
黄河下游冲积平原试验区土壤水分特征曲线的传递函数研究
湛江, 刘颜珲, 王琳, 潘登, 蔡金龙
黄河下游冲积平原试验区土壤水分特征曲线的传递函数研究
Study on Pedo-Transfer Functions of Soil Water Characteristic Curve in an Experimental Area of Alluvial Plain in the Lower Yellow River
黄河下游冲积平原区是我国重要的粮食生产基地和工业集聚地,该地区土壤水力学参数的获取,对于保障用水安全和指导农业生产具有重要意义。为建立黄河下游冲积平原区的土壤水分特征曲线传递函数(PTFs),以河南省兰考县闫楼乡作为黄河下游冲积平原试验区,基于多元非线性逐步回归和单因素扰动方法,建立了包气带土壤水分特征曲线的PTFs,并分析了影响因素敏感性。研究结果显示,实测土壤水分特征曲线以及土壤理化性质参数呈现较强变异性,所建立的PTFs精度良好,具备应用推广价值。多元回归结果表明,土壤颗粒组成是本文PTFs的主要影响因素,其中黏粒含量对PTFs最为敏感,砂粒含量次之,粉粒含量最弱。在其余5个土壤理化特性中,对PTFs的相对敏感的因素是pH值和分形维数。研究表明,土壤物理化学性质与土壤水分特征曲线模型参数的关系复杂,并非一般线性或单调关系。尽管土壤物理化学性质参数可以提高PTFs精度,但土壤颗粒组成是影响土壤水分运动的根本原因,其作为构建PTFs的关键因素不可忽略。在实际应用中,PTFs使用者可根据所掌握数据的实际情况,结合影响因素敏感性,决定影响因素的取舍。
The alluvial plain area in the lower reaches of the Yellow River is an important grain production base and industrial agglomeration in China. The acquisition of soil hydraulic parameters in this area is crucial for ensuring water safety and guiding agricultural production.To establish the PTFs of soil water characteristic curves in the alluvial plain area of the lower Yellow River, Yanlou Township, Lankao County, Henan Province, is selected as a representative area in the alluvial plain of the lower Yellow River. Based on multiple nonlinear stepwise regression and single factor perturbation methods, the PTFs of soil water characteristic curves in the vadose zone were established, and the sensitivity of influencing factors was analyzed. The research results show that the measured soil water characteristic curves and soil physicochemical parameters exhibit strong variability. The PTFs established in this research demonstrate good accuracy, which have application and promotion value. The multiple regression results indicate that soil particle composition is the main influencing factor of PTFs in this study. The clay content is the most sensitive, followed by sand content, and the silt content is the weakest. Among the other five soil physicochemical properties, the relatively sensitive factors are pH value and fractal dimension. This study indicates that the relationship between soil physicochemical properties and soil water characteristic curve model parameters is complex. It is not a general linear or monotonic relationship. Although soil physical and chemical property parameters can improve the accuracy of PTFs, soil particle composition is the fundamental reason affecting soil water movement, and cannot be ignored as a key factor in the constructing of PTFs. In practical applications, PTFs users can determine the choice of influencing factors based on the actual situation of the data they have mastered, combined with the sensitivity of influencing factors.
黄河下游冲积平原 / 土壤水分特征曲线 / 土壤传递函数 / 多元逐步回归 / 分形维数 / 参数敏感性 {{custom_keyword}} /
alluvial plain in the lower reaches of the Yellow River / soil water characteristic curve / pedo-transfer function / multiple stepwise regression / fractal dimension / parameter sensitivity {{custom_keyword}} /
表1 测试项目和方法Tab.1 Test items and methods |
测试项目 | 测试方法或装置 | 装置型号或参考规范 |
---|---|---|
土壤水分特征曲线 | 压力膜仪 | 500 kPa和1 500 kPa压力膜仪 |
土壤粒径分布 | 激光粒度仪 | QT-2012型激光粒 度仪 |
干容重(BD) | 环刀法 | 土工试验方法标准(GB/T 50123-2019) |
总孔隙度(TP) | 烘干法 | 土工试验方法标准(GB/T 50123-2019) |
pH值 | pH检测计 | 农业行业标准(NY/T1121.6-2006) |
有机质含量(OM) | 重铬酸钾滴定法 | 农业行业标准(NY/T1121.6-2006) |
电导率(EC) | 电导率测定仪 | 农业行业标准(NY/T1121.6-2006) |
表2 土壤理化特性参数的经典统计学特征Tab.2 Classical statistical characteristics of soil physicochemical parameters |
参数 | 最小值 | 最大值 | 均值 | 标准差 | 变异系数 |
---|---|---|---|---|---|
黏粒含量/% | 0 | 91.28 | 8.82 | 18.20 | 2.063 5 |
粉粒含量/% | 0.05 | 99.95 | 55.03 | 34.91 | 0.634 4 |
砂粒含量/% | 0 | 99.95 | 36.38 | 39.18 | 1.077 0 |
分形维数 | 1.046 | 2.470 | 1.867 | 0.401 | 0.215 0 |
干容重/(g•cm-3) | 1.23 | 1.68 | 1.43 | 0.09 | 0.063 4 |
总孔隙度/% | 32.30 | 67.00 | 45.17 | 4.73 | 0.104 7 |
pH值 | 7.80 | 9.70 | 8.68 | 0.34 | 0.039 2 |
有机质含量/(g•kg-1) | 0.57 | 31.20 | 5.75 | 5.41 | 0.940 9 |
电导率/(μS•m-1) | 8.67 | 1142.20 | 246.83 | 203.14 | 0.823 0 |
表3 不同吸力段主要基质势下含水率的经典统计学指标Tab.3 Classic statistical indicators of water content under main matrix potentials in different suction stages |
基质势/cm | 统计学指标 | 表层 | 第2层 | 第3层 | 第4层 |
---|---|---|---|---|---|
40 | 最小值 | 0.392 2 | 0.352 8 | 0.448 4 | 0.333 0 |
最大值 | 0.666 5 | 0.474 3 | 0.657 5 | 0.478 7 | |
均值 | 0.503 6 | 0.432 9 | 0.537 9 | 0.383 4 | |
变异系数 | 0.124 9 | 0.065 3 | 0.097 4 | 0.062 9 | |
200 | 最小值 | 0.126 5 | 0.063 9 | 0.226 4 | 0.110 6 |
最大值 | 0.651 5 | 0.371 3 | 0.629 4 | 0.263 7 | |
均值 | 0.345 9 | 0.251 2 | 0.407 6 | 0.192 4 | |
变异系数 | 0.357 3 | 0.282 1 | 0.233 0 | 0.179 1 | |
12 000 | 最小值 | 0.041 3 | 0.031 3 | 0.052 3 | 0.016 9 |
最大值 | 0.293 7 | 0.111 0 | 0.414 4 | 0.067 9 | |
均值 | 0.105 8 | 0.065 3 | 0.130 7 | 0.036 4 | |
变异系数 | 0.519 7 | 0.266 9 | 0.553 8 | 0.305 3 |
表4 van Genuchten模型参数的经典统计学特征Tab.4 Classical statistical characteristics of van Genuchten model |
参数 | 最小值 | 最大值 | 均值 | 标准差 | 变异系数 |
---|---|---|---|---|---|
θr /(cm3•cm-3) | 0.011 1 | 0.265 2 | 0.056 0 | 0.032 7 | 0.584 4 |
θs /(cm3•cm-3) | 0.352 6 | 0.680 0 | 0.487 0 | 0.070 5 | 0.147 7 |
参数α/cm-1 | 0.001 7 | 0.014 2 | 0.008 1 | 0.002 5 | 0.310 4 |
参数n | 1.215 0 | 3.307 3 | 1.883 3 | 0.388 5 | 0.206 3 |
表5 建模集和验证集的经典统计学特征Tab.5 Classical statistical characteristics of the modeling and validation sets |
数据集 | 样品数 | 参数 | 最小值 | 最大值 | 均值 | 标准差 | 变异系数 |
---|---|---|---|---|---|---|---|
建模集 | 175 | θr /(cm3•cm-3) | 0.011 1 | 0.265 2 | 0.059 1 | 0.035 7 | 0.604 1 |
θs /(cm3•cm-3) | 0.361 6 | 0.680 0 | 0.496 0 | 0.077 0 | 0.155 3 | ||
α/cm-1 | 0.001 7 | 0.014 2 | 0.008 0 | 0.002 8 | 0.348 1 | ||
n | 1.232 1 | 3.307 3 | 1.896 2 | 0.396 5 | 0.209 1 | ||
验证集 | 58 | θr /(cm3•cm-3) | 0.016 3 | 0.079 3 | 0.046 7 | 0.017 8 | 0.382 1 |
θs /(cm3•cm-3) | 0.352 6 | 0.582 7 | 0.469 0 | 0.060 4 | 0.128 7 | ||
α/cm-1 | 0.006 1 | 0.010 5 | 0.008 5 | 0.001 1 | 0.126 5 | ||
n | 1.215 0 | 2.778 9 | 1.844 4 | 0.356 5 | 0.193 3 |
表6 逐步回归方程及其精度评价Tab.6 Stepwise regression equation and its accuracy evaluation |
参数 | 逐步回归方程 | RMSE | R 2 |
---|---|---|---|
θr /(cm3•cm-3) | | 0.015 9 | 0.454 |
θs /(cm3•cm-3) | | 0.031 7 | 0.839 |
α/cm-1 | | 0.001 0 | 0.406 |
n | | 0.321 4 | 0.504 |
表7 含水率验证指标的经典统计学特征Tab.7 Classic statistical characteristics of moisture content validation indicators |
指标 | 最小值 | 最大值 | 均值 | 标准差 | 变异系数 |
---|---|---|---|---|---|
RMSE | 0.006 856 | 0.134 984 | 0.036 148 | 0.028 385 | 0.785 2 |
R 2 | 0.941 7 | 0.999 3 | 0.987 6 | 0.013 685 | 0.013 857 |
1 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
2 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
3 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
4 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
5 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
6 |
胡振琪,张学礼. 基于ANN的复垦土壤水分特征曲线的预测研究[J]. 农业工程学报, 2008, 133(10): 15-19.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
7 |
李彬楠,樊贵盛. 基于灰色理论-BP神经网络方法的土壤水分特征曲线预测模型[J]. 干旱区资源与环境, 2018, 32(7): 166-171.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
8 |
高鹏飞,冉卓灵,韩珍,等. 含岩屑紫色土水力特性及饱和导水率传递函数研究[J]. 土壤学报, 2021, 58(1): 128-139.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
9 |
赵红光,樊贵盛,于浕,等. 基于BP神经网络的Gardner模型参数预测[J]. 节水灌溉, 2017(10): 22-25, 30.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
10 |
秦文静,樊贵盛. 基于粒子群优化算法-支持向量机的原状黄土Van Genuchten模型参数土壤传输函数[J]. 干旱区资源与环境, 2020, 34(11): 132-137.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
11 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
12 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
13 |
聂春燕,胡克林,邵元海,等. 基于支持向量机和神经网络的土壤水力学参数预测效果比较[J]. 中国农业大学学报, 2010, 15(6): 102-107.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
14 |
孙志祥,邓建波,吕玉娟,等. 长江上游低山丘陵区土壤水分特征曲线传递函数研究[J]. 灌溉排水学报, 2022, 41(6): 97-104.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
15 |
张昭,程靖轩,刘奉银,等. 基于颗粒级配参数描述砂土持水及非饱和强度特性的土壤转换函数[J]. 水利学报, 2020, 51(4): 479-491.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
16 |
廖凯华,徐绍辉,程桂福,等. 基于不同PTFS的流域尺度土壤持水特性空间变异性分析[J]. 土壤学报, 2010, 47(1): 33-41.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
17 |
姚姣转,刘廷玺,王天帅,等. 科尔沁沙地土壤水分特征曲线传递函数的构建与评估[J]. 农业工程学报, 2014, 30(20): 98-108.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
18 |
王子龙,常广义,姜秋香,等. 灰色关联及非线性规划法构建传递函数估算黑土水力参数[J]. 农业工程学报, 2019, 35(10): 60-68.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
19 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
20 |
李浩然,樊贵盛. 土壤水分特征曲线Gardner模型参数预报研究[J]. 人民黄河, 2019, 41(4): 149-152, 158.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
21 |
左炳昕,查元源.基于机器学习方法的土壤转换函数模型比较[J]. 灌溉排水学报, 2021, 40(5): 81-87.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
22 |
廖凯华,徐绍辉,吴吉春,等. 基于PCR和ANN构建的土壤转换函数的适用性研究[J]. 灌溉排水学报, 2014, 33(1): 17-21.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
23 |
李彬楠,樊贵盛,申丽霞. 基于BP神经网络方法的黄土水分特征曲线预测模型比选[J]. 中国农村水利水电, 2023(1): 171-175. LI B N, FAN G S, SHEN L X. Comparison of loess water characteristic curve prediction model based on BP neural network method[J]. China Rural Water and Hydropower, 2023 483 (1):171-175.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
24 |
崔俊芳,邓建波,刘传栋,等.若尔盖高寒草甸表层土壤水分特征曲线传递函数研究[J]. 山地学报, 2021, 39(4): 483-494.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
25 |
仝长水,靳孟贵,王献坤,等. 黄河故道兰考段地下水水化学特征[J]. 工程勘察, 2011, 39(12): 36-41.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
26 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
27 |
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.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
28 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
29 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
30 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
31 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
32 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
33 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
34 |
秦文静,樊贵盛. 冲洪积平原土壤低吸力阶段水分特征曲线影响因素研究[J]. 节水灌溉, 2019(10): 38-42.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
35 |
王志涛,缴锡云,韩红亮,等. 土壤垂直一维入渗对VG模型参数的敏感性分析[J]. 河海大学学报(自然科学版), 2013, 41(1): 80-84.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |