Irrigation Parameterization Confusion

[Evan]

Hello all,

I am using the irrigation parameters (SWAT2012) for the first time, and had some questions I couldn't find in the manuals or the forum (please let me know if I've missed something obvious). I used USGS data to get irrigation water use for groundwater and surfacewater, and then calculated/ estimated water use per subbasin. Now I am trying to parameterize these numbers and I'm not sure I understand how the model wants the input.

1) In the Management input file>Edit Management Parameters>General Parameters, the Irrigation can be 'turned on' and several parameters can be defined. However, under the Operations tab, there is another suite of parameters for 'auto-irrigation' that seem similar. I tried just adding irrigation under the 'General Parameters' tab, and had no change in model output. I am assuming that an irrigation operation needs to be added to the .mgt operation schedule. So I guess my question is: why are the 'general parameters; there, and what are they used for? If I schedule an irrigation operation, are the 'General Parameters' even used?

2) Also, since there are surface and groundwater withdrawals within a subbasin (reach and groundwater), is it possible to have the model irrigate from both water sources within the same watershed, perhaps by adding a second operation?

Thanks for any advice,

-Evan

[Evan]

As an update, I had some good advice and went ahead and started parameterizing water withdrawls for the model. However, a new question has arisen. If HRUs can only have one irrigation water withdrawal (one source), and there are groundwater and surface water withdrawals in a subbasin, should the irrigation be split between ground/ surface water sources, or be combined and a single source used? My current thought is that the resulting changes in stream flow (which is the main focus of this model) will be impacted similarly regardless of source (reach vs shallow aquifer), and therefore combining them will be the most efficient way of inputting this information into the model.

I'd appreciate any thoughts or advice.

-Evan