Accuracy of methods for field assessment of rill and ephemeral gully erosion
Accuracy of methods for field assessment of rill and ephemeral gully erosion
复制标题
DOI：
10.1016/j.catena.2006.03.005
复制
发表时间：
2006-10-15
期刊：
影响因子：
6.2
通讯作者：
Alvarez-Mozos, J.
中科院分区：
文献类型：
--
作者：
Casali, J.;Loizu, J.;Alvarez-Mozos, J.
为了正确地评估农业区的土壤侵蚀，有必要通过直接测量程序精确地确定田间短暂的沟壑和细沟的体积。然而，关于所使用的不同方法的准确性的信息很少。本文的主要目的是为利用这种直接测量方法对细沟和短暂沟壑侵蚀进行适当的评估提供信息。为实现这一目标：a)探讨了用于现场估算河道横截面积的三种方法的测量误差；b)分析了单位河道长度所使用的横截面数对估算精度的影响；c)考察了河道尺寸和形状对测量误差的影响。被考虑确定横截面面积的三种方法是：微地形轮廓仪(1)；用磁带详细测量截面特征长度(2)；用磁带测量横截面宽度和深度(3)。选取了5个14.0或30.0不同类型的浅沟河段和6个20.4~29.4m的细沟河段。在每个沟渠段上，用上述三种方法测量横截面面积，并在I in的横截面之间进行分隔(S)。对于细沟，采用方法I和方法3测量横截面积，S=2m，然后计算相应的总侵蚀量。以方法I计算的冲沟体积S=i，细沟体积S=2 in作为参考方法。对于每个通道以及对于S和测量方法(M)的可能组合中的每一个，计算相对测量误差和相对于基准定义的相对测量误差的绝对值(E-sm(R)和垂直条E-sm(r垂直条))。即使对于小的S值和明显的准棱柱形通道，也很容易获得远高于10%的垂直条E-sm(r垂直条)。通道大小和形状对测量误差有很大影响。事实上，对于某种沟壑形状和大小，选择更合适的方法似乎比S重要得多，至少在S时代是如此。方法我总是提供最精确的测量，其结果对S的依赖性越小，但S必须在5米以下才能保证误差小于10%。不推荐使用方法2，因为该方法难度大、耗时长，可能会导致较大的错误。方法3似乎对小沟、宽沟、浅沟和小沟都足够，但只有当S的身高小于5英寸时才行。通过对细沟测量误差的分析，得到了与沟谷相似的结果。在使用唯一的代表性断面计算渠道体积时，对E-sm(R)和垂直杆E-sm(r垂直杆)的分析突出了正确选择适当断面的重要性。由于这种方法可能会带来很高的误差值，所以无论何时进行准确的侵蚀测量，都不被认为是可取的。(C)2006爱思唯尔B.V.保留所有权利。
To properly assess soil erosion in agricultural areas, it is necessary to determine precisely the volume of ephemeral gullies and rills in the field by using direct measurement procedures. However, little information is available on the accuracy of the different methods used. The main purpose of this paper is to provide information for a suitable assessment of rill and ephemeral gully erosion with such direct measurement methods. To achieve this objective: a) the measurement errors associated to three methods used for field assessment of channel cross sectional areas are explored; b) the influence of the number of cross sections used per unit channel length on the assessment accuracy, is analysed and; c) the effect of the channel size and shape on measurement errors is examined. The three methods considered to determine the cross sectional areas were: the micro-topographic profile meter (1); the detailed measurement of section characteristic lengths with a tape (2); and the measurement of cross section width and depth with a tape (3). Five reaches of different ephemeral gully types 14.0 or 30.0 in long and a set of six 20.4 to 29.4 m long rill reaches were selected. On each gully reach, the cross sectional areas were measured using the three above mentioned methods, with a separation (s) between cross sections of I in. For rills, the cross sectional areas were measured with methods I and 3, with s = 2 m. Then, the corresponding total erosion volumes were computed. The volume calculated with method I with s = I in for gullies and s = 2 in for rills was taken as the reference method. For each channel, and for each one of the possible combinations of s and measurement method (m), the relative measurement error and the absolute value of the relative measurement error (E-sm(r) and vertical bar E-sm(r vertical bar)), defined with respect to the reference one, was calculated. vertical bar E-sm(r vertical bar) much higher than 10% were obtained very easily, even for small s values and for apparently quasi prismatic channels. Channel size and shape had a great influence on measurement errors. In fact, the selection of the more suitable method for a certain gully shape and size seemed to be much more important than s, at least when s < 10 in. Method I always provided the most precise measurements, and its results were the less dependent on s. However, s must be < 5 m to guarantee an error smaller than 10%. Method 2 is not recommended, because it is difficult, time consuming and can lead to large errors. Method 3 seems to be enough for small, wide and shallow gullies, and for small rills, but only if s is shorter than 5 in. Results obtained after the analysis of rill measurement errors were similar to those of gullies. The analysis of E-sm(r) and vertical bar E-sm(r vertical bar) when calculating channel volumes using a unique representative cross section highlighted the importance of correctly selecting the adequate cross section. Due to the high error values that this method can entail, it is not considered as advisable whenever accurate erosion measurements are pursued. (c) 2006 Elsevier B.V. All rights reserved.