Untitled Document
About Water Level Management
Keeping near Full Pond
Water Level Management at SMLA

SMLA’s volunteer scientists explore ways to improve the AEP predictive model for lake levels, with a particular focus on anticipating dry periods and increasing lake levels during those dry periods, and periodically provide our recommendations to AEP.

Please go to http://www.smithmountainlakelevel.com to see the present water level as well as historical water levels.

Water Level Management Committee

  • Smith Mountain Lake (SML) water levels regularly fall to low levels, below 792 ft adjusted height, about every 4 years
  • The actual Smith Mountain Lake water level can be as much as 2 ft below the adjusted height.
  • Every 5 years, AEP will “…solicit comments on the performance of the project in maintaining lake levels and in providing flows necessary to protect instream beneficial uses.”
  • In order to provide such comments to AEP, The Water Level Management Committee was formed in 2013 with the mission:

To address with data and science the water shed, water level and water release issues at SML and LVL and to participate in any updates to the current Water Release Permit, and to participate in the construct of a new 401 Permit in 2015

Water Level Management Committee

  • The Committee demonstrated that late summer/early fall rainfall was the controlling factor in lake level, with groundwater flows from the winter recharge period being a contributing factor
  • Heavily localized rainstorms accounted for the large deviations regularly seen between actual lake level and the Hydrologics predictive model
  • The following recommendations were provided to AEP in the Spring of 2015 (* indicates recommended responsibility)
  • Use actual Leesville discharges since 2010 to improve predictive accuracy in the June-December timeframe (Hydrologics*)
  • Maintain closer adherence to following the minimum Leesville discharges specified in the Water Management Plan (AEP*)
  • Request adding a probability value for the lake falling below 791 ft. in the Trigger 3 definition (Bedford Regional Water Authority*)
  • Request that DEQ renew the AEP permit for the next 5 years and review  our recommendations and issues (DEQ*)
  • Continue to assess predictive model performance and address issues during the next 5 years (SMLA*)

Water Level Management Committee

  • Continue to analyze the factors that control SML water levels over the next 5 years
  • Expand the committee membership to include additional expertise from the SML community
  • Specific areas of interest include:
    • Evaluate using more realistic groundwater models
    • Explore the utility of NOAA drought predictions for identifying low rainfall periods further ahead of their occurrence
    • Examine whether there is a better scaling relationship between the USGS streamflow gages and the larger SML drainage area
    • Improve estimates of the groundwater loss from the Smith Mountain and Leesville Lakes
    • Examine alternate methods for generating synthetic streamflows
  • Provide appropriate comments to AEP in the Spring of 2020
Water Level Management Reports
Learn what determines lake water levels
Water Level Management Reports




Many of our residents and interested parties have asked about the “recent” low water. SMLA has been responding to quite a few, as has the Bedford Regional Water Authority (BRWA) and Appalachian Power Company (Appalachian).  The following consolidated contributing factors are in order of their impact.


First, and foremost, our area has not had measurable rain since well before Labor Day.  If you look at the “one year” view of the actual lake level chart at www.smithmountainlakelevel.com you can see that the level trend started down around mid July. While that timeframe is not unusual, the level usually goes up in Oct/Nov when the hurricane remnants come through and the plants and trees go dormant.


Second, Appalachian is required to follow the release protocol which remains constant until certain “triggers” are activated based on weather forecasts, inflow models, power generation needs, and the adjusted water level of Smith Mountain Lake.  (The adjusted level is the level Smith Mountain Lake would be, if the available water in Leesville Lake were pumped back for re-use at Smith Mountain.)  Trigger 1 is activated when the forecasted lake level has a 20% chance of dropping below 790.5 feet, adjusted, in 16 weeks.  Trigger 2 is activated if the forecasted lake level has a 2% chance of dropping below 790 feet, adjusted, in 10 weeks.  Trigger 3 is activated if Trigger 2 is in effect and the adjusted elevation is less than 795 feet between December 1 and March 31, or anytime the adjusted elevation drops below 791 feet after September 30. 


Current lake levels are available on Appalachian’s website at http://www.aep.com/environment/conservation/hydro/ . For example, as of 10:50am this morning (January 25, 2018), the actual water level at Smith Mountain Lake was 790.87 feet.  The adjusted lake level (taking into account the water in Leesville Lake above 600 ft. available to be pumped back for re-use) was 791.87.  Based on these current numbers, and considering future forecast and inflow models, no triggers have been activated at this time that would reduce the amount of water that is required to be released downstream.  Appalachian continues to monitor lake levels to determine if any triggers can be implemented.


Keep in mind that the primary function of the Smith Mountain Project is the generation of electricity.  The demand for electricity generally increases during periods of extreme cold weather, requiring more power generation at Smith Mountain.  When generating electricity at Smith Mountain, Appalachian utilizes a 2-foot “power pool” – meaning that the water level in Smith Mountain Lake could drop as much as 2 feet during generation, resulting in a potential 13-foot rise in the water level at Leesville Lake (from 600 to 613 ft.).  Afterwards, most of this water in Leesville is pumped back into Smith Mountain Lake for re-use.  Hence, if you are out on the water boating for a few hours, the water levels at your dock might be lower – or higher – when you return.        


Third, evaporation rates at the lake are substantial - about 73M gallons a day.


Fourth, the Smith Mountain Lake Water Treatment Facility currently withdraws about 1.8M gallons a day from the lake – which is 40 times less than the rate of evaporation.  If all inflow (rivers, streams, and groundwater infiltration) stopped flowing into the lake, and there was no precipitation, and if the water withdrawal rate were raised to 3M gallons a day, the lake level would only be lowered by less than 2” per year. In other words, the impact of the water withdrawal on the lake is negligible


As you can see, BRWA has the least impact on the lake. The main culprit is no rain. Other factors will cause variable impacts, but only Leesville releases can be negotiated.  With each renewal of the FERC license, SMLA is fighting for reduced outflows. SMLA was actually successful the last time, however downstream advocates have a strong lobby in the House of Delegates.  SMLA did make Appalachian Power aware that they were releasing MORE than the required amounts.  This has had a positive impact on lake levels over the past couple of years since we advised them of the impacts.

So how does 2017 compare to previous years in terms of lake levels and what can we expect in 2018?  First, groundwater flow makes up about 65% of the water flowing into Smith Mountain and Leesville Lakes on an annual basis.  We did have heavier rain than normal in April and May of 2017, which increased the groundwater flow throughout the summer.  As a result, lake levels in 2017 were about 1.5 ft. higher than they were in 2012, the last time lake levels were really low.  The lake level has always recovered to full pond during the winter, typically in the January to March time period, even during periods of low rain.  Sources expect 2018 to be no different. About 60% of the time when there has been a low lake event in the fall, like 2017, the lake has also been low the following year.  Another way of saying this is that droughts in the area typically last more than one year.  The Virginia Department of environmental Quality publishes their drought assessment on their website

http://www.deq.virginia.gov/Programs/Water/WaterSupplyWaterQuantity/Drought/CurrentDroughtConditionsMap.aspx.   Interested parties can follow the drought outlook for our area on that website.

Why is the Smith Mountain Lake water level so low (2017)

To answer this question, one must first understand that Smith Mountain Lake (SML) and Leesville Lake together form a pumpback project managed by Appalachian Electric Power(AEP). Typically, water is released from SML to Leesville in the afternoon to generate electricity for sale, and water is pumped back from Leesville to SML during the night when electricity is cheaper. In this pumpback system, Leesville Lake can vary between 600 and 613 ft, which corresponds to roughly a 2 ft variation in SML water level. Because of this pumpback operation, the actual water level of SML is constantly changing.

The attached figure shows the calculated values of the actual SML water level on a daily basis since January, 1995 up to the present. The time of the calculation is midnight, which generally corresponds to a period between electricity generation and pumpback and usually represents the lowest actual level for that day. As the figure shows, the actual SML level typically varies between 793.5 ft and 795 ft.; currently the actual level is around 792 ft. While this level may be low enough that some residents having trouble getting their boats into the water, the SML water level has been this low many times in the past during periods of low rainfall, in 17 out of the last 23 years.

During droughts, the SML water level has gotten even much lower, below 790 ft. Fortunately, these especially low lake levels have occurred much less frequently, between every 5 to 7 years on the average. The double low levels in 1998-99, 2001-02 and 2007-8 are actually extended periods of very low rain. AEP is also now using a predictive model to anticipate these situations and reduce releases from the Leesville Dam to avoid these low lake levels. The few spikes in actual level above 795 ft. correspond to temporary flood conditions, typically only lasting a day or so, where SML overflows the spillway.

The actual SML and Leesville Lake levels are published hourly by AEP on the web at http://www.aep.com/environment/conservation/hydro/. The actual SML levels daily for the past year are also available on the web at http://www.smithmountainlakelevel.com/.

The AEP website also gives the SML adjusted water level. This is the level SML would be at if the water in Leesville Lake above 600 ft. were pumped back to SML. The SML adjusted level is the level used in the AEP predictive model to anticipate low lake levels and to reduce the water releases from Leesville Dam.


Executive Summary

Study of Water Level and Release Issues at Smith Mountain and Leesville Lakes

Smith Mountain Lake Association Board Water Management Committee* February 15, 2015

The Smith Mountain Lake Association (SMLA) formed the Water Management Committee two years ago to study and prepare comments on the performance of the project in maintaining lake levels and in providing flows necessary to protect instream beneficial uses. The Committee adopted a scientific approach with a study focus on keeping lake levels above trigger points, thereby avoiding having to reduce downstream flows below the “no trigger” protocol levels as much as possible. The results of this study are documented in an annotated set of briefing slides that are included as an attachment. The key points of this study are summarized in this Executive Summary.

Figure 1 shows the Smith Mountain Lake (SML) adjusted lake level from 1995 to 2013. The Committee’s primary interest was in the very low lake level events where the lake fell close to or below 791 ft. The figure shows these low lake level events primarily occur in the second half of the year, after June. The 2012 event was of particular interest since the current predictive model used by AEP did not predict any trigger events until the lake finally fell below 791 ft. in late November.


Figure 1: Effect of Protocol discharges

The blue squares in Figure 1 show the lake levels that would result if trigger points can be accurately predicted and the appropriate protocol discharges in the current AEP Water Management Plan are followed. In all these instances the lake level is kept above 791 ft. except in 1998 where it falls below by only about 1 inch. Based on this analysis, the Committee felt that current protocols are adequate to maintain lake levels and protect downstream flows provided accurate lake level predictions can be made sufficiently far in advance and the discharge protocols are followed.

Over the past two years, the Committee has examined the factors that might control lake level and analyzed the performance of different simulation methods. Based on the results of this analysis, the Committee now believes:

  1. Heavy localized rain storms can cause significant changes in lake level, sometimes in excess of 2 ft. within a day or two, which cannot be predicted by the current model. However, since the current predictive model is always re-initialized to the new lake level after such events, these localized storms will not affect long-range predictive accuracy.

  2. Most of the other factors examined by the Committee do not appear at this time to cause lake level fluctuations more than a few inches and are not considered responsible for the very low lake events seen in Figure 1.

  3. The very low lake level events seen in Figure 1 do appear to be a combination of prolonged rainfall deficits after June coupled with low winter groundwater recharge in the January-April time period. Figure 2 shows the lowest lake level reached in the second half of the year plotted against the rainfall surplus or deficit in the June to December time period, converted to feet of lake height. All of the lake levels below 792 ft. are associated with rainfall deficits

  4. Prolonged rainfall deficits appear to be the primary cause for low lake events with reduced groundwater flows in the June-July time period from low winter recharge acting as a contributing factor. This result suggests there is a possibility to better predict the likelihood of such low lake events based on good groundwater models and NOAA drought predictions.


    Figure 2: Lowest Lake Level vs Rainfall Surplus/Deficit

  5. Based on the past history shown in Figure 1, low lake levels that can cause trigger points typically occur every 4 to 7 years, although the new discharge protocols may change this behavior.

Based on the Committee’s analysis, the following recommendations and requests are proposed with the organization considered most appropriate to handle them shown in parentheses:

  1. Use actual Leesville discharges since 2010 to improve predictive accuracy in the June- December timeframe (Hydrologics)

  2. Maintain closer adherence to following the minimum Leesville discharges specified in the Water Management Plan (AEP) (particularly in years predicted to be low lake event years)

  3. Request adding a probability value for the lake falling below 791 ft. in the Trigger 3 definition (Bedford Regional Water Authority)

  4. Request that DEQ renew the AEP permit for the next 5 years and review our recommendations and issues (DEQ)

  5. Continue to assess predictive model performance and address issues during the next 5 years (SMLA)

The Committee also identified a number of issues that it feels need to be examined over the next five years. These issues are described in more detailed in the attached final report and include:

  1. Use NOAA Climate Prediction Center drought predictions to identify low rainfall years

  2. Examine more “realistic” groundwater models

  3. Examine the interaction of rainfall and groundwater deficits

    1. Moderate rainfall and groundwater deficits are additive

    2. Large rainfall deficits dominate

  4. Examine the effect of inflow/discharge correlations

  5. Consider alternate methods for generating synthetic streamflows

  6. Reevaluate the 40 cfs addition to the Roanoke gage

  7. Consider using the Niagara rather than the Roanoke gage for drainage area scaling

  8. Reconsider the “groundwater loss” term

  9. Examine evaporation loss variability

The Water Management Committee would be very willing to work with any of the appropriate agencies to address these recommendations and study these issues over the next five years. In closing, the Committee wishes to especially recognize and thank AEP and Hydrologics for making their data, procedures and modeling techniques available to us for this study and answering our many questions and Dr. Eric Anderson for his contributions to the Committee’s analysis.

Water Management Committee Members: Russ Johnson (chairman), Bill Piatt (vice- chairman), Jim Colby, Bill Brush, Rob Whitener, Chuck Sinex and John Lindsey

Click on the links below to download a report. Acrobat is required to read these documents.

Download full text of Water Level Management Report
Download Water Level Management Briefing
Download Water Level Management Executive Summary


Save Our Streams: Reports
Results that matter
Reports on Save Our Streams at SMLA

Summary 2015 SOS Data

Of the 21 stream sites monitored for ecological conditions:
  • 14 sites (67%)had acceptable scores between 9-12 compared with 13 sites in 2014
  • 4 sites (19%) were classified in the gray zone with a score of 8 –ecological conditions could not be determined compared with 3 sites in 2014
  • 3 sites (14%) had unacceptable ecological scores between 0 –7 compared with 2 sites in 2014

Click on the link below to download a report.
Adobe Acrobat is required to read these documents.

Download 2015 Save Our Streams Briefing
Download 2014 Save Our Streams Briefing
Download 2013 Save Our Streams Briefing


Water Safety Council Education Programs
Know safety!
Education at Water Safety Council

Virginia law requires that all operators ages 14 and older of a personal watercraft (PWC - jet ski/seadoo) must receive training either in the form of a boaters' safety course or via a PWC rental provider.  All operators of a boat must also have this training. For more information on this training please click the VDGIF website.

SMLA Top Navigation Bar
E-Mail CALL 540.719.0690