New EPIC user: Switchgrass yield for Connecticut

[Adam Gallaher]

Hello, 

I am a PhD student new to EPIC, I downloaded winEPIC v6 and am looking to assess switchgrass yields for Connecticut. I have a few questions about weather data setup and farm setup. Currently, I have data for ten weather stations across Connecticut. Will EPIC simulate for the whole state with this data or only for those ten locations? Also, when running the model we are asked to select a "farm", how does one run the model for multiple "farms"? 

Sorry if these questions are generic, this is my first time using a crop model. 

Thank you in advance, 

Adam 

[Luca]

Dear Adam,

Welcome to the EPIC/APEX user forum.

Starting with the weather data, I think the main question is how well the ten weather stations represent the weather across Connecticut? If they represent well the weather variability across the state, you might be able to use them to do your simulation across the entire state. If they are not, the results of the simulation will be affected and probably the quality of the results will be reduced for areas where the weather is not well represented by any of the ten weather stations available.

To run multiple farms, you have to create one simulation for each farm you want to simulate. Then, from the main window click on "run batch" (it should be the button in the middle), in the following windows select to show all the points available, in the next window select all the farms you want to simulate and you should be good to go.

Let me know if you need more information.

Best,

Luca

[Adam Gallaher]

Hello Luca, 

Thank you for getting back to me so quickly. I think that is a good question and after some consideration I think that they would, the idea being that I am looking to estimate crop yield for specific locations in Connecticut not the entire state. Unless my line of thinking is off and it would be important to know the weather conditions across the whole state. I would then ask the same question about the soils, would I need to have data on the soils for the entire state to have a successful simulation or just data on soils for the specific locations that I am interested in within the state. 

Between posts, I was able to run a test simulation using the default data and have a better understanding of how to build multiple models, thank you! 

I appreciate your time. 

Adam

[Luca]

Adam,

since your goal is to estimate the crop yield for specific locations in Connecticut, you do not need to have weather and soil data for the entire state. So, for both soil and weather you will need the information to describe the characteristics of the areas you want to simulate. It should be "easy" for the weather and one weather station for each location will be enough. The same weather data can be used for several locations if they are in the same general area with the same weather. The same is valid for the soil considering that the soil variability can be greater than the weather variability but, if you are simulating a specific field where more than one soil can be identified, you can run the model using the dominant soil.

Let me know if you need more information.

Best,

Luca

[Adam Gallaher]

Hello Luca,

Thank you for clearing that up for me, this was very helpful. I have one more question as of now, that is what would be the optimal time period to optimally model crop yield for a given area, 1 years worth of weather data, 4 years worth? The dominate soil was also a good suggestion, thank you again for all your help! If I think of anymore questions I will reach out. 

All the best, 

Adam 

[Luca]

Adam,

When collecting the weather data, I suggest getting the longest period possible. Along with the weather data, you will need information on how the crop was managed during the period you want to simulate (information like planting and harvesting date, amount, type, and date of fertilizer used (if any), tillage operations, etc.).

Then, for the period you want to simulate, you will need observed data. You can have direct observations from the field you want to simulate (for instance crop yield or leaf area index), or you can have information like the average yield in the region where your study area is located. Then, you have to consider that observed data is needed for the calibration and validation of the model (this is an extensive topic, I can give you more information if needed or you can read about it, I am sure there are many resources available). So, to give you a final number, I consider one year not enough unless you have several observed data (for instance multiple daily observations in one year). If you are going to use only crop yield to calibrate/validate the model, I would say that three years (three observations) is the minimum to be able to do some statistical analysis to evaluate the model results.

I hope this is not confusing you. Let me know if you have more questions.

Best,

Luca

[Adam Gallaher]

Hello Luca, 

This was very helpful and has given me a lot to think about. I will need to do some more research to see if there are plots that I can validate against. Currently, to my knowledge, there are no switchgrass plots in Connecticut. I might be wrong however. What I might end up doing is what you suggested, taking averages for the region and using those. 

Thank you again for all your help. 

Adam