The discriminant function with grain size of floodplain and aeolian sediments and its application in the quaternary red clay
Received date: 2013-11-13
Request revised date: 2014-05-21
Online published: 2014-10-10
Copyright
In recent years, increasing interest has been focused on the Quaternary Red Clay (QRC) in the middle and lower reaches of the Yangtze River. The QRC profiles generally contain two parts, the Xiashu Loess(XSL) or Homogeneous Red Clay (HRC) on the topside and the Vermicular Red Clay (VRC) below. The provenance of the QRC was discussed furiously. Some scholars believed that Xuancheng, Jiujiang, and Jinhua etc. share the similar characteristics with the XSL, which may originate from aeolian dust deposit. But the others believed that the QRC in Jinhua and Changsha etc. was deposited by flood. Because of these arguments, this paper collected the QRC samples from Jiujiang (JJ), Jiangxi province; Jinhua (TX), Zhejiang province; Changsha (CS), Hunan province; Nanchang (NC), Jiangxi province; Xinyu (XY) Jiangxi province; Nanping (NP), Fujiang province.Discriminant analysis is a usual method in statistical analysis; it can be used to identify the objects in two groups, and have been widely applied. In this paper, the two groups of training samples were floodplain deposits (FPD) from Nantong (NT) and Jinhua (TG), aeolian dust deposits (ADD) of Malan Loess from Luochuan (LC) and Yili (ZS); the testing samples contained the QRC, FPD in Qidong (QD), Xiashu Leoss from Nanjing (XG) and Malan Loess from Zeketai (ZKT). The discriminate results of QRC were complicated. The samples from NP were all identified as ADD; the samples from JJ, NC, TX and CS were identified as both FPD and ADD. On the topside of the QRC profile, the XSL and HRC were always identified as ADD, and the VRC at the bottom of the profile were not only FPD but also ADD.The identification result of QRC suggests that the provenances of QRC from the middle and lower reaches of the Yangtze River are not only floodplains but also aeolian dust deposits. The north plains of Yangtze River are the dominant dust sources for the QRC which are identified as FPD. The QRC which is identified as ADD is deposited by the winter monsoon, and the aeolian characteristics decreased with latitude reduction and terrain fluctuation.
YANG Lihui , YE Wei , ZHENG Xiangmin , SU You . The discriminant function with grain size of floodplain and aeolian sediments and its application in the quaternary red clay[J]. GEOGRAPHICAL RESEARCH, 2014 , 33(10) : 1848 -1856 . DOI: 10.11821/dlyj201410006
Fig. 1 The distribution of sampling site图1 采样点分布图 |
Tab. 1 The description of sediment profiles表1 沉积物剖面记录 |
| 编号 | 地点 | 坐标 | 样品数量 | 野外描述 |
|---|---|---|---|---|
| ZS | 新疆伊犁昭苏 | 42°58.05′N, 81°29.12′E | 20 | 浅黄色黄土,疏松多孔,无层理,结构均匀,不含砾石。 |
| LC | 陕西洛川 | 35°44.07′N, 109°26.11′E | 57 | 马兰黄土,剖面高290 cm。浅灰黄色,质地疏松,以粉砂为主,孔隙度较大,质地均匀,垂直节理发育。 |
| ZKT | 新疆伊犁则克台 | 43°31.00′N, 83°08.57′E | 215 | 黄色黄土,疏松多孔,无层理,结构均匀,不含砾石。剖面高215 cm。发育三层古土壤。 |
| XG | 江苏南京 新港大道 | 32°09.722′N, 118°53.301′E | 80 | 下蜀黄土,黄棕色,质地均匀,疏松,剖面高420 cm,可划分为四层黄土与古土壤层。 |
| NT | 江苏南通 | 32°04.53′N, 120°42.48′E | 32 | 河漫滩沉积,柱样高64 cm,共可分六层,粉砂与粗粉砂交替。 |
| TG | 浙江金华通济桥 | 29°21.617′N, 119°42.573′E | 9 | 河漫滩相沉积,粗粉砂,颜色为土黄色,颜色均一,质地疏松。 |
| QD | 江苏启东 | 32°01.58′N, 121°43.39′E | 44 | 河漫滩相沉积,含较多粗粉砂,疏松,土黄色,深90 cm岩柱。 |
| TX | 浙江金华汤溪 | 29°02.943′N, 119°24.579′E | 147 | 第四纪网纹红土,上部为红棕色均质红土层,下部为紫红色网纹红土层,质地均匀。 |
| CS | 湖南长沙 | 28°07.424′N, 112°56.593′E | 60 | 第一层黄棕色土层,厚110 cm;第二层黄色土,厚55 cm;第三层黄红色网纹红土层,厚约60 cm;第四层为紫红色网纹红土层,网纹粗大,白色,基质为紫红色。 |
| JJ | 江西九江 | 29º42.034΄N 116º01.702΄E | 920 | 剖面高18 m,可分5层,第一层为现代土壤层,质地较松,厚度较薄,约5 cm;第二层为黄土层;第三层为红色古土壤层;第四层为黄色土层;第五层为发育弱网纹;第六层为紫红色网纹红土层。 |
| NC | 江西南昌 | 28°38.124′N, 115°48.822′E | 29 | 第一层均质红土层,厚150 cm;第二层紫红色网纹红土层,厚250 cm;第三层黄红色网纹红土层,厚200 cm。 |
| XY | 江西新余 | 27°48.522′N, 114°54.362′E | 29 | 第一层为弱发育网纹红土,厚约110 cm;第二层为浅紫色网纹红土层,厚约80 cm;第三层为网纹化砾石层,砾石直径1~2 cm,厚约40 cm;第四层为黄红色弱发育网纹红土层,有大量砾石,厚约90 cm。 |
| NP | 福建南平 | 26°37.977′N, 118°12.56′E | 30 | 第一层红土,厚100 cm,有粒状团聚体和植物根系;第二层厚190 cm,有块状团结构。 |
,其中
,
;Tab. 2 Equality test of group means表2 组均值的均等性检验 |
| Wilks' Lambda | F | df1 | df2 | Sig. | |
|---|---|---|---|---|---|
| Mz | 0.792 | 27.914 | 1 | 106 | 0.000 |
| Sd | 0.808 | 25.191 | 1 | 106 | 0.000 |
| Sk | 0.793 | 27.640 | 1 | 106 | 0.000 |
| Kg | 0.757 | 34.077 | 1 | 106 | 0.000 |
Tab. 3 Wilks' Lambda表3 λ值 |
| 函数检验 | Wilks' Lambda | 卡方(chi-square) | 自由度(degree of freedom) | Sig. |
|---|---|---|---|---|
| 1 | 0.130 | 212.043 | 4 | 0.000 |
Tab. 4 Box's M test results of equality covariance matrixes表4 协差阵相等的Box's M检验结果 |
| Boxe's M | F | |||
|---|---|---|---|---|
| 近似Approx | df1 | df2 | Sig. | |
| 393.053 | 37.563 | 10 | 33501.010 | 0.000 |
Tab. 5 Eigenvalues of canonical discriminant function表5 典则判别函数方差附加特征值 |
| 函数 | 特征值 | 方差的% | 累积% | 正则相关性 |
|---|---|---|---|---|
| 1 | 6.682a | 100.0 | 100.0 | 0.933 |
Tab.6 Classification function coefficients表6 分类函数系数 |
| 变量 | 组别 | |
|---|---|---|
| 分类函数1 | 分类函数2 | |
| Mz | 171.222 | 170.673 |
| Sd | 616.316 | 672.373 |
| Sk | -82.621 | -82.242 |
| Kg | 3252.797 | 3414.969 |
| (常量) | -3193.422 | -3515.928 |
Tab. 7 The verification of the discriminant function表7 判别函数的验证结果 |
| 样品 编号 | 样品数量 | 实际类别 | 判别结果 | 正确率(%) | |
|---|---|---|---|---|---|
| 类别1 | 类别2 | ||||
| XG | 80 | 1 | 79 | 1 | 98.75 |
| ZKT | 215 | 1 | 215 | 0 | 100 |
| QD | 44 | 2 | 0 | 44 | 100 |
Fig. 2 Forecasting group of the Quaternary red clay in the profiles图2 第四纪红土的判别结果 1 上覆黄棕色土与古土壤;2 均质红土;3 弱发育网纹红土;4 典型发育网纹红土;5 判别结果 |
Tab. 8 Forecasting group of the Quaternary red clay表8 第四纪红土的判别结果 |
| 样品名称 | 样品总数 | 判别结果 | |
|---|---|---|---|
| 类别1 | 类别2 | ||
| JJ | 920 | 895 | 25 |
| TX | 147 | 117 | 30 |
| CS | 60 | 48 | 12 |
| NC | 29 | 24 | 5 |
| XY | 29 | 6 | 23 |
| NP | 30 | 0 | 30 |
Tab.9 Forecasting group of dust fall in Wuhu city表9 芜湖春夏季大气降尘的判别结果 |
| 月份 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|
| 样品总数量 | 14 | 17 | 13 | 5 | 14 | 7 | 5 |
| 类别1个数 | 6 | 15 | 2 | 1 | 3 | 1 | 0 |
| 类别2个数 | 8 | 2 | 11 | 4 | 11 | 6 | 5 |
The authors have declared that no competing interests exist.
| [1] |
[
|
| [2] |
[
|
| [3] |
[
|
| [4] |
|
| [5] |
[
|
| [6] |
[
|
| [7] |
|
| [8] |
|
| [9] |
|
| [10] |
[
|
| [11] |
[
|
| [12] |
[
|
| [13] |
[
|
| [14] |
[
|
| [15] |
[
|
| [16] |
|
| [17] |
[
|
| [18] |
[
|
| [19] |
[
|
| [20] |
[
|
| [21] |
[
|
| [22] |
[
|
| [23] |
[
|
| [24] |
|
| [25] |
[
|
| [26] |
[
|
| [27] |
[
|
/
| 〈 |
|
〉 |