Subject: calibration Posted: 10/14/2015 Viewed: 5143 times
I'm a master student. I am using soil moisture method to allocated water resource in basin, I have problem with model calibration.. I don't know how can I calibrate WEAP? and whichone of parameter should be calibrate?
In a paper I saw the result of calibrate was observed and simulation stream flow in stations. I should select a river and the station of it in PEST calibration,I know, but what is the lower bound and upper bound? and after calibrating wich one of output must be cheek? and I don't know wichone of parameter shows the simulation and wichone is observed?
Is it possible to help me? please!
Subject: Re: calibration Posted: 10/14/2015 Viewed: 5137 times
I'm also a master student. Right now, I’m also focusing on calibrate WEAP model, I don't know so much about WEAP calibration, but according to some documents that I already read, usually, some parameters you can be calibrated are:
For catchment: area, crop coefficients, effective precipitation, reference evapotranspiration...
For rainfall and infiltration: runoff/infiltration ratio
For Groundwater: Storage capacity, initial storage, hydraulic conductivity, specific yield, horizontal distance, wetted depth, storage at river level...
I think that it also depends on your catchment to determine what kind of parameters should be calibrated
Lower bound and upper bound in PEST calibration is parameter range, I think it depends on your study area, you need to understand your catchment. In addition, if you do not know so much about your study area, just can try to run PEST such as set up for "effective rainfall" from 0-100% (0 is lower bound and 100 is upper bound)
Hope we can share information to understand more about WEAP calibration process.
Subject: Re: calibration Posted: 10/16/2015 Viewed: 5065 times
Thank for your suggestion.
I hope that we can help together about WEAP.
Subject: Re: calibration Posted: 10/17/2015 Viewed: 5019 times
Hope see sharing from you.
Subject: Re: calibration Posted: 10/19/2015 Viewed: 4993 times
I quote from your reply:
" some parameters you can be calibrated are:
For catchment: area, crop coefficients, effective precipitation, reference evapotranspiration... "
Can you please tell me how to calibrate one of the above parameters.
Should I get the calculation algorithm "for example" for Reference evapotranspiration that is used in the weap to calculate the flow in river and get it into and excel sheet and start changing the equation parameters until the real observed data curve for the flow in river fits the moduled data?
thank you in advanced.
Subject: Re: calibration Posted: 10/19/2015 Viewed: 4989 times
With my understanding about WEAP, I suggest you that:
Firstly, you should try to use PEST calibration, which is an automatic calibration program available in WEAP. It can calibrate many parameters simultaneously, thereby you can save your time and maybe more effectively.
In addition, If you use PEST calibration, you don't need to get the output from the model into excel. You want to compare observed streamflow data and modeled streamflow data? Let’s do it as following:
Set up a stream gauge station in your schematic and input observed data into it.
After that, you run PEST calibration, it will automatically compare observed and modeled data with each other.
Hope it can useful to you!
El Hassani Layla
Subject: Re: calibration Posted: 11/5/2015 Viewed: 4869 times
I have had some experience in calibrating WEAP model. You can do it using PEST calibration or manually. The purpose of calibrating hydrology is to bring observed and simulated values in same level. For that you have to select results tab, then streamflow, then river, then streamgage and the catchment inflow node. The catchment inflow node data (simulated) should match with the streamgage data (observed).
You have to go to the concerned catchment and try to modifie the parameters you have : start with soil water capacity and root zone conductivity then extend (if needed) this to the other parameters you have like kc, prefered flow direction, Z1 and Z2.Start with default parameters and see,once you change a parameter, what is goind to happen in result tab (after running the model).
Once visually you have an acceptable accordance between observed and simulated data, you have to compte some quality parameters like NSE and Bias. For that Just use excel and you can find the equations for computation through internet (if not, I can send you something on that).
I hope this will help you, best of luck.
Subject: Re: calibration Posted: 11/6/2015 Viewed: 4861 times
Dear El Hassani Layla
First of all, thank you so much about your useful suggestion.
I have some questions about calibration for my catchment, hope receives your suggestions.
I divided my catchment into five sub-catchment, and I have two stream gauge stations, one on the upstream which influenced by one sub-catchment, and another one in downstream which influenced by four sub-catchment. I need to calibrate for streamflow at two stream gauge stations. In my opinion, the calibration process at the downstream station is more complex, because I have to adjust many parameters in four sub-catchment that influence it. Can you suggest to me something that makes my calibration process become easier.
I have no groundwater data, so I just use rainfall – runoff method in “catchment” (NOT soil moisture method).
How much big are the 4 sub-catchments in the downstream streamgages ? If they do not have large surface could you consider them as one big-catchment ? and then you will have only one catchment to deal with ?
Subject: Re: calibration Posted: 11/8/2015 Viewed: 4830 times
In my catchment area, the 4 sub-catchment is about 5765 km2. It can consider them as one big-catchment?
El Hassani Layla
Subject: Re: calibration Posted: 11/11/2015 Viewed: 4800 times
Dear Quoc Pham,
I think it may work. Just make sure you are entering right data about land use and kc to calibrate agriculture demand which will be the sum of the demand in the sub-basins.
El Hassani Layla
Subject: Re: calibration Posted: 11/11/2015 Viewed: 4797 times
If you are considering agriclture demand in your basin
Subject: Re: calibration Posted: 11/11/2015 Viewed: 4792 times
I think I still have a problem, if I consider 4 sub-catchments as big sub-catchment. In this region, I have 3 reservoirs like cascade reservoirs along main river. If I did that, when I draw schematic on WEAP, I only have one catchment object. If I assigned that water from this catchment flow into main river in the upstream of the first reservoir (among three cascade reservoir), so it’s not real, the first reservoir will be taken so much water (actually this reservoir just receive water from one sub-catchment, but right now it seems that receive water from 4-subcatchment). How can I solve this problem?
El Hassani Layla
Subject: Re: calibration Posted: 11/11/2015 Viewed: 4782 times
Hmm I see, you have a complexe situation for calibration. I would suggest you to do the following : devide each catchment with a streamgage into upper, middle and low basins. For example if the river located in catchment A have one streamgauge, you devide the catchment into Upper catchment A and low catchment A. If it contains two streamgauges, you devide it into upper catchment A, middle catchment A and low catchment A. Then calibrate by changing parameters in those basins. I hope this new suggestion will works. Good luck
El Hassani Layla
Subject: Re: calibration Posted: 11/11/2015 Viewed: 4778 times
I forget to tell you, if you choose to do that, you have to delete original catchment
Subject: Re: calibration Posted: 11/12/2015 Viewed: 4761 times
I have a little bit confuse, May I send my "WEAP areas" folder to you ? Therefore, you can more understand my issue. And after, you might suggest me effectively.
Subject: Re: calibration Posted: 11/18/2015 Viewed: 4670 times
Dear El Hassani Layla First of all, thank you so much about your suggestion. I have a question. I don't know which node is "catchment inflow node"? Or what's the mean of it?
Can you explain more about it. Thanks a lot..
Subject: Re: calibration Posted: 11/18/2015 Viewed: 4660 times
Hello again Dear layla I have littel confuse. I forget to tell you I have gauge station that influenced by two sub-catchment and then I have two catchment inflow node. I think they are too big to consider them as one catchment. And now I change the soil water capacity and root zone conductivity but it dosen't match with obzerved data. Can you help me?
Thsnks a lot.