تخمین پارامترهای هیدرولیکی منحنی مشخصه شاخه مرطوب شدن خاک با استفاده از حل معکوس فرآیند نفوذ تحت بار افتان توسط مدل HYDRUS-1D

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشیار/ بخش مهندسی آب، دانشکده کشاورزی، دانشگاه شیراز، ایران و پژوهشگر مرکز مطالعات خشکسالی، دانشگاه شیراز، ایران

2 دانشجوی کارشناسی ارشد /بخش مهندسی آب، دانشکده کشاورزی، دانشگاه شیراز، ایران

3 دانشیار/ بخش مهندسی آب، دانشکده کشاورزی، دانشگاه شیراز، ایران

چکیده

تعیین ویژگیهای هیدرولیکی خاک مانند منحنی مشخصه رطوبتی که در شبیهسازی جریان آب و آلاینده کشاورزی و هیدرولوژی مورد استفاده هستند، امری ضروری است. در پژوهش حاضر، دقت تخمین پارامترهای هیدرولیکی منحنی مشخصه شاخه مرطوب شدن خاک با استفاده از حل معکوس فرآیند نفوذ تحت بار افتان توسط مدل HYDRUS-1D مورد بررسی و مطالعه قرار گرفت. برای این منظور، از دو گروه دادههای آزمایشگاهی و میدانی نفوذ تجمعی آب به درون خاک تحت بار افتان استفاده شده است. گروه اول شامل داده‌هایی است که در یک ستون خاک آزمایشگاهی با بافت ماسه ای و با دو رطوبت اولیه متفاوت خاک انجام شده است. گروه دوم شامل داده‌های میدانی است که از اندازه‌گیری نفوذ تجمعی تحت بار افتان و با استفاده از استوانه مضاعف در شرایط سه مزرعه با بافت لوم، لومی شنی و لومی رسی بدست آمد. در هر دو گروه از داده‌ها، ابتدا با استفاده از صفحات فشاری و اندازه‌گیری رطوبت، منحنی‌های مشخصه خاکهای مورد مطالعه به دست آمد. سپس مقدار رطوبت باقی‌مانده، رطوبت اشباع شاخه مرطوب شدن و داده‌های نفوذ تجمعی به عنوان داده ورودی به مدل HYDRUS-1D داده شد و با استفاده از روش حل معکوس سایر پارامترهای هیدرولیکی منحنی مشخصه شاخه مرطوب شدن خاک تخمین زده شدند. مقایسه دادههای منحنی مشخصه اندازهگیری شده و شبیهسازی شده در شرایط آزمایشگاه نشان داد که در هر دو خاک و برای هر دو مدل ون‌گنوختن و بروکز-کوری مقادیر R2 و NRMSE به ترتیب بیشتر از 95/0 و کمتر از 02/0 بدست آمد، که بیانگر این است که مدل بخوبی توانسته است شبیه سازی را انجام دهد. نتایج مشابهی برای شرایط مزرعه نیز حاصل شد. نتایج حاصله بیانگر این است که مدل HYDRUS-1D میتواند با استفاده از روش حل معکوس فرآیند نفوذ تحت بار افتان، پارامترهای هیدرولیکی منحنی مشخصه شاخه مرطوب شدن خاک را با دقت بالایی تخمین بزند.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Estimation of hydraulic parameters of wetting soil water characteristics curve using a reverse solution of falling head infiltration process by HYDRUS-1D model

نویسندگان [English]

  • Fatemeh Razzaghi 1
  • Tayebeh Marzban 2
  • Jahanshir Mohammadzadeh Habili 3
1 Associate Professor, Water Engineering Department, School of Agriculture, Shiraz University, Iran and Researcher, Drought Research Center, Shiraz University, Iran
2 Master students of Water Engineering Department, School of Agriculture, Shiraz University, Iran
3 Associate Professor, Water Engineering Department, School of Agriculture, Shiraz University, Iran
چکیده [English]

Introduction:

Soil hydraulic parameters play an important role in the water cycle and are part of the basic information for designing irrigation and drainage systems, hydrological issues, and soil quality assessment. These parameters can be measured or estimated through direct and indirect methods. Direct measurement requires a lot of money and time, and also due to the large spatial and temporal variability, a large number of samples must be taken for direct measurement under field conditions. An indirect method used to estimate soil hydraulic properties is the inverse solution method, which estimates the physical and hydraulic properties of soil using a numerical solution of the Richards equation. The results of the different researches show that this method has an acceptable accuracy in estimating the hydraulic characteristics of the flow into and out of the soil column, under both laboratory and field conditions. The inverse solution method is cheaper and faster than other methods. One of the advantages of this method is that it estimates the effective parameters and has the ability to simultaneously estimate the hydraulic properties and solute transfer. The HYDRUS-1D model developed by Simonek et al. (2002) in the laboratory, is an advanced model and able to simulate the one-dimensional movement of water, vapor, salts, and heat in soil. This model can simulate the wetting soil water characteristic curve with a very high accuracy by using the cumulative water infiltration data under constant head. Also, this model is capable of simulating infiltration into layered soils under falling head. Moreover, this model can estimate the hydraulic parameters and simulate the characteristic curve of the soil by using infiltration data and inverse solution method. The purpose of this study is to estimate the hydraulic parameters of the wetting soil water characteristic curve using the inverse solution of the infiltration equation under falling load by the HYDRUS-1D model.

Methods:

Laboratory and field experiments were conducted to measure water infiltration into the soil under falling head in two and three soil textures, respectively. For the laboratory experiment, the data of Mohammadzadeh-Habili and Heidarpour (2019) were used, where they measured the falling head infiltration in the transparent acrylic cylinder with an inner diameter of 7 cm with different initial soil water contents. The soil column was filled with sandy texture soils (passed through two different sieves). Another experiment was conducted under field conditions to measure cumulative falling head infiltration using a double ring. The field experiment was conducted in 3 different sites with different soil textures of loam, sandy loam, and clay loam named Kooye Asatid, Shekarbani, and Daneshkade, respectively. In both laboratory and field experiments, the characteristic curves for different soil types were first obtained using pressure plates and moisture measurement. Then the amount of residual moisture, saturated water content of the wetting soil water characteristic curve, and cumulative infiltration data were given as input data to the HYDRUS-1D model, and thereafter, the hydraulic parameters of the wetting soil water characteristic curve were estimated using the inverse solution method. In the HYDRUS-1D model, van Genuchten and Brooks-Corey equations were used for laboratory conditions, and the van Genuchten model was used for field conditions. Finally, the SSQ index was used by the HYDRUS-1D model to optimize the parameters estimated by the model. Also, to determine the accuracy of the HYDRUS-1D model in simulating characteristic curve, the correlation coefficient (R2) and normalized root mean square error (NRMSE) were used.



Results:

For the laboratory experiment, the results showed that the HYDRUS-1D model was able to simulate the characteristic curve with high accuracy using both the van Genuchten and Brooks-Corey equations, but the van Genuchten equation was slightly performed better than the Brooks-Corey equation for both A and B soil textures. The comparison of the measured characteristic curve of Kooye Asatid, Shekarbani, and Daneshkade soils with the characteristic curve simulated by the HYDRUS-1D model showed the effect of hysteresis quite well, as the measured values obtained by drying soil, while the HYDRUS-1D model simulated the wetting front of characteristic curve. In conclusion, a comparison of the values of the simulated parameters with the measured values showed that the HYDRUS-1D model can estimate the hydraulic parameters of the wetting soil water characteristic curve with high accuracy using the reverse solution method of infiltration process under the falling head.

کلیدواژه‌ها [English]

  • Infiltration
  • Falling head
  • Soil water characteristic curve
  • HYDRUS-1D