Reservoirs: Full Pools & Environmental Flows

Recent analysis projects have brought to mind a question pertaining to reservoir storage status and its impact on the higher end of the hydrograph. In other words, when reservoirs are refilling after a significant draw-down and “skimming” or “absorbing” high flows, we see some of our most substantial alterations to the higher end of the hydrograph (see Figure 1).

flow_duration_refill

Figure 1: Flow duration comparison of inflows versus outflows in an impoundment during a refill period. The inflow duration curve is shown in green and the outflow curve is shown in blue.

Conversely, when reservoirs are full, they tend to pass these flows in a manner similar to the natural flow regime (albeit with some dampening due to the nature of reservoir dynamics). Nowadays we are seeing more “percent of inflows” type of management schemes, that is, we require operators to pass-through 70, 80, or even 90% of the inflow on a given day in an attempt to recover storage gradually thereby preserving the shape of the natural hydrograph (see Figure 2).

flow_duration_pct

Figure 2: Flow duration comparison of inflows versus outflows in an impoundment which releases 90% of inflows, therefore mimicking the natural flow regime. The inflow duration curve is shown in green and the outflow curve is shown in blue.

These types of management schemes are great in theory; however, there are a couple of practical obstacles to their implementation:

  1. The sophistication & capacity of the outlet structure, and,
  2. The reasonable level of management that can be expected out of the facility.

Outlet Structure Capacity
The most serious of these issues is likely the capacity of the outlet structure. For example, an impoundment on a medium sized river, say 200 square miles of drainage area, may regularly see peak inflows of over 1,000 cfs, but there actual outlet structure may be limited to passing less than 400 or even 300 cfs. Even those operations that feature hydropower and can pass some larger flows, the turbines are generally sized to be an economical investment for the stream capacity, and thus may have a capacity that is far lower than the regularly seen peak inflows. Also, these structures may lack precision, that is, their ability to pass flows may be such that they can only increase flows in a certain “increment” (for example, turning on another turbine) or with an uncertain level of accuracy (necessitating them to be conservative so as to avoid loosing too much water). Figure 3 shows the actual hydrograph of the 90% release rules from Figure 2. Due to the maximum release rate from the structure, the hydrograph cannot be effectively mimicked above a certain level.

flow_duration_limtations

Figure 3: Flow duration comparison of inflows versus outflows in an impoundment which releases 90% of inflow, but with a maximum release of 250 cfs due to structural limitations. This common situations cause unavoidable flow alterations when the reservoir is allowed to draw down below full. The inflow duration curve is shown in red and the 90% outflow curve in green. The actual outflow curve, based on the limitations of the outlet structure is shown in blue.

Operational Expectations
As water resources become dearer, we begin to manage more small impoundments with greater and greater levels of expectation. However, many of these smaller operations have a low level of manpower, a seasonal focus (such as in agriculture), and/or older infrastructure. It is not uncommon in these settings for it to be impractical to make daily adjustments to flow release rates, with permit conditions that call for weekly release adjustment reflecting this situation. Estimation of inflows to the impoundment on these smaller streams is similarly challenging as they often lack nearby upstream flow gages, or the ability to monitor conditions in their impoundment with a high degree of accuracy. The precision of the outlet structures themselves are generally lower than those in more modern/larger facilities and are therefore even more difficult to control with a high level of precision, and have smaller maximum flow rates.

Full Pool Management
The idea of “full pool” management is not exactly novel. In impoundments used for municipal water supply, there are often goal dates for achieving full pool that coincide with the summer peak season. Many larger managed reservoir systems, with extensive hydro power capacity are already being used to provide elevated minimum releases during spawning seasons with good results. However, due to the reasons mentioned above, we are often in the situation where practical limitations prevent our ability to pass high flows that resemble the natural inflow regime. In these cases, we may find that our best strategy is simply to try to keep the pool full. As noted above, when the reservoir pool is full and high flows occur, the reservoir is spilling these inflows (minus whatever withdrawals are occuring) resulting in an outflow regime that is very close to that of the natural inflow. This may be especially true in smaller impoundments, whose backwater effects are less pronounced, and which therefore pass inflows with little impedance (sometimes referred to as “short-circuiting”).

Opportunities for “Full Pool” Management
In the mid-Atlantic and southeastern regions of the United States, we commonly see our highest flows in late winter and early spring, and these flows often coincide with the spring-time spawning period for native fish. These spring inflows, even during non-storm conditions, can be much larger than capacity of the dam outlet structures. If we can craft release rules during the winter months to maximize the probability of entering the spawning period with a full pool, we can achieve a much more natural flow regime during the subsequent days and weeks. For seasonal agricultural impoundments this can result in virtually no loss of flows, since irrigation withdrawals have yet to begin and evaporation is low. Figure 4a and 4b are show the difference predicted by modeling 2 different sets of release rules for an agricultural impoundment on a small headwater stream. Note that in this case, the same amount of water is impounded and consumed during the year; however the pattern of flow-alteration during the target period is drastically different.

time_series_notfull

Figure 4a: The simulated inflows and outflows of an impoundment that has NOT achieved full pool before a large flow event. The minimum release is shown in green, and the downstream flows are essentially static until refill is achieved.

time_series_full
Figure 4b: The simulated inflows and outflows of an impoundment that has achieved full pool just before a large flow event. The minimum release is shown in green, but because the pond is full, the outflow (orange) is essentially equal to inflow (blue).

Summary
As the demands on our hydrologic resources intensify, we are looking for ways to meet off-stream needs while imposing a minimal disturbance on the natural flow regime. These two goals are often at odds with one another. The ability of “full-pool” management to simplify operational demands and simultaneously mimic the natural flow regime more faithfully may have benefits to both operators and downstream aquatic resources. However, reducing winter release rates to achieve full pool for a spring spawning period may not be advantageous in every situation – the risk of reduced winter flows has to be outweighed by the gains made for downstream spawning fish. Also, we have to consider the size of the impoundment relatives to the size of the watershed – clever management of resources does not create water, but rather maximizes our efficiency in its use. If models and or operation realities find that it is too difficult to achieve refill by the end of winter, we must realize that the impoundment or demands placed upon it are too large for the contributing watershed to sustain, and demands need to be restricted. Nevertheless, as we attempt to manage more and more of our impacts to the aquatic system, we will need to look for ways that are practicable given the realities of the technical infrastructure and logistical capacity of operators. For the special case of small rivers that need significant spring spawning flows and can accommodate significant diversion of winter flows, “full-pool” management may be an important tool.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s