Welcome to WEAP's Website WEAP
WEAP is an initiative of the Stockholm Environment Institute.

About WEAP

What's New?
Sample Screens
History and Credits

Using WEAP
User Guide

User Forum
Members List
Edit Profile

Additional Support
University Courses

About Us
SEI-US Water Resources Program
Please Contact Us

Environmental News
Interested in Energy?
Read about LEAP: SEI's software for energy planning.

User Forum

All Topics | Topic "How to model a water source distribution"
Log in to post new messages or reply to existing messages.
Author Message
Mr. Michael Mutz

Subject: How to model a water source distribution   
Posted: 9/27/2017 Viewed: 1726 times
Dear all,

I want to model a villages' water demand with respect to different modes of water supply. We determined the percentages that a certain mode of supply covers through household surveys.
Such a mode might be a waterline, a handpump or a public borewell, although some of them have common primary sources, such as an aquifer.

I tried to set up intermediate demand nodes that calculates its demand via global variables from the village demand node and has zero consumption, i.e.

demand_handpump = demand_village*0.1

where 0.1 represents the fraction of the total village demand covered by handpumps.

If I do a test setup however with one intermediate node that requires a total of 1m^3 per year, consumes none of it, and is linked to the "real" demand node (city) that requires 10m^3 per year, that results in an unmet demand of 0.8 m^3 per year at the city demand node.
How is this possible?

In addition, I get the error message
"Not all the inflow to "mode 1" is consumed but there are not any return flo links to carry away the wastewater. You probably need to create return flow links"
However, it is not possible to connect the intermediate node with the city node via a return flow link. Ho can I set this up properly?
It also does not seem to make a difference if I set the consumption in the intermediate node to 100%, there still is an unmet demand of 0.8m^3 per year instead of 10m^3 at the city.
Screenshots of the setup and its results can be found here:

Thanks alot in advance!

Ms. Stephanie Galaitsi

Subject: Re: How to model a water source distribution   
Posted: 9/28/2017 Viewed: 1690 times
Dear Michael,

Because you've added a sequential order to your demand sites by insisting that the water first go through your intermediate node, you must also tell WEAP that the demand priority of you "city" demand site is 2. When you do that, WEAP will send 1m^3/year to the intermediate node, and it in turn will send 1m^3/year to the city, which I believe represents the fraction of the overall demand that you're simulating.

Are you aiming to connect various supplies with different intermediate nodes that will all send the water along to the city?

You will continue to receive those errors, but they are not concerning in your case because you are not worrying about return flow - just make sure that the Supply Delivered results variable is correct.
Mr. Michael Mutz

Subject: Re: How to model a water source distribution   
Posted: 10/16/2017 Viewed: 1594 times
Dear Stephanie,

Thanks for your answer! The priorities did the trick, thank you!
However, some things remain unclear to me. Say I have a source node with 4 demand sites attached to it, with three of them (As) having priority 1 and the other one (B) having a priority 2.

If there is a shortage and the source cannot satisfy all of the demands of the As, I was under the impression that the water is distributed proportionally among the As:

delivered(A1) = demand(A1)/sum(demand(As)) * water_available(source)

However, if I run a test case scenario, water reaches B. Why is that? I thought lower priority demand nodes only get water if all of the higher priority nodes have been satisfied AND there is still excess water.


Yes I have a groundwater node and a surface water node (we only know that it supplies roughly 2MLD - is using a reservoir the best option for this?) and I want to use intermediate nodes that represent the different modes of supply.

Ms. Stephanie Galaitsi

Subject: Re: How to model a water source distribution   
Posted: 10/16/2017 Viewed: 1549 times
Hi Michael,

For the question of delivery to B, are you by any chance viewing your results as "annual total" or "monthly average" because that may misrepresent the exact time steps were the deficit is split between the A. Remember, all the A demand sites will receive the same *percentage* of their demand, not the same quantity.

If you want to supply roughly 2MLD for all time steps, you could also use "other supply" to model this in WEAP.
Topic "How to model a water source distribution"