of rainfall intensity and slope gradient on steep colluvial deposit erosion in southeast China[J]. Soil Science Society of America Journal, 2014, 78(5): 1741-1752. DOI:10.2136/sssaj2014.04.0132 (0)
[13]
Bissonnais Y. Aggregate stability and assessment of soil crustability and erodibility:Ⅰ. Theory and methodology[J]. European Journal of Soil Science, 1996, 47(4): 425-437. (0)
[14]
Issa O M, Bissonnais Y L, Planchon O, et al. Soil detachment and transport on field- and laboratory-scale interrill areas:Erosion processes and the size-selectivity of eroded sediment[J]. Earth Surface Processes and Landforms, 2006, 31(8): 929-939. DOI:10.1002/esp.1303 (0)
[15]
Wang L, Shi Z H, Wang J, et al. Rainfall kinetic energy controlling erosion processes and sediment sorting on steep hillslopes:A case study of clay loam soil from the Loess Plateau, China[J]. Journal of Hydrology, 2014, 512: 168-176. DOI:10.1016/j.jhydrol.2014.02.066 (0)
[16]
Slattery M C, Burt T P. Particle size characteristics of suspended sediment in hillslope runoff and stream flow[J]. Earth Surface Processes & Landforms, 1997, 22(8): 705-719. (0)
[17]
Proffitt A, Rose C W. Soil erosion processes.Ⅱ. Settling velocity characteristics of eroded sediment[J]. Soil Research, 1991, 29(5): 685. DOI:10.1071/SR9910685 (0)
[18]
Shi Z H, Fang N F, Wu F Z, et al. Soil erosion processes and sediment sorting associated with transport mechanisms on steep slopes[J]. Journal of Hydrology, 2012, 454/455: 123-130. DOI:10.1016/j.jhydrol.2012.06.004 (0)
[19]
Wu F Z, Shi Z H, Fang N F, et al. Temporal variations of clay content in eroded sediment under different rainfall condition (In Chinese)[J]. Environmental Science, 2012, 33(7): 2497-2502. [吴凤至, 史志华, 方怒放, 等. 不同降雨条件下侵蚀泥沙黏粒含量的变化规律[J]. 环境科学, 2012, 33(7): 2497-2502.] (0)
[20]
Berger C, Schulze M, Rieke-Zapp D, et al. Rill development and soil erosion:A laboratory study of slope and rainfall intensity[J]. Earth Surface Processes & Landforms, 2010, 35(12): 1456-1467. (0)
[21]
Wu X L, Wei Y J, Li Z X, et al. Composition of sediments of erosion from different red soil slopes in subtropical area (In Chinese)[J]. Acta Pedologica Sinica, 2014, 51(6): 1223-1233. [吴新亮, 魏玉杰, 李朝霞, 等. 亚热带地区几种红壤坡面侵蚀泥沙的物质组成特性[J]. 土壤学报, 2014, 51(6): 1223-1233.] (0)
[22]
Lin J S, Huang Y H, Zhao G, et al. Flow-driven soil erosion processes and the size selectivity of eroded sediment on steep slopes using colluvial deposits in a permanent gully[J]. Catena, 2017, 157: 47-57. DOI:10.1016/j.catena.2017.05.015 (0)
[23]
Wu X L, Wei Y J, Wang J G, et al. Effects of erosion degree and rainfall intensity on erosion processes for Ultisols derived from quaternary red clay[J]. Agriculture, Ecosystems & Environment, 2017, 249: 226-236. (0)
[24]
Ni S M, Yang W, Wang J, et al. Soil water repellency of different types of soils aggregate and its effect on the splash erosion (In Chinese)[J]. Journal of Soil and Water Conservation, 2018, 32(1): 167-173. [倪世民, 杨伟, 王杰, 等. 不同类型土壤团聚体斥水性及其对溅蚀的影响[J]. 水土保持学报, 2018, 32(1): 167-173.] (0)
[25]
Ding G M. Causes and control patterns of collapse gully erosion in Fujian Province (In Chinese)[J]. Bulletin of Soil and Water Conservation, 2001, 21(5): 10-15. [丁光敏. 福建省崩岗侵蚀成因及治理模式研究[J]. 水土保持通报, 2001, 21(5): 10-15.] (0)
[26]
Yan D C, Wen A B, Shi Z L, et al. Critical slope length of rill occurred in purple soil slope cultivated land in Three Gorges reservoir area (In Chinese)[J]. Journal of Yangtze River Scientific Research Institute, 2010, 27(11): 58-61, 66. [严冬春, 文安邦, 史忠林, 等. 三峡库区紫色土坡耕地细沟发生的临界坡长[J]. 长江科学院院报, 2010, 27(11): 58-61, 66.] (0)
[27]
Ni S M, Zhang D Q, Feng S Y, et al. Quantitative relationship between hydraulics parameters and soil erosion rate on remolded soil slopes with different textures (In Chinese)[J]. Acta Pedologica Sinica, 2019, 56(6): 1336-1346. [倪世民, 张德谦, 冯舒悦, 等. 不同质地重塑土坡面水沙定量关系研究[J]. 土壤学报, 2019, 56(6): 1336-1346.] (0)
[28]
Liu Y H, Cui D J. Effects of long-term fertilization on fractal dimension (In Chinese)[J]. Chinese Journal of Soil Science, 2005, 36(3): 324-327. [刘永辉, 崔德杰. 长期定位施肥对潮土分形维数的影响[J]. 土壤通报, 2005, 36(3): 324-327.] (0)
[29]
Ma Y, He B H, He J L, et al. Effects of herba andrographitis hedgerow on soil fractal characteristics and erodibility on sloping cropland in Three Gorges reservoir region (In Chinese)[J]. Journal of Soil and Water Conservation, 2011, 25(4): 79-82, 87. [马云, 何丙辉, 何建林, 等. 三峡库区皇竹草植物篱对坡面土壤分形特征及可蚀性的影响[J]. 水土保持学报, 2011, 25(4): 79-82, 87.] (0)
[30]
Wang Q J, Zhao G X, Tao W H, et al. Mathematical model of soil nutrient along surface runoff under water scouring condition (In Chinese)[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(7): 189-195. [王全九, 赵光旭, 陶汪海, 等. 径流冲刷条件下坡地养分随地表径流迁移数学模型[J]. 农业机械学报, 2016, 47(7): 189-195.] (0)
[31]
Mandelbrot B B, Wheeler J A. The fractal geometry of nature[J]. American Journal of Physics, 1983, 51(3): 286-287. (0)
[32]
Tyler S W, Wheatcraft S W. Fractal scaling of soil particle-size distributions:Analysis and limitations[J]. Soil Science Society of America Journal, 1992, 56(2): 362. (0)
[33]
Wu F Q, Fan W B. Effects of soil encrustation on rainfall infiltration, runoff and sediment generation (In Chinese)[J]. Science of Soil and Water Conservation, 2005, 3(2): 97-101. [吴发启, 范文波. 土壤结皮对降雨入渗和产流产沙的影响[J]. 中国水土保持科学, 2005, 3(2): 97-101.] (0)
[34]
Chen X Y, Zhao Y, Mi H X, et al. Estimating rill erosion process from eroded morphology in flume experiments by volume replacement method[J]. Catena, 2016, 136: 135-140. (0)
[35]
Asadi H, Moussavi A, Ghadiri H, et al. Flow-driven soil erosion processes and the size selectivity of sediment[J]. Journal of Hydrology, 2011, 406(1/2): 73-81. (0)
[36]
Zhang G H, Liu Y M, Han Y F, et al. Sediment transport and soil detachment on steep slopes:Ⅰ. transport capacity estimation[J]. Soil Science Society of America Journal, 2009, 73(4): 1291. (0)
[37]
Yang W, Zhang Q, Li Z X, et al. Estimating sediment characters of several typical red soils by simulated rainfall (In Chinese)[J]. Resources and Environment in the Yangtze Basin, 2016, 25(3): 439-444. [杨伟, 张琪, 李朝霞, 等. 几种典型红壤模拟降雨条件下的泥沙特征研究[J]. 长江流域资源与环境, 2016, 25(3): 439-444.] (0)
[38]
Yan F L, Shi Z H, Cai C F, et al. Effects of topsoil aggregate stability on soil erosion at hillslope on ultisoils (In Chinese)[J]. Acta Pedologica Sinica, 2007, 44(4): 577-583. [闫峰陵, 史志华, 蔡崇法, 等. 红壤表土团聚体稳定性对坡面侵蚀的影响[J]. 土壤学报, 2007, 44(4): 577-583.] (0)
[39]
Moss A J, Walker P H, Hutka J. Raindrop-stimulated transportation in shallow water flows:An experimental study[J]. Sedimentary Geology, 1979, 22(3/4): 165-184. (0)
[40]
Asadi H, Ghadiri H, Rose C W, et al. An investigation of flow-driven soil erosion processes at low streampowers[J]. Journal of Hydrology, 2007, 342(1/2): 134-142. (0)
[41]
Zhang S R, Deng L J, Zhou Q, et al. Fractal dimensions of particle surface in the plowed layers and their relationships with main soil properties (In Chinese)[J]. Acta Pedologica Sinica, 2002, 39(2): 221-226. [张世熔, 邓良基, 周倩, 等. 耕层土壤颗粒表面的分形维数及其与主要土壤特性的关系[J]. 土壤学报, 2002, 39(2): 221-226.] (0)
[42]
Liu M, Wu Y Y, Yang M Y, et al. Variation characteristics of fractal dimension of eroded sediment under different rainfall conditions (In Chinese)[J]. Science of Soil and Water Conservation, 2015, 13(2): 37-43. [刘淼, 吴媛媛, 杨明义, 等. 次降雨过程中侵蚀泥沙分形维数的变化特征[J]. 中国水土保持科学, 2015, 13(2): 37-43.] (0)
[43]
Wang G L, Zhou S L, Zhao Q G. Volume fractal dimension of soil particles and its applications to land use (In Chinese)[J]. Acta Pedologica Sinica, 2005, 42(4): 545-550. [王国梁, 周生路, 赵其国. 土壤颗粒的体积分形维数及其在土地利用中的应用[J]. 土壤学报, 2005, 42(4): 545-550.] (0)