Regarding Issues with SWAT-MODFLOW Coupling Results

[Min Jing Li]

Hi, all

I have encountered the following problems in my simulations:

Q1: During the SWAT-MODFLOW mapping phase, in normal simulations, only two grid cells display exchange, and this exchange is limited to the river domain. However, theoretically, there should be multiple grid cells with exchange. Do you have any suggestions on how to address this issue?

Q2: The recharge values in the "swatmf_out_MF_recharge_yearly" file seem unusual, as depicted in the attached figure. They do not accurately simulate the alluvial fan area I am working on. This situation is perplexing to me. Have you ever encountered a similar situation, and do you have any advice to offer?

I sincerely appreciate your time in reading this problem, and I am eagerly looking forward to receiving your valuable professional insights, which would greatly benefit my research.

Min-Jing, Li

[Seonggyu Park]

Hi,

I believe there might be an issue with the "mf_grid (MODFLOW)" layer. Please check the layer, following this link.

https://github.com/spark-brc/QSWATMOD2-plugin#add-or-subtract-rows-and-columns

Best,

Seonggyu

[Min Jing Li]

Hi, Seonggyu

Q1: However, I used ArcGIS to create it, and I have verified that the MODFLOW grid matches the SWAT-MODFLOW grid in the overlapped areas, as shown in the figure. So, what other potential issues could there be?

Q2: The exchange of surface and groundwater in my model only occurs in two grid cells, whereas normally it should involve the entire river. What could be the possible reasons for this?

Sincerely,

Min Jing

env...@gmail.com 在 2023年10月11日 星期三晚上9:28:31 [UTC+8] 的信中寫道：

[Seonggyu Park]

Dear Min Jing,

Did you link the models using QSWATMOD or ArcGIS?

Thanks,

Seonggyu

[Min Jing Li]

Dear: Seonggyu

It's ArcGIS. (Using  SWAT-MODFLOW_Tutorial_version3)

Thanks,

Min Jing

env...@gmail.com 在 2023年10月26日 星期四上午11:43:35 [UTC+8] 的信中寫道：

[Seonggyu Park]

Hi Min Jing,

It appears there may be an issue with your linkage files. Could you kindly share your SWAT-MODFLOW model inputs, linkage table files, and GIS files with me? Once I have these, I'll take a look and see what I can do to help.

Best,
Seonggyu

[Min Jing Li]

Hi, Seonggyu

When you replied to my last email, I tried to use QSWATMOD 2.6 in QGIS 3.32.1. The issue with the garbled characters has been resolved, but I have encountered new problems. Here are my questions:

Q1: When I try to add RT3D to SWAT-MODFLOW, I encounter the following error and cannot generate ICBOUND, even though the RT3D layer has been created.

Q2: In the "Hydrographs for streamflow and hydraulic head" section, I am encountering an issue as shown in the attached figure. 

I have created the "modflow.obd" file, but when I click on the display option, I receive the error message displayed in the figure. I would like to understand which part of the process is causing this problem.

Below is all the data for my QSWATMOD package. I would greatly appreciate your assistance in resolving these issues.
https://drive.google.com/file/d/1zjzsfU7KiFzZ4RKxAymgzrVr4wmzy-Pv/view?usp=sharing
I look forward to receiving your response as soon as possible.

Thank you

Min Jing

env...@gmail.com 在 2023年11月8日 星期三上午11:52:16 [UTC+8] 的信中寫道：

[Seonggyu Park]

Hi Min Jing,

I will review your project regarding RT3D option and get back to you soon. For Q2, please upgrade to QSWATMOD version 2.7.2 using the link (https://github.com/spark-brc/QSWATMOD2-plugin/releases/tag/v2.7.2) and make the necessary adjustments to the observation file format. Additional details can be found here (https://github.com/spark-brc/QSWATMOD2-plugin/discussions/16).

Best,

Seonggyu

[Min Jing Li]

Hi Dr. Seonggyu,

Regarding version 2.7.2, I have encountered an issue where the water level column in the output graph displays the y-axis in units of water flow rate (m³/day) rather than the water table (m), as illustrated in the attached video. 

I would be immensely grateful if you could spare a moment to examine this issue and offer any advice or solutions.

Additionally, I am eagerly awaiting your response regarding the RT3D option we previously discussed. 

Your insights and expertise in this area are highly valued, and I am looking forward to your valuable input on this matter.

Thank you very much for your time and assistance.

Your help and expertise are greatly appreciated.

Thank you

Min Jing

env...@gmail.com 在 2023年11月22日 星期三下午2:20:43 [UTC+8] 的信中寫道：

[Seonggyu Park]

Hi Min Jing,

The current version of QSWATMOD2 doesn't support to build a multi-layer RT3D model. You'll have to construct it manually. Regarding the high groundwater level problem, it typically pertains to the structure and parameters of your MODFLOW model from both SWAT and MODFLOW. I recommend examining the water balance in the *.lst file (MODFLOW output file) and then adjusting certain parameters (such as CN2 and Sol_Z from SWAT, and SS, SY, HK from MODFLOW).

Best regards,
Seonggyu

[Min Jing Li]

Hi Dr. Seonggyu,

I am writing to ask about the capabilities of SWAT-MODFLOW3.exe in relation to RT3D simulations. I question whether this software supports multi-layer simulation in RT3D or is it limited to simulating only a single layer?

Additionally, I am curious if it's possible to simulate a single layer in RT3D while using a multi-layer setup in MODFLOW. I have successfully run SWAT-MODFLOW alone, but I encounter issues when trying to link a manually created single-layer RT3D with SWAT-MODFLOW. Upon running SWAT-MODFLOW3.exe in this configuration, the simulation fails. Attached, you will find my RT3D files and swatmf_log file. I would greatly appreciate any insights or suggestions you might have on this matter.

Furthermore, I would like to clarify a point regarding the latest version of QSWATMOD. It appears that in the water level charts, the Y-axis represents flow rates instead of water levels.

Thank you for your time and assistance. I look forward to your valuable input.

Best regards,

Minjing

env...@gmail.com 在 2024年1月20日 星期六凌晨4:32:43 [UTC+8] 的信中寫道：

[Seonggyu Park]

Hi Minjing,

1. The SWAT-MODFLOW model can perform a multi-layer RT3D simulation if you input multi-layer RT3D data, and the number of layers should match the MODFLOW model layers. Please refer to the SWAT-MODFLOW tutorial documentation for more details: https://github.com/spark-brc/QSWATMOD2-plugin/blob/main/docs/SWAT-MODFLOW.Tutorial_v3.pdf

2. Thank you for reporting the issue. I will check and resolve it in the upcoming version.

Best,

Seonggyu