Seeking Guidance on Integrating LULC Data into QSWAT+ Model

43 views
Skip to first unread message

Muhammad Nawaz

unread,
Apr 25, 2024, 1:05:29 PMApr 25
to QSWAT+
Dear QSWAT+ User,

I'm seeking guidance on incorporating Land Use Land Cover (LULC) data into my model. I have flow data spanning from 1994 to 2006 for my river basin and a LULC dataset with defined crop types for 2023. My goal is to simulate flow and manage water resources for 2023.

I'm uncertain about the best approach. Should I use the 2000 LULC data initially for model calibration and validation, and then switch to the 2023 dataset? Or is there a different strategy I should consider?

Any relevant documents or resources you could share would be greatly appreciated. Thank you for your assistance.

Best,




natalja....@gmail.com

unread,
Apr 29, 2024, 3:26:28 PMApr 29
to QSWAT+
Hello,
If your goal is to simulate 2023, then use data for 2023.
The earlier dataset can be used to hindcast or to calibrate/validate the model, considering the uncertainty, which will need to be quantified separately. There are many papers on LULC change for SWAT, please use the database to find relevant examples for your study: https://swat.tamu.edu/publications/

Best,
Natalja

Muhammad Nawaz

unread,
Apr 30, 2024, 10:08:06 AMApr 30
to QSWAT+
Hi Natalja,

Thank you for your response. I aim to simulate conditions for 2023, but unfortunately, I lack runoff data for that year. Therefore, I plan to calibrate and validate my model using data from 1994 to 2006, with land use/land cover (LULC) data from 1997. My overarching goal is to analyze water resource distribution in agricultural land, particularly across different crop types, using LULC data from 2023 that includes cropping patterns. Could you advise me on whether it's logical to first calibrate and validate my model and then update the LULC data? 
There's a gap of over 20 years between 1997 and 2023, I see a significant change in land use/land cover during that period.

Best,
Nawaz

natalja....@gmail.com

unread,
May 31, 2024, 2:46:30 PMMay 31
to QSWAT+
Hello  Nawaz,

Because SWAT+ is a process-based model, hence it is logical to calibrate the model on the LULC representative for that period. Otherwise you will be misrepresenting the processes and their functioning.

HTH,
Natalja
Reply all
Reply to author
Forward
0 new messages