Subject: HELP please Posted: 11/4/2017 Viewed: 14394 times
I did not understand exactly the mechanism in design WEAP model
For example, I have river flow data for the period 1994-2016. Do I design the whole model first by entering the demand and supply data and all the sites of demand and supply and then calibrate
Or just enter the flow data then calibrate?
What is the historical model? For example, did the 1994-2015 data for calibration be used so that 1994 was considered the base year(current account)?
When I entered 1994 base year, how will I enter the demand and supply sites and data?
Did you just enter the base year data only in 1994 and then calibrate? Or should the year-water method be used in the reference scenario before calibration between observed and simulated data?
Please help
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/6/2017 Viewed: 14374 times
Dear Rouba,
I advise you to first think about where your streamflow gauge (that gave you the streamflow data) is located, specifically, is it upstream or downstream of the area you are modeling? It could also be in the middle of the area. Basically, before you can calibrate your model, you must be enter all the data that is upstream of the streamflow gauge, because that observed data should match the calibrated historic data of your model. That's how you know that the model upstream is producing/consuming water in a way that is representative of the actual system. That means you'll have to data and/or assumptions for upstream demands and supplies.
A historic model is used for calibration. For your data, the Current Accounts year would be 1994, and the model would extend until 2016. Whenever you are ready to start modeling scenarios you can change the years of your model. If you change the Current Accounts year, be sure to consider initial conditions.
If you're using the Water Year Method, it will be difficult to calibrate your model very well because the Water Year Method isn't designed to represent historic streamflow, it's designed to produce very simplistic possibilities of future streamflow. When we calibrate historic models against observed streamflow, we tend to use catchments and precipitation records to model hydrology. I recommend finding a period when you have data for both precipitation and streamflow records, and creating a historic model for that period, which you can calibrate.
Eng. rouba ali
Subject: Re: HELP please Posted: 11/6/2017 Viewed: 14356 times
My streamflow gauge (that gave my the streamflow data) is located,is downstream of my area
Eng. rouba ali
Subject: Re: HELP please Posted: 11/6/2017 Viewed: 14348 times
Is there another way to calibrate the historical model than
?to use the catchment method and precipitation
Note that my study area is a one watershed
Please reply
In other words, can I introduce monthly river flows in 1994 (the current year
of accounts)or entre flow data in many years and then compare modeled and observed
Or do you enter the flow data in the current account year by (read from file
Then compare the modeled flows with the observed flows in the results
Or should use the method of catchment?
please reply
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/6/2017 Viewed: 14337 times
Dear Rouba,
Really the only way to calibrate streamflow is to use a hydrological model (one of the catchment methods in WEAP). If your streamflow gauge is downstream of your study area, you need to have information about the upstream area (supplies and withdrawals) to "sum" up to the observed data for the streamflow gauge. The withdrawals would be any demands, and the supply would be the water that percolates into the river from precipitation plus any return flows from the demands (there are other possible supplies as well, but these are usually the most important ones).
You can enter the flow data for the Current Accounts year, but WEAP will repeat this data for the entirety of the model, unless you use the Water Year Method, which will either make the monthly flow greater or smaller according to a limited number of percentages. Neither of these techniques are detailed enough to allow a good calibration of the model. Plus, if your data comes from streamflow downstream, you should not be entering it as the data for streamflow upstream. Those will be very different, according to the other elements in the system.
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/6/2017 Viewed: 14331 times
Dear Rouba,
I have an amendment: you can enter data for the headflow if you have a separate hydrological modeling that you are using to generate the streamflow upstream. Those model outputs could be entered into the headflow for the WEAP model, and then compared with the gauge data downstream. In that case, calibration would involve either changing your demand data (if uncertain) or your hydrological model.
Eng. rouba ali
Subject: Re: HELP please Posted: 11/6/2017 Viewed: 14326 times
Do you mean that it is enough to enter the flow data for the base year and then compare the modeled flows with the
?observed flows at the downstream gauge site in results
Or should enter, for example, flow data for several years, for example (flow data 1994-2000) and then I compare Modeled flows with the observed flows where the gauge
please explain
and can I use that without having to use the catchment
method?
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/7/2017 Viewed: 14288 times
Dear Ruba,
The only way you could enter flow data into the WEAP model and use it for calibration is if you have a hydrological model elsewhere that is generating that flow data. It is the parameter values of the hydrological model that change based on calibration, so if you do not have a hydrological model, you cannot calibrate. In WEAP, the hydrological model is within the catchments.
[If you have two stream flow gauges, upstream and downstream, and you are trying to calculate the sum of demand and supply in the distance separating them, you can do that, but that is not calibrating streamflow.]
In terms of base year vs. many years, calibration requires many years. You would want to build the historic model for the time period with an overlap of both gauge data and precipitation data available.
Eng. rouba ali
Subject: Re: HELP please Posted: 11/7/2017 Viewed: 14281 times
Thanks very much for the reply
I want to ask if I used aRainfall Runoff (simplified coefficient method) method in modeling
Should I enter the crop coefficient and evaporatranspiration
?in the data if is not available
Or is it enough to precipitation and flow data in the base year only
Please explain how it works
My other question is if the calibration was successful after using the catchment model, then creating the future model. Can I use the water year method in the future model where I use the catchment method in the historical model
I will use the historical model 1994-2015 and the future model 2016-2050
Is that correct?
please reply
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/7/2017 Viewed: 14274 times
Dear Rouba,
You should enter the data for all the parameters, if you have it. If you do not have it (which is very likely because some parameters are abstract and difficult to measure), those parameters should become part of your calibration process.
You will need to enter records of historic precipitation and historic flow data into your model, and you should enter it for as many years as possible. You can upload the CSV files containing the data into the base year (current accounts year), but in order to calibrate you will need to have CSV files with several years of data.
You cannot use the Water Year Method with a catchment method, you have to choose either one or the other to generate model headflow in your river.
For your years 1994-2015 for your historic model, you will need to have precipitation records and flow records. If you do, those years for the historic model make sense.
Eng. rouba ali
Subject: Re: HELP please Posted: 11/8/2017 Viewed: 14268 times
Dear Stephanie
can the period 1972 to 1978 has been used for calibration and the period of 1972
to 1992 has been used to generate the temperature pattern for the period of 2010 to 2030 in order
to assess its impact on water available.
؟Is that correct
I've seen it in a master's thesis
Could a historical period be used as observational values and compared with a future period as modeling values for calibration?
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/9/2017 Viewed: 14250 times
Dear Rouba,
I'm not sure what you mean when you say using a historical period as observational values.
It's fine to use the period 1972-1978 for calibration, although a more recent time period would be preferable. It could be difficult to know what sort of changes have happened since 1972 that could impact the model.
Eng. rouba ali
Subject: Re: HELP please Posted: 11/17/2017 Viewed: 14130 times
I have become confused
You said:
1-If you're using the Water Year Method, it will be difficult to calibrate your model very well because the Water Year Method isn't designed to represent historic streamflow, it's designed to produce very simplistic possibilities of future streamflow. When we calibrate historic models against observed streamflow, we tend to use catchments and precipitation records to model hydrology.
2-Really the only way to calibrate streamflow is to use a hydrological model (one of the catchment methods in WEAP)
3-It is the parameter values of the hydrological model that change based on calibration, so if you do not have a hydrological model, you cannot calibrate. In WEAP, the hydrological model is within the catchments.
4-You cannot use the Water Year Method with a catchment method, you have to choose either one or the other to generate model headflow in your river.
So what is the use of the water year method and how it is used?
And also if i want to study the impact of climate change in the future. what should I do? If I had used the catchment method in my project
Please explain and clarify things
Eng. rouba ali
Subject: Re: HELP please Posted: 11/19/2017 Viewed: 14090 times
if I used the catchment method, where I can use the water year method.and if I used the catchment method in my project and could not use the water year
?method, how can I study the impact of climate change in the future
please reply
Ms. Stephanie Galaitsi
Subject: Re: HELP please Posted: 11/20/2017 Viewed: 14066 times
Dear Rouba,
You cannot use the Water Year Method with the catchment method. They are different methods and cannot be combined in WEAP.
You can either:
1) Use the catchment method, calibrate your model using historic data, and then project future climate conditions in your catchment, either by obtaining climate projections from GCMs, or by repeating your historical time series climate data into the future while incorporating gradual changes such as a 2 degree temperature increase in 50 years or a 4 degree increase in 50 years. These are just examples. For the reference scenario, I like to repeat some length of historical time series climate data so I can ask the question "what would my basin be like if the climate stayed the same" and compare that future against a climate change scenario.
2) Using the Water Year Method, you can enter a single year of data in the Current Accounts and then study variations (floods/droughts) of that year in the future, using that exact seasonal pattern. You cannot calibrate using the Water Year Method.
