More rainfall, but where is it all going?

Things are starting to look up for the Savannah River Basin as each of the sub-basins received above average rainfall for the third straight month.

Russell led the charge in June, netting 7.2 inches compared to its 3.8-inch average. (This was Russell’s fourth best in the past 69 years; the most for June came in 1989 at 8.21 inches.)

Thurmond and Hartwell received 6.6 and 5.7 inches on 3.8- and 4.8-inch averages, respectively.

These numbers are especially encouraging as the precipitation has come at a time of year not normally associated with heavy rainfall.

So with all this rain – for instance, the Hartwell sub-basin has collected 6.4 inches above its average in the last three months – why haven’t the reservoirs fully recovered?

The answer is complicated.

For starters, the basin is still recovering from one of the worst droughts in its history. In 2016, the sub-basins had the worst or second worst single year for rainfall on record.

In addition, the effects of the drought were muted due to the torrents of rain we received at the end of 2015. Essentially, that rainfall carried us through the first six months of the drought.

So even though the basin started receiving below-average to severely below-average rain from January to July 2016, the water table and reservoir levels were still being buoyed by the rain we received from October through December 2015.

Finally, timing is everything.

Much in the way that the winter 2015 rainfall delayed the drought’s effects, the drought is prolonging the time it takes for the water table and reservoirs to recover.

In addition, the excess rainfall we’ve been receiving is now being “taxed” by transpiration and evaporation, so a smaller percentage is actually making it to the reservoirs. This effect happens every summer, but was exacerbated by the drought.

Thurmond observed and average rainfall from July 2015 to present.

So while Thurmond received 14.3 inches above its average in the winter 2015 compared to an excess of 6.8 inches from April to June 2017, those 14.3 inches were amplified further because a greater percentage of that precipitation actually reached the reservoir.

Similarly, had those additional 6.8 inches come earlier in the year, say from January to March, it would have been a different story.

~ Jeremy S. Buddemeier, Corporate Communications Office

About US Army Corps of Engineers Savannah District

The U.S. Army Corps of Engineers Savannah District oversees a multi-million dollar military construction program at 11 Army and Air Force installations in Georgia and North Carolina. We also manage water resources across the Coastal Georgia region, including maintenance dredging of the Savannah and Brunswick harbors; operation of three hydroelectric dams and reservoirs along the upper Savannah River; and administration of an extensive stream and wetland permitting and mitigation program within the state of Georgia. Follow us on Twitter @SavannahCorps and on Facebook.com/SavannahCorps
This entry was posted in Drought in the News, Rainfall Update and tagged , , , , , , , , , . Bookmark the permalink.
  • Littlebigpaw

    What happened to all the good comments?

    • Great question! and you asked before I could offer an explanation. We were using Facebook as a temporary comment plug-in for the past year while DoD worked out contract details with Disqus (our original and now current plugin). All the Facebook comments can still be viewed on our Facebook page. The direct link to this post feed is here:

      All the posts from the past year or so have also lost their comments due to the shift from Facebook, but the conversations are all still available on our FB page.

      Thanks for your patience and understanding! ~Russell Wicke

  • Ferris

    Pool Balance Suggestion- Summer droughts target noticeably lower Thurmond pools relative to Hartwell, beginning on weekends. Project weeks and pool balances begin at midnight on Friday. Hartwell and Russell typically have no weekend releases for Thurmond, but Thurmond must maintain minimum drought release rates. In addition, Russell continues pumping water from Thurmond on weekends for releases the following week. Thurmond may fall 0.6′ below balance with Hartwell by Sunday midnight. Thurmond targets regaining the loss by Friday midnight, but consistently remains lower than Hartwell except at midnight Friday. Targeting Thurmond 0.3′ higher than Hartwell at midnight Friday would more evenly balance Thurmond lake recreation and Thurmond lakefront property owners relative to Hartwell since the actual balance would occur around midnight Saturday. Thanks Scott L for observing this imbalance on the BtB Aug 12 2015 topic. Although small relative to the overall drop, every few inches help. In addition, a relatively higher weekend Thurmond pool slightly improves Russell pumping efficiency.

    • Thanks for the recommendation Ferris. I’ll be sure to alert the water managers to your comment here. ~Russell Wicke

  • Ferris

    How is SEPA replacing SRB contract generation shortages? Declarations and historical data show generation averaging 4K MW-HR below contract in Jul, or 13% low.

    • Good question Ferris. I had to contact SEPA to confirm: They are purchasing replacement energy on the open market because of restrictions in the drought plan. In July, SEPA’s cost for replacement energy is $94.98 per megawatt hour (MW-HR). The current declaration shows SEPA falling 5,732 MW-HR short of contract. For the week of the current declaration that math comes to $544,425.36 — just for one week.

      It’s important to note that before SEPA resorts to purchasing energy on the open market they have recourse to pump back, but there are still costs associated with that. Due to insufficient “stream flow” (that is discharge from our dams) they made rigorous use of the pump back option, purchasing 196 hours of pump energy. Each “unit hour” of pumping requires the purchase of 97 MW-HR of energy – so 196 hours of pump energy is equivalent to 19,012 MW-HR. In their response to my inquiry, they mentioned the cost of pumping energy in July is $31.49 per MW-HR. That math comes to $598,687.88 for utilizing pump back.

      So as you can see, drought conditions certainly impact all the project purposes.

      Hope this helps. ~Russell Wicke

      • Ferris

        Russell: Thanks for the research- an extra million a week is serious money! What would happen if SEPA saved this money and simply provided less power? Would that affect operating costs and debt payment?

        • Yes – it puts a strain on the operation. Your question is purely hypothetical because I don’t think SEPA has the authority to make that call. But if hydropower as a purpose were marginalized enough, this could happen as a matter of course. Hydropower is one of the original purposes set by Congress and it had significant influence on the justification for building the dams. It is what allows the federal government to repay the taxpayer for the construction of the dams and it also contributes to keeping power costs low in the Southeast. SEPA wouldn’t make that kind of call (save money by providing less power.) But if persistent drought led to strict enough policy constraints on hydropower, SEPA would not be able to consistently provide power when their customers (the power companies) demand it. As a result those customers would arrange to contract for peaking power elsewhere – at greater costs. And I don’t think it’s unrealistic to suppose that cost would be passed on to the end user. I hope this helps. ~Russell

  • Ferris

    An often offered suggestion reduces releases to 3,600 cfs at elevations that vary from Full Pool to DL1. The comprehensive study showed that changing to this operation greatly increases downstream flooding risks. ALT4 in the 2017 Draft EA reduces Thurmond releases to 3,600 cfs beginning at DL1, with 3,100 cfs releases Nov-Jan, until Hartwell and Thurmond rise 2′ above DL1. It states on page 105, “ALT4 is the only alternative that would increase flood risks”. Appendix B Table 8-11 shows that ALT4 increases the average of daily flows over 30,000 cfs by 24%, a potentially catastrophic and significant adverse impact to flood mitigation.

  • K Miller

    After a casual review of the data in the app over the last 6 months, it would appear that every time there is a rain event, the CFS outflow spikes (particularly at the Hartwell dam), essentially holding the lake level at or near the same water level. What causes this? Is there a flow gate of some sort that is set at a certain elevation and any additional water volume flows out?

    • Hello K – thanks for your question. In order to answer questions about specific days, I’d need more specific times you are referring to. However, I may be able to address it generally. First, we don’t respond to rainfall by releasing water. That is a misconception. The markets and the contracts don’t work like this, and there is no benefit. Rather, Hartwell is a peaking hydroelectric dam. This means water isn’t released at a steady rate throughout the day, but only during the few hours the market peaks in its demand for power. During these hours release rates are high, but when averaged out with the rest of the day (when nothing is flowing), you get rates closer to what you might expect: between 2,000 and 4,000 cfs. The declaration currently has Hartwell scheduled for 2,500 cfs for the rest of this week. This may mean at certain hours of the day release rates go above 10,000 cfs, but for the rest of the day it is zero, and this would average out to 2,500 cfs over the 24 hour period.

      I hope this helps. ~Russell Wicke

      • Fish

        It looks to me from viewing USGS Lake Hartwell data page summaries for 6-4 thru 7-4-17 that we have gotten a little over a wonderful 8 inches of rain… and the lake has gone up…oops, no, down about 0.2 in. I take this as a sign that our Corps’ daily 4k cfs releases from Thurmond/Clarks Hill are not working very well to refill our lakes, since they are not working at all.

        • Hello Nick – I believe you posted this comment on the Facebook post, and had an engagement with Ferris over this? ~Russell Wicke

          • Fish

            Sort of…I was hoping for a straight answer from you, as I would guess are many adjoining property owners on Lake Hartwell, and probably Thurmond. K. Miller’s comment raised the issue again.
            If the system can not support the lakes ability to refill significantly with 8 in of rain fall in a month, how much rain is required, surely not a hurricane? It seems to me from this window that the hydropower expectations from the SRB are excessive and should be reduced. I have never understood the logic of hydropower contracts which are dependent on rain as the source not having a significant drought clause. I would think that an appropriate drought clause would be required. It also puzzles me that I get comments from lake Hartwell that SRB lakes and Lanier seem to suffer most in droughts, while ‘power company’ lakes like Murray, Wateree, Norman, and Wylie suffer less…I haven’t looked at the numbers, but those are the comments.

          • Understood. There are several contributing factors. Hydropower isn’t one of them: the drought restrictions are holding releases at Thurmond to no more than 4,000 cfs. SEPA is falling short on the contracts you mentioned due to the drought restriction. It’s costing them in the neighborhood of a million dollars a week to either purchase replacement energy or buy pumpback energy.

            Concerning other reservoirs, the SRB isn’t comparable to other river basins because we have multiple competing purposes, and an environmental requirement to put a minimum amount of water downstream to dilute the state-permitted treated waste water going into the river. Other systems don’t have the same demands as ours.

            The required minimum 4,000 cfs is only one factor that has an impact. Evaporation at this time of year adds roughly 1,200 cfs outflow to the equation. There’s also transpiration, which is harder to measure, but no less significant. Trees and vegetation pull a great deal of water out of the ground before it can turn into runoff, which explains how inflows can sometimes almost double in the winter.

            And finally, although we’ve gotten around 150% of normal rainfall in the period you outline, inflow for the same period is still only 75% of normal. This means the ground is still claiming disproportionate amounts of water due to the preceding 15 month period of rain deficits. The good news is, we are in much better condition than we projected before April, and if we keep getting above average rainfall, conditions will return to normal more quickly.

            I hope this helps. ~Russell

          • Ferris

            Russell, it sounds like the drought plan favors lake elevations by increasing SEPA power costs for everyone; is that ethical? Do you consider that a “drought clause”?

          • Fish

            Ferris- You may be more familiar with the term ‘Force Majeure’ which I think is the essence of a drought clause. I gather that they are used in the US from Calif to Conn to deal with unpredictable ‘Acts of God’ such as droughts, and their impact on contracts. At some point, contractual obligations may become impossible or commercially impracticable.

          • Ferris

            Fish: Please describe how a SEPA hydropower drought clause would benefit lake levels.

          • Ferris

            Fish: You have questioned the lack of a SEPA drought clause several times over the past few years, and again on this topic, with the implication that a drought clause would benefit lake levels. BtB topics, discussions, and direct answers to your questions have repeatedly explained why hydropower does not determine drought release rates. A SEPA hydropower drought clause provides no additional benefit to lake levels; if you have evidence to the contrary, please present it.

            The responsibility for economic decisions including supplemental power costs, customer charges, customer retention, and drought clauses belongs to SEPA and has no effect on the 2012 Drought Plan releases. Discussions regarding additional costs incurred by SEPA during droughts confirm the necessity of SEPA securing alternate power sources since the drought plan prevents increasing release rates to meet power demand.

            As Russell explained, hydropower is not preventing lake level recovery because downstream flow requirements set drought release rates, not hydropower. Drought studies require hydropower consideration; however, hydropower did not determine the selection of ALT2 in the 2012 EA, the current drought plan. Page 118 shows the approved ALT2 with a 9.8% power cost increase over the prior plan. Page 45 compares ALT releases. EA Appendices excerpts from my post on the Sep 3 2015 topic show that seven natural resource agencies objected to 3,600 cfs release rates for DL3 pending further data determined by the long awaited 2017 Draft EA.

            References: USACE Savannah District Home Page:
            Find documents for the 2012 Drought Plan and the DL4 Study under the Plans & Reports Button

      • K Miller

        Yesterday appears to be an example of what I’ve been observing. 1.465 inches of rainfall for Hartwell, 6972 cfs discharge.

        • Thanks for the specific example K.

          Consider that on Aug. 1 and 2, Hartwell discharge was higher than normal, above what was published on the declaration. But there was no rain at all from July 30 to Aug. 3. What’s happening here is a balancing operation with Thurmond. It’s not so much a response to rainfall as it is to reservoir levels (therefore localized rainfall can be an indirect influence).

          Also consider Hartwell had zero discharge on Aug. 5 and 6, and got three times the amount of rain on the 7th (contributing to the imbalance of levels) while Thurmond fed the river alone Aug. 5 and 6 at 4,000 cfs. Now that both reservoirs are better balanced you will see discharge at Hartwell drop off.

          Keep in mind that the meaningful discharge point is Thurmond, not Hartwell. Water leaving Thurmond leaves the “system” and ultimately that is the discharge point affecting Hartwell and Thurmond lake levels together. The balancing operation ensures that when Thurmond gets a great rain event, Hartwell benefits too, and vice versa.

          I hope this helps. ~Russell Wicke

          • Fish

            I finally found a current SRB Lakes daily report page and note that Thurmond has released close to 42-4300 for the last week. 4k was too much, and this is obviously worse. And the Corps’reason for the releases beyond the hurtful Drought Plan levels is??
            Hartwell is now down 7 ft…so much for good Labor Day recreation.

          • Yes, the discharges have been slightly higher recently due to our agreement with Duke Power’s licensing with the Federal Energy Regulatory Commission (FERC). When they must send extra water downstream to meet their license requirements, we must pass that additional water through our system to the downstream.

            This is temporary. It only happens in the summer when we are in a drought trigger and must run a minimum of 11 days. We have returned to our standard drought management plan. As one of the cooperating agencies on Duke’s FERC license, we must adhere to the agreement, too. In effect, it simply became a “pass-through” of the extra 200 cfs for the period.

            – Billy Birdwell, Corporate Communications Office

  • Ferris

    Russell, your highlights of SEPA communications confirm what informed stakeholders already know- HRT hydropower represents an integral and mandatory drought plan consideration in spite of unremitting and endlessly disproven FALSE claims to the contrary. The first 3 weeks of Jul, SEPA purchased expensive peak power on the open market to compensate for HRT shortages, also in spite of unremitting and endlessly disproven FALSE claims to the contrary. Russell pumping contributed to replacement power, also at an additional cost. 

    My 2015 posts and tables demonstrated that SEPA has little recourse for drought energy replacement because of insufficient extra power available from the other 7 projects in the GA-AL-SC System. Carters Lake persistently maximizes drought pumpback generation with essentially no extra applicable to HRT. Lanier usually suffers our droughts and is 5.3′ low. Allatoona is full but generates only 8% of HRT annual generation. West Point is full but generates only 11% of HRT annual generation. The other 3 are river run projects with little elevation flexibility and must conserve water for navigation purposes. 

    Hypothetically, and ignoring contributions to projects with navigation purposes, order of magnitude calculations suggest that lowering Allatoona and West Point pool elevations through a 10% increase in annual generation could reduce average annual Thurmond releases by 38 cfs. This very small number falls well below release rate precision, but illustrates the futility of replacing HRT power with other projects in the system. 

    0.10*(0.08+0.11)*(1/3 Generation/Flow for JST/HRT)*6000 cfs = 38 cfs

  • Elisabeth White Putnam

    I will still never understand how it rained three inches at Thurmond on Wednesday 7/26/17, but yet we loose a foot of lake level???? What are the excuses now???? WHERE IS IT ALL GOING other than down the river????

    • Hello Elisabeth. I understand your frustration. There are good explanations to your questions, so I will do my best to provide sensible answers:

      First, to clarify, the Thurmond sub-basin received less than a quarter of an inch on 26 July. Localized areas may have gotten up to 3 inches, but when you average that out over the thousands of square miles that make up the drainage basin, the number is much smaller. Thurmond has still not quite reached its July average for rainfall, but it is close.

      2. Thurmond levels did decline about 10 inches over the weekend, but the levels will return throughout this week. The weekend decline is typical when we are making heavy demands on the pump back capability, due to drought. Over the weekend and at nights we pump water from Thurmond up to Russell, and release the water back into Thurmond throughout the week when energy demands spike. This helps keep levels higher in general. You can see a brief animation on this concept here:

      3. We are still recovering from 15 months of rain deficits. The ground went dry during this time, which decreases the runoff ratio. The basin is 10,580 square miles. It takes a while to re-saturate the ground after 15 months of deficits. But conditions indicate we are heading in the right direction.

      4. During the peak of summer, even in normal conditions reservoirs tend to decline (this happens on natural lakes too). There are two reasons: 1- Evaporation is high (accounting for about 1,200 cfs in July/August), and 2- Transpiration significantly lowers the runoff rate. Trees are fully leafed; shrubs, grasses, and all other vegetation in the basin get first dibs on rain when it falls, which decreases the amount of water that actually makes it into tributaries. A single large oak, for example, can transpire up to 40,000 gallons in one year.

      While summer conditions persist, recovery will be slow even with good rainfall. But runoff will increase as the weather cools.

      I hope this explanation helps. ~Russell Wicke

      • Elisabeth White Putnam

        I guess the pumpback isn’t working because lake is still steady dropping….!!!!

        • Boar Master

          It drout!!!! Lakes drop!!!!

          • Elisabeth White Putnam

            We are not in a drought…..!!!

          • Boar Master

            What?!?!? Send more water!!!!!

          • Ferris

            Elisabeth: You may be referring to reports based on the Drought Monitor that shows our region no longer in drought. The Drought Monitor places a lower emphasis on hydrological drought that typically affects “only” river and lake levels. In spite of recent rainfall, persistently below normal local inflows keep the Savannah River and upstream lakes in the hydrological drought designated as Drought Level 2.

            Some Types of Drought
            – Meteorological Drought: Based on rainfall deficit and length of dry period
            – Hydrological Drought: Based on the impact of rainfall deficits on the water supply such as stream flow, lake levels, and ground water table decline
            – Agricultural Drought: Based on impacts to agricultural by factors such as rainfall deficits, soil water deficits, reduced ground water, or reservoir levels needed for irrigation
            – Socioeconomic drought: Based on impact of drought conditions on supply and demand of some economic goods

            Additional information for those interested.
            NWS Climate Prediction Center:
            NWS Drought Fact Sheet:
            NWS Types of Drought Discussion:

          • Ferris

            The chart illustrates the longer duration to reach and therefore recover from hydrological drought.

          • Fish

            Russell- Ferris raises an issue I don’t understand. Since we are in Drought 2 and in an ongoing hydrological drought, why aren’t releases from Thurmond reduced at least to 3800 if not 3600? Why doesn’t the drought plan address this deprivation of much of the recreational use of Thurmond and Hartwell? Many good-faith purchasers bought adjoining lake-front property with the expectation that the Corps would look after their interest. Many families planned vacations to use the Corps ‘safe swim’ beach areas for their families- the markers are laying on dry ground. We have lost too many recent recreation seasons. With many millions of visitors a year to the SRB lakes (about 15 million now?), the Corps Drought Plan needs to be much more responsive.

          • Thanks for the question Nick. We are following the Drought Plan, as part of the established SRB Water Manual. This plan was developed by the Corps in coordination with natural resource agencies in Georgia and South Carolina, with federal resource agencies, with municipalities, and with public input.

            For this time of year, the plan calls for for 4,000 cfs discharge for Thurmond in DL2. Reference here:

          • Fish

            And obviously a very flawed plan…again and still flawed despite many Corps ‘adjustments’ over the years.

      • Elisabeth White Putnam

        Why does the corps drop lake levels drastically every year starting in August? We were finally making some ground by first to mid July , & then first of August it’s like the corps pull the plug…!!! We have had good rains!! I’m sooo tired of seeing the muddy banks & hearing excuse after excuse!!!! We the people have to make a change!