Subject: generate a random set of reservoirs Posted: 3/16/2012 Viewed: 14557 times
Is it possible to have WEAP automatically create a large set of in-stream reservoirs (say 1000 of them) in a certain area? One way to do it is to draw them one by one, but it will be very time consuming and tedious. Is there a better way to do it?
Thank you very much!
Subject: Re: generate a random set of reservoirs Posted: 3/16/2012 Viewed: 14556 times
No, it is not possible to automate the creation of objects on the Schematic, including reservoir. However, you might want to rethink the need for so many reservoirs. Would it be possible to combine many smaller reservoirs into a few larger reservoirs?
Subject: Re: generate a random set of reservoirs Posted: 3/22/2012 Viewed: 14510 times
Let share what is your idea for building a model with 1000 dams. I can guess something, but let us know.
Subject: Re: generate a random set of reservoirs Posted: 3/22/2012 Viewed: 14504 times
The reservoirs that I was referring to are small reservoirs that generally have a surface area at FSL of a few hectares. They are tiny and are primary for irrigation/animal watering. They are common in rural Asia and Africa.
Jack has suggested that it can be modeled in WEAP by combining them into a few large reservoirs. But in my research, I have found that it would substantially decrease evaporation if I do that in the semi-arid region of west Africa where evaporation is very high. I ran the model in matlab and found that if I have 100 reservoirs that have the same total storage capacity as those 1000 ones, the resulting evaporation loss from the surface of the reservoirs is 50% of that would result from 1000 ones. But even 100 reservoirs is many to model. The total storage capacity of those 1000 reservoirs is equivalent to about 5 of a large irrigation scheme in the area of my study. The resulting evaporation loss of those 5 reservoirs is only about 5% of the total amount from the 1000 individual ones.
1000 dams is one of the scenarios that I modeled. Actually, this number is not too many for the White Volta Basin, which is the river basin that I have studied. That is about the current density of small reservoirs in one region of Ghana. People there want more small reservoirs built there. Therefore, I modeled 5000 of them for a future scenario. So my conclusion is that it is quite hard to model this many dams in WEAP and, therefore, WEAP is not the right program to use in this case.
Subject: Re: generate a random set of reservoirs Posted: 3/22/2012 Viewed: 14503 times
Why would evaporation be so much lower if the surface area of the fewer combined reservoirs was the same as the sum of the smaller reservoirs? In defining the Volume-Elevation curve for the aggregate reservoir, make sure that the elevation is the same as the average of the many smaller reservoirs. For example, if the sum of the volume of the 1000 reservoirs is 20 million cubic meters (MCM) with an average elevation of 3 meters, the volume-elevation point for the aggregate reservoir would be 20 MCM and 3 meters.
Subject: Re: generate a random set of reservoirs Posted: 3/22/2012 Viewed: 14500 times
Thanks! That's a good suggestion. Let me think it over and I will let you know what happens.
Subject: Re: generate a random set of reservoirs Posted: 3/22/2012 Viewed: 14493 times
Now I get your point. It's interesting.
I think that there are three important things in your model:
1- Reservoir volume
3- Allocation from this reservoirs
So, what if we think that these tiny reservoirs are stream? you know that in WEAP you can define Evaporation for each reaches, hence the second problem can be solved.
As WEAP is a water balance model (like the other models), the first problem already was solved.
And for the third, you can just keep tie with priorities.
In Iran, specially north of Iran, farmers get water from main stream and then keep them on farm dams/ponds. during non irrigation season they have fishery productions, and in summer, they use ponds as a water reservoir. All of these dams/ponds are off-stream dam, but actually they are a part of stream. So, my suggestion can work (I hope!).
instead of defining 1000 dams, you can define 1000 stream or branch.
But dear Xuan... I think sometimes in water resources proper defining of a project/watershed is more important than using software. Software are just tools. you MAY do everything that WEAP does, in an Excel sheet, but it is bloody time consuming, and sometimes impossible. If I were you, re-define my research. You may run MATLAB or any Mathematical Optimisation software, but are they really define your goal?
Well... what do you think?
honestly, I am a WEAP addicted guy! :D