Restriction curves - SRG

Within Source, Restriction Curves are an approach to controlling water supply within a model and are used to reduce the water volumes released to users when the volume in the storage is declining. For instance, if dam water volumes drop below 60%, restrictions on water supply may be imposed for a period of time to reduce the use of a declining resource. Restrictions are usually lifted when dam water volumes exceed another dam volume higher than the volume at which the restriction was applied. In this way a particular restriction is not lifted until there is some significant recovery in stored water volume above the volume at which the restriction was applied. 

The approach is particularly relevant to urban water supply but can also be used to trigger events such as releases from a certain storage, desalination plant supply rates or environmental flows based on salinity levels. The aim of the trigger/restriction curves is to provide a time series that tracks what level of restriction the system is in. This time series can then be used to modify some other variable such as demand or inflow (desalination) as well as being a measure of the standard of service delivered. 

Restriction Curves provide water management rules that aim to avoid the situation where all water in the storage is used up and aim to extend the period when a certain minimum supply can be ensured. Users can be informed about these rules in advance so that all stakeholders know what to expect under the range of possible conditions of water supply. Restriction levels are best estimated using a long historical record of daily inflow rates and water storage levels and volumes.

The Restriction Curves approach relates Restriction Levels to a Return Value, which can then be used to modify the model. For example:

• At Restriction Level 0, water volume released for a water user is equal to the original allocation or entitlement (no change, return value = 1)
• At Restriction Level 1, water volume released to a user will be reduced to 65% of the original allocation or entitlement (return value = 0.65)
• At Restriction Level 2, water volume released to a user will be reduced to 40% of the original allocation or entitlement (return value = 0.4)
• And so on … where increasing restriction levels are applied as the stored water volume decreases to lower and lower volumes

Trigger values are entered as Rising and Falling values for each restriction level. An example set of Restriction Curves implemented in Source is shown in Figure 1. In this case, the trigger values are the water volume expressed as a proportion of the stored volume at full supply level. Note that Rising and falling refer to the change in the stored water volume not the Restriction Level. For example:

• When stored water volumes are declining and the proportion of stored water to full supply volume becomes equal to or less than 0.6 for the preceding day, the Restriction Level 0 changes to the more severe Restriction Level 1 (Figure 1, blue dotted line). Hence, a Falling restriction curve lists the monthly trigger values for entering that restriction level from a lower, less severe restriction level.  
• When the proportion of stored water to full supply volume rises to 0.7 or above for the preceding day the Restriction Level returns from 1 to 0 (Figure 1, solid red line). Hence, a Rising Restriction Curve lists the monthly trigger values for entering that restriction level from a higher, more severe restriction level.  

The falling Restriction Curve for Restriction Level 0 is redundant and is not used (there is no less severe restriction than Level 0).

Figure 1. Example Restriction Curves

Figure 1 also shows that the Falling and Rising trigger values for switching Restriction Levels can change on a monthly basis. For example, lower levels in a water storage might be tolerated at the end of a dry season because of water use through this season – and an upcoming wet season is expected to provide inflows to the water storage and reduction in water demand.

An example consistent with the restrictions above is provided in Figure 2, which shows that the change in Restriction Level (shown on the y axis on the right) goes from Level 0 to Level 1 at 60% storage and from Level 1 to Level 2 at 45% storage as storage volumes are getting lower and then returns from Level 2 to Level 1 at 55% storage and Level 1 to Level 0 at 70% storage.

Figure 2. Example model using Restriction Levels