|
|
The identification of bedrock types based on soil chemical composition |
Jia-Yi WANG, Li-Bo HAO, Xin-Yun ZHAO, Cheng-You MA, Ji-Long LU, Yu-Yan ZHAO, Qiao-Qiao WEI |
College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China |
|
|
Abstract Geological mapping of areas with shallow overburden generally has less bedrock outcrop to work with, and is thus characterized by poor accuracy. Soils formed by weathering of rocks have significant inheritance of chemical composition from the bedrocks. In view of such a situation, the authors propose a soil chemical composition based multi-layer perceptron neural network model to recognize bedrock types. Taking Alongshan area in the northern part of the Da Hinggan Mountains as an example, the authors used geochemical data of major and lithophile trace elements of soil samples overlying volcanic bedrocks to identify bedrock types, and identified 4 types of bedrocks, i.e., basalt, andesite, dacite and rhyolite. The results show that the prediction accuracy of the model in the identification of bedrock types in the shallow overburden area of the Da Hinggan Mountains reaches up to 90%. The authors have reached the conclusion that soil chemical composition based multi-layer perceptron neural network model used for the identification of bedrock types has the advantage of high convenience, high speed and efficiency, and can provide an effective way for improving the geological mapping quality in areas with shallow overburden.
|
Received: 13 August 2018
Published: 19 December 2018
|
|
|
|
|
|
|
|
|
指标 | 玄武岩类(n=21) | 安山岩类(n=47) | 英安岩类(n=38) | 流纹岩类(n=61) | 最小值 | 最大值 | 平均值 | 最小值 | 最大值 | 平均值 | 最小值 | 最大值 | 平均值 | 最小值 | 最大值 | 平均值 | SiO2 | 43.04 | 53.82 | 49.38 | 54.12 | 66.90 | 60.60 | 67.04 | 71.78 | 69.22 | 72.09 | 86.25 | 75.52 | Al2O3 | 10.21 | 19.24 | 15.95 | 13.82 | 18.78 | 16.63 | 12.64 | 17.83 | 15.83 | 8.77 | 16.03 | 13.26 | TFe2O3 | 6.36 | 20.50 | 10.26 | 3.23 | 9.54 | 5.81 | 1.85 | 6.83 | 3.23 | 0.96 | 3.31 | 1.79 | K2O | 0.65 | 3.72 | 1.84 | 1.36 | 7.12 | 3.60 | 2.17 | 7.01 | 4.50 | 0.18 | 7.16 | 4.38 | Na2O | 1.71 | 4.07 | 3.05 | 1.37 | 7.22 | 4.14 | 0.58 | 5.93 | 4.00 | 0.17 | 8.38 | 3.76 | CaO | 1.43 | 12.64 | 5.92 | 0.13 | 6.01 | 2.63 | 0.10 | 3.66 | 0.74 | 0.07 | 17.0 | 0.53 | MgO | 1.08 | 16.72 | 4.86 | 0.27 | 6.29 | 2.01 | 0.09 | 2.00 | 0.70 | 0.03 | 0.85 | 0.31 | Ti | 2248 | 29451 | 10552.7 | 2699 | 9704 | 5938.4 | 1460 | 4565 | 3170.2 | 474 | 3503 | 1536.3 | Mn | 555 | 3131 | 1404.3 | 402 | 2588 | 939.2 | 230 | 2052 | 809.5 | 93.8 | 1439 | 475.8 | P | 173 | 4922 | 2083.3 | 211 | 3155 | 1613.0 | 138 | 1514 | 726.7 | 62.4 | 797 | 242.9 | Cu | 0.52 | 63.6 | 22.2 | 0.48 | 69.6 | 14.8 | 0.70 | 8.63 | 21.6 | 0.18 | 13.1 | 2.9 | Pb | 2.85 | 31.0 | 11.7 | 1.69 | 80.8 | 20.4 | 9.08 | 50.3 | 21.6 | 5.71 | 201 | 28.6 | Zn | 59.7 | 100 | 103.6 | 34.2 | 143 | 82.5 | 42.0 | 159 | 67.3 | 18.4 | 105 | 49.1 | Co | 15.6 | 62.0 | 36.3 | 4.26 | 33.2 | 17.6 | 0.21 | 13.3 | 6.99 | 0.02 | 9.10 | 2.93 | Ni | 2.42 | 102 | 33.25 | 1.05 | 55.8 | 15.79 | 2.03 | 20.7 | 6.32 | 1.98 | 28.40 | 6.41 | V | 123 | 414 | 236.2 | 9.22 | 181 | 104.5 | 6.90 | 74.6 | 26.5 | 2.49 | 42.7 | 15.5 | Rb | 13.1 | 139 | 59.0 | 21.9 | 294 | 108.9 | 53.6 | 214 | 119.0 | 1.46 | 272 | 141.1 | Sr | 128 | 1525 | 570.1 | 20.2 | 1892 | 557.8 | 14.0 | 743 | 249.3 | 9.85 | 336 | 83.2 | Ba | 71.7 | 1793 | 625.0 | 122 | 2325 | 945.9 | 119 | 1655 | 933.6 | 48.5 | 1483 | 557.4 | Nb | 0.96 | 20.8 | 12.0 | 0.83 | 68.6 | 12.8 | 2.82 | 77.2 | 20.1 | 0.83 | 94.6 | 19.4 | Zr | 23.1 | 322 | 193.9 | 85.2 | 1649 | 286.2 | 190 | 1157 | 382.4 | 85.4 | 655 | 257.0 | Y | 5.81 | 43.5 | 20.6 | 7.87 | 74.4 | 17.8 | 9.1 | 99 | 25.5 | 3.34 | 129 | 19.4 | Th | 0.46 | 12.7 | 6.64 | 1.52 | 28.5 | 14.16 | 8.61 | 37.60 | 19.27 | 4.95 | 39.1 | 24.12 |
|
|
指标 | 玄武岩类(n=21) | 安山岩类(n=47) | 英安岩类(n=38) | 流纹岩类(n=61) | 最小值 | 最大值 | 平均值 | 最小值 | 最大值 | 平均值 | 最小值 | 最大值 | 平均值 | 最小值 | 最大值 | 平均值 | SiO2 | 47.70 | 70.05 | 61.09 | 46.88 | 71.64 | 63.83 | 56.74 | 76.30 | 67.53 | 59.13 | 79.58 | 67.83 | Al2O3 | 13.89 | 18.82 | 15.90 | 12.08 | 18.50 | 15.52 | 10.10 | 17.37 | 14.77 | 11.03 | 18.85 | 15.15 | TFe2O3 | 4.45 | 11.33 | 7.37 | 3.80 | 8.69 | 5.65 | 2.18 | 10.38 | 4.86 | 2.32 | 6.71 | 4.27 | K2O | 1.34 | 2.98 | 2.17 | 1.46 | 3.98 | 2.63 | 1.62 | 4.14 | 2.67 | 1.72 | 5.69 | 2.91 | Na2O | 1.04 | 2.57 | 1.64 | 1.22 | 3.26 | 1.98 | 0.92 | 3.22 | 1.69 | 0.48 | 4.90 | 1.77 | CaO | 0.59 | 5.19 | 1.49 | 0.55 | 4.95 | 1.30 | 0.50 | 1.93 | 0.82 | 0.32 | 1.54 | 0.63 | MgO | 0.85 | 4.90 | 2.01 | 0.69 | 4.24 | 1.49 | 0.41 | 1.95 | 1.01 | 0.42 | 1.75 | 0.89 | Ti | 4811 | 11009 | 6908.5 | 4007 | 9939 | 5909.7 | 3803 | 7562 | 5457.9 | 1349 | 7901 | 4685.8 | Mn | 392 | 2515 | 966.0 | 269 | 2433 | 853.6 | 280 | 5625 | 919.9 | 219 | 4432 | 687.9 | P | 368 | 2971 | 864.7 | 315 | 3052 | 961.5 | 285 | 2560 | 768.7 | 152 | 1177 | 507.3 | Cu | 5.64 | 62.0 | 18.3 | 6.32 | 31.8 | 15.7 | 4.10 | 21.8 | 10.5 | 1.62 | 50.3 | 10.1 | Pb | 8.78 | 37.7 | 21.3 | 3.14 | 125 | 26.4 | 4.43 | 127 | 26.0 | 6.53 | 338 | 35.3 | Zn | 69.6 | 189 | 113.2 | 44.4 | 162 | 98.9 | 57.1 | 494 | 115.0 | 48.7 | 736 | 112.9 | Co | 10.4 | 34.3 | 21.9 | 8.8 | 60.5 | 18.2 | 6.33 | 22.5 | 13.0 | 0.54 | 17.7 | 10.1 | Ni | 2.56 | 62.4 | 21.5 | 2.51 | 33.9 | 17.2 | 4.60 | 1403 | 52.2 | 3.43 | 33.8 | 14.7 | V | 60.6 | 380 | 141.1 | 18.0 | 164. | 97.6 | 31.5 | 128 | 75.2 | 14.4 | 113 | 65.1 | Rb | 62.8 | 143 | 103.8 | 52.1 | 201 | 116.3 | 11.0 | 165 | 118.1 | 61.2 | 221 | 128.8 | Sr | 119 | 431 | 233.0 | 141 | 884 | 306.5 | 66 | 384 | 178.5 | 34 | 268 | 131.6 | Ba | 471 | 704 | 600.7 | 475 | 964 | 658.3 | 444 | 1006 | 662.9 | 108 | 997 | 579.9 | Nb | 7.40 | 36.3 | 18.8 | 2.16 | 64.3 | 20.2 | 8.53 | 56.5 | 21.1 | 9.12 | 55.9 | 22.1 | Zr | 95.5 | 430 | 263.3 | 180 | 371 | 277.8 | 167 | 978 | 309.8 | 190 | 589 | 309.9 | Y | 5.52 | 45.0 | 21.2 | 9.85 | 41.2 | 18.9 | 12.8 | 54.2 | 22.6 | 12.6 | 43.2 | 21.1 | Th | 0.61 | 19.1 | 10.6 | 1.95 | 28.1 | 12.4 | 2.27 | 35.3 | 12.5 | 5.86 | 21.8 | 14.1 |
|
|
[1] |
刘德鹏, 丁峰, 汤正江 . 区域化探在森林沼泽区地质填图应用初探[J]. 物探与化探, 2004,28(3):209-212,217.
|
[2] |
时艳香, 纪宏金, 郝立波 , 等. 利用水系沉积物地球化学数据判别浅覆盖区岩性与构造—欧氏距离法[J]. 物探化探计算技术, 2004,26(3):243-246.
|
[3] |
Ji H J, Zeng D M, Shi Y X , et al. Semi-hierarchical correspondence cluster analysis and regional geochemical pattern recognition[J]. Journal of Geochemical Exploration, 2007,93(2):109-119.
|
[4] |
郝立波, 陆继龙, 李龙 , 等. 区域化探数据在浅覆盖区地质填图中的应用方法研究[J]. 中国地质, 2007,34(4):710-715.
|
[5] |
时艳香, 郝立波, 陆继龙 , 等. 因子分类法在黑龙江塔河地区地质填图中的应用[J]. 吉林大学学报:地球科学版, 2008,38(5):899-903.
|
[6] |
田密 . 水系沉积物低弱地球化学异常提取方法研究[D]. 长春:吉林大学, 2017.
|
[7] |
Zhao X Y, Hao L B, Lu J L , et al. Origin of skewed frequency distribution of regional geochemical data from stream sediments and a data processing method[J]. Journal of Geochemical Exploration, 2018,194:1-8.
|
[8] |
徐国志, 徐锦鹏, 段玲玲 . 化探资料在地质填图中的应用[J]. 物探与化探, 2015,39(3):450-455.
|
[9] |
王大勇, 郝立波, 陆继龙 . 人工神经网络在识别浅覆盖区地质体中的应用[J]. 吉林大学学报:地球科学版, 2006,36(S2):185-187.
|
[10] |
郝立波, 蒋艳明, 陆继龙 , 等. 利用多目标地球化学数据识别第四纪沉积物类型——基于概率神经网络方法[J]. 吉林大学学报:地球科学版, 2008,36(6):1081-1084.
|
[11] |
蒋宗礼 . 人工神经网络导论[M]. 北京: 高等教育出版社, 2001.
|
[12] |
郝立波, 陆继龙, 马力 . 浅覆盖区土壤化学成分与基岩化学成分的关系及其意义——以大兴安岭北部地区为例[J]. 中国地质, 2005,32(3):477-482.
|
[13] |
郝立波, 马力, 赵海滨 . 岩石风化成土过程中岩石均一化作用及机理——以大兴安岭北部地区为例[J]. 地球化学, 2004,33(2):131-138.
|
[14] |
邱家骧 . 岩浆岩岩石学[M]. 北京: 地质出版社, 1985.
|
[15] |
李曦 . 神经网络信息传输函数Sigmoid与tanh比较论证[J]. 武汉理工大学学报:交通科学与工程版, 2004,28(2):312-314.
|
[16] |
Tavakolipour H, Mokhtarian M . Neural network approaches for prediction of pistachio drying kinetics[J]. International Journal of Food Engineering, 2012,8(3):1-15.
|
[17] |
连克强 . 基于Boosting的集成树算法研究与分析[D]. 北京:中国地质大学(北京), 2018.
|
[18] |
Sharkey A J C . Boosting using neural networks[G]//Combining artificial neural Nets. London: Springer, 1999: 51-78.
|
[19] |
于玲, 吴铁军 . 集成学习: Boosting算法综述[J]. 模式识别与人工智能, 2004,17(1):5259.
|
[1] |
WANG Zhi-Qiang, YANG Jian-Feng, WEI Li-Xin, SHI Tian-Chi, CAO Yuan-Yuan. Geochemical characteristics and bioavailability of selenium in alkaline soil in Shizuishan area, Ningxia[J]. Geophysical and Geochemical Exploration, 2022, 46(1): 229-237. |
[2] |
ZHAO Xiao-Yuan, YANG Zhong-Fang, CHENG Hui-Yi, MA Xu-Dong, WANG Jue, LI Zhi-Kun, WANG Chen, LI Ming-Hui, LEI Feng-Hua. Geochemical characteristics and ecological health-related ranges of Copper in soil in Huaying Mountain-Xicao in Linshui County, Sichuan Province[J]. Geophysical and Geochemical Exploration, 2022, 46(1): 238-249. |
|
|
|
|