1 2 Direct withdrawals are limited to major rivers and lakes. - - PDF document

1

2

Direct withdrawals are limited to major rivers and lakes. Minimum in-stream flow requirements are more than drought period flows, so safe yield without storage is zero. Withdrawals from an existing reservoir is a reservoir alternative, not a direct river withdrawal.

3

On-stream reservoirs capture water from within the reservoir’s watershed and use stored water for supply.

4

Review bullets Depending upon the physical setting and the location of the demand center, these projects can be operated as direct withdrawals from the reservoir or for flow augmentation releases to a downstream diversion. If the reservoir releases into a much larger receiving stream, the project’s yield can be significantly enhanced.

5

Water withdrawn directly form the reservoir generally provides high quality source water because it is pre-settled and and the watershed can be monitored for spills and releases. As an example, Greensboro, North Carolina’s Lake Townsend project is a direct- withdrawal, on-stream reservoir (with a recently added emergency backup diversion). The Lake Townsend project is a 6.3 billion gallon reservoir on about a 105 sq. mi. watershed. It has a safe yield of about 35 mgd.

6

However, if the reservoir releases to a fairly large watershed with enough flow to meet needs during normal to high flow periods, augmentation operation of the reservoir can be quite beneficial. Water supply needs can to drawn from the larger receiving stream when flows are high and water supply releases need be provided from the reservoir only when the receiving stream flows are insufficient to meet those needs and the receiving stream in-stream flow requirements. Therefore, more storage can be conserved for use during droughts, thereby increasing safe yield yield.

7

Most modern water supply reservoirs are constructed on smaller watersheds and provided with diversions from a larger nearby stream or river. Nearly all diversion reservoirs are supplemented by pumping. Occasionally, an adjacent watershed is located at a higher elevation, allowing a gravity flow diversion.

8

Review bullets

9

Discuss bullets Similar to on-stream reservoirs, these projects can also be operated as direct

• withdrawals. In these circumstances, natural runoff is supplemented with diversions

and water supply is taken directly form the reservoir.

10

In the right setting, they can also be effective as augmentation reservoirs with releases to a downstream water supply diversion. As discussed previously, if the reservoir releases into a much larger receiving stream, the project’s yield can be significantly enhanced. Like Greensboro’s Lake Townsend Reservoir, Cobb-Marrietta Water Authority’s Hickory Log Creek project is a 6 billion gallon reservoir. Unlike Lake Townsend, it is

• n an 8 sq. mi. watershed and receives up to 39 mgd in pumped diversions from the

600 sq. mi. Etowah River (located immediately downstream). This project provides augmentation releases to an intake located further downstream on the Etowah

• River. Since water supply releases are only made when flows in the Etowah are

low, this project yields nearly 50 mgd.

11

To illustrate some of the permutations and combinations considered, I’ll touch on Carroll County, GA’s proposed Indian Creek Reservoir Project. The county (outlined in green) is divided into two major river basins. To avoid interbasin transfer issues, the Carroll County Water Authority has planned for separate supplies to meet needs in each basin. For the Tallapoosa, the primary demand center is to the northeast, near I-20 and closest to Atlanta. The best reservoir site is located in the northwest corner of the county It’s a 4 8 sq The best reservoir site is located in the northwest corner of the county. It s a 4.8 sq.

• mi. watershed that can accommodate 10 Bg of water supply storage. The major

long-term diversion to the reservoir is the Little Tallapoosa River – far enough downstream to have a large watershed and far enough away from the Alabama state line to address interstate concerns. Since the reservoir is located well downstream of the demand center, an augmentation reservoir would not serve the county. Therefore, this project needs to be a direct withdrawal that sends water back to the eastern demand center.

12

Here’s the whole project in schematic format. This shows only the western end of the county. The proposed reservoir would be located in the headwaters of Indian

• Creek. For initial operations - up to 6 mgd – a 10 mgd diversion would be located

downstream on Indian Creek. However, Indian Creek can’t provide much more yield, so the second and third phases of the project - 6 mgd each in safe yield – would require additional diversions from the Little Tallapoosa River (10 mgd and 20.5 mgd increments). This project’s yield is 18 mgd despite 40 mgd of diversions and 10 Bg of storage because Georgia, like many states, bases yield on the drought of record. The Tallapoosa River was hit with a very severe drought in 2007 and 2008. Feb. 2007 to Feb. 2008 flows were 54% of the previous minimum 365 day average for the prior 55 years of record.

13

Now let’s take a look at some interesting planning process considerations.

14

You know that you are in a drought once reservoir storage becomes sufficiently depleted to command your attention. Also, it’s impossible to know if conditions will improve or get worse, so water managers must assume the latter. Since supplies can’t be planned, permitted and developed in time to help, unless supplies are ample, demand management is the water managers only viable tool. With our evolving awareness of climate change and not knowing if the current drought is going to be the “big one”, I recommend ample supplies.

15

It is imperative to plan ahead, well ahead. …

16

17

Environmental regulators and advocacy groups want to maintain high levels of flow throughout the year. For a given safe yield, the higher the in-stream flow standard, the larger the reservoir and the larger the diversion pumps. Also, stream and wetland impacts weigh heavily in project selection and mitigation costs are very high. Also, higher in-stream flows require bigger reservoirs which increase stream and wetland impacts. It can get incredibly complex, so finding and retaining the very best team (based on track record) is a wise investment.

18

Because of the time and cost and complexity of these projects … Review bullets From initial planning to operation can take from 10 to more than 20 years. Very few projects approach the lower limit. Let’s tune up the time machine!

19

That’s right, it’s 1961. We’ve mastered time travel. Gunsmoke is the #1 TV show, Alfred Hitchcock Presents is in its last year of production and for you folks from Georgia, Hank Aaron and the Braves are still in Wisconsin, along with a 20 year-old catcher named Joe Torre. Take a minute to tell me what your world looks like, and think about how accurate you could project service conditions to appropriately meet water supply needs in 2011?

20

“To better understand the context and difficulty of the challenge, let’s reflect for a moment on our world of 50 years ago”

21

Having put ourselves in a 1961 perspective, would anyone have been able to picture today’s world? Would we have estimated that the state of North Carolina would more than double its population in 50 years or that Georgia would grow to more than 2.5 times its 1961 population. Remember that phones still had wires and many of them still needed an operator interface to work. Most people would have laughed at the concept of a ‘personal’ computer and there would have been no doubt that our climate was firmly fixed

• computer. … and there would have been no doubt that our climate was firmly fixed

within the bounds of weather cycles we had experienced. Brace yourselves – we’re going Back to the Future.

22

Here we are back in 2011, and now we’re charged with planning for a water supply to meet needs through 2061. How accurately can we define what our grandchildren’s world will look like and what its needs will be.

23

Fast forwarding and looking from our current perspective … to plan well, we need to get a handle on a number of issues and develop educated guesses for others. I’d be 114, so it doesn’t matter much to me. It will be critical for our younger people and for the many people not yet born!

24

We all recognize that long-term water supply planning necessarily includes measures of uncertainty. The recognized uncertainties most commonly play out in population projections, per capita demand and industrial demand forecasts. Many regulators and project opponents jump on these issues because they know that the numbers can readily be questioned. In recent years, the opposition has attacked each of these elements and fought to minimize each.

25

We also have unrecognized or, at the least, unspoken uncertainties – climate and streamflow records are at the top of the list.

26

What has not been recognized as entailing uncertainty is the supply side – “weather is variable but climate is fixed”. It is tacitly accepted within the regulatory and project opposition camps that the past 50 years of hydrologic data fully captures the range of climatological events to occur over the next 50 years (planning period). What is most interesting is that these are the same folks that most readily support man-made global warming. However, in this case, applying that argument leads to bigger dams and bigger pumps, so the issue is suppressed. If climate dynamics is discussed, requests for detailed analytical justification are common, knowing that contingencies and factors of safety are the only currently available approaches.

27

Regulators and independent project opposition now scrutinize every consideration in the planning process and vocally demand that each be minimized. Integrating a set of minimized considerations results in output that is likewise minimized and unable to meet planning period needs.

28

For the Indian Creek Reservoir in Carroll County, Georgia, the drought of record yield steadily declined through the 1990s and then took a major hit in the past few

• years. None of us can tell whether the Tallapoosa River flows reflect climate

change or erratic longer term hydrologic behavior that expresses itself with greater frequency as the amount of available data grows. The reason for the decline may be arguable. What isn’t arguable is the volatility of climate and the considerable risk posed by reliance on 30 to 60 year streamflow

• records. With a growing awareness of climate volatility, whether natural or

manmade, such records are coming to be recognized as inadequate for capturing longer terms climatic cycles. If human activity is measurably rebalancing the climate equation, it further compounds the difficulty in projecting available water supplies 50 years into the future.

29

Each of these graphs represent safe yield over time for a fixed set of project facilities and operating conditions, and each had at least 30 years of record in the early 1950s. West Georgia, where the Indian Creek site is located, has had the most extreme droughts of late. The only site showing reasonable resistance to significant decline in safe yield with an extended flow record is a proposed reservoir which would be located in the mountains of North Georgia,. The trends for the other plotted projects vary, with the overall trend (red line) ominously downward. The decline could be approaching a steady state, but given the linearity of the average decline, some measure of yield deterioration is expected to continue into the future. This provokes serious concerns about the veracity of currently accepted yield y y y analysis procedures, but more importantly, it implies that the actual safe yield of Georgia’s existing water supply reservoirs may be significantly less than the permitted values on file. There is also no reason to suspect that this trend isn’t

• perating over a much larger geography.

30

In looking at a plot of current vs. 1982 yields for the projects plotted on the previous slide, the West Georgia phenomena shows very clearly. But it also begs the question of why that particular region has been so hard hit? Up to about 10 years ago, the streamflow records in West Georgia and Central North Georgia (Atlanta Metro Area) had been remarkably similar. There are no developmental,

• rographic, geologic, geomorphic or other significant factors or

considerations to explain differences in recent drought intensity. The only bl l i i th t th t d i d i tl fl ti reasonable conclusion is that the recent record is predominantly a reflection

• f the chance distribution of precipitation. An Atlanta area drought of West

Georgia magnitude (2007-2009) appears to be no less likely than what

• ccurred in West Georgia.

31