Pompton Lake Dam. March 12 15, 2010 Nor easter Post-Flood Report

Transcription

1 Pompton Lake Dam March 12 15, 2010 Nor easter Post-Flood Report 31 July 2010

2 EXECUTIVE SUMMARY This report documents the meteorologic, hydrologic, and hydraulic aspects of flooding along the Ramapo River during March The river and flash flooding that resulted from this heavy rainfall and snowmelt damaged residences and businesses in many communities and severely impacted local and state municipal infrastructure. Area residents affected by the flood questioned the impact that the operation of the flood gates at the Pompton Lake Dam had on their area. Residents reported seeing flood water rise faster and sooner during this event than during past floods. In response to these concerns, the Corps of Engineers was asked prepare a post-flood report on the operation of the flood gates for the March 2010 nor easter and determine if these operations contributed to downstream flooding as compared to pre-project conditions. The original gate operation was revised after the April 07 event to improve performance. A review of the gate operation data indicates that the gates operated as intended in accordance with the revised Gate Operation Rule Curve. The 3-day rainfall total for the March event was 6 inches in the Ramapo River basin; the antecedent precipitation consisting of the snowpack resulting from the February snowstorms was equivalent to 4.1 in. of liquid water. Peak outflow for the event was recorded to be cfs at the USGS gage immediately downstream of the dam. Peak inflow was estimated to be cfs based on storage computations. The without project outflow was simulated to be cfs, resulting in an increase in peak flow of 214 cfs, which was less than 2% of the total. The increase in discharge resulted in an estimated increase in stage of 0.25 ft at the USGS gage 400 ft downstream of dam, and less than 0.1 ft at Dawes Hwy. There was no increase in Pompton Plains. The gate operation shifted the timing of the peak discharge 3.25 hours earlier, and the hydrograph in general about 1.5 hours earlier. Options to the gate operations including revisions to minimum gate opening, timing, and target elevations can be investigated to further reduce downstream impacts. 1

3 INTRODUCTION This report documents the meteorologic, hydrologic, and hydraulic aspects of flooding along the Ramapo River during March The Ramapo is a tributary of the Pompton River, which flows into the Passaic River. The latter enters tidewater in Newark Bay within the Greater New York Harbor Area. Recorded rainfall from this event varied from 2.5 inches at Cape May to approximately 6.0 inches within the study area. The river and flash flooding that resulted from this heavy rainfall and snowmelt damaged residences and businesses in many communities and severely impacted local and state municipal infrastructure. Due to the extent of damage, the State of New Jersey declared a state of emergency. Area residents affected by the flood questioned the impact that the operation of the flood gates at the Pompton Lake Dam had on their area. Residents reported seeing [flood] water rise faster and sooner during this most recent May 2010 nor easter than during past floods. In response to these concerns, the Corps of Engineers was asked prepare a post-flood report on the operation of the flood gates for the March 2010 nor easter and determine if these operations contributed to downstream flooding. DESCRIPTION OF THE STORM A severe nor'easter impacted New Jersey from Friday March 12th through Monday March 15th. A slow moving upper level low pressure cell drifted into the eastern United States and spawned a surface low pressure system off the South Carolina coast on the morning of Friday March 12th. Light to moderate rain spread north across the entire region through the day on Friday as the low pressure drifted north and slowly strengthened. With a strong fetch off the Atlantic Ocean, rainfall rates became heavy (0.3 in/hr or more) overnight Friday and through the evening of Saturday March 13th, resulting in small stream and eventually major river flooding. Meanwhile, a high-pressure system anchored in the Canadian Maritimes also strengthened and a very strong pressure gradient developed between these pressure systems overnight Friday March 12th. This resulted in strong, damaging easterly winds across much of the area through the day Saturday March 13th, especially along the New Jersey coast where minor to moderate coastal flooding also occurred. The strong winds and widespread heavy rains slackened off overnight Saturday 2

4 March 13th as the surface low was stationary across the Delmarva region. Showers and even a few thunderstorms continued to rotate in off the Atlantic through the day on Sunday March 14th and continued into Monday March 15th as the low pressure system slowly moved eastward and finally out to sea by Tuesday morning March 16th. Precipitation Totals Light to moderate rain fell over the area from south to north starting early Friday morning March 12th and continued through the day. Rain became heavy throughout most of the area from overnight Friday through the evening of Saturday March 13th before letting up. Periods of showers and areas of steadier rain would continue to rotate in off the Atlantic Ocean through the day on Monday March 15th, adding to the already high rainfall totals. The heaviest rainfall amounts averaged 4-6 inches; isolated amounts as high as 7 inches were reported in the northeastern sections of the Mount Holly county warning area, including the northern New Jersey coastline, east-central New Jersey and northeastern New Jersey. Rainfall totals averaged 3-5 inches with isolated higher amounts in southern New Jersey, west central New Jersey and extreme southeastern Pennsylvania. In a band stretching from northwest New Jersey to just northwest of Philadelphia and down into Delaware and eastern Maryland, lighter amounts averaging 2-4 inches were observed. The lightest amounts were observed over the western sections of the Mount Holly area in the Pennsylvania counties of Berks, Lehigh, Carbon, and Monroe Counties where only 1 to 3 inches of rain fell. The National Weather Service 3-day rainfall total map showed a total of 6 inches in the Ramapo River basin at the New Jersey-New York border and a minimum total of 2.5 inches at Cape May, New Jersey. A total storm rainfall of 5 inches is estimated from this map at the Charlotteburg Reservoir, NJ rain gage in the Pequannock River Basin to the west of the Ramapo River basin. A total storm rainfall of 3.82 inches was recorded at the Caldwell NJ Essex County Airport rain gage just south of Wayne NJ and the Ramapo and Pompton River basins, and a total storm rainfall of 3.76 inches was recorded at the Woodcliff Lake NJ rain gage operated by the USGS 10.5 miles east northeast of the Pompton Lake Dam, which is the USGS rain gage closest to the Pompton Lake Dam. 3

5 In the case of the March nor easter storm, the antecedent precipitation consisted of the snowpack that resulted from the snowstorms of February 10 and 23-28, A preliminary estimate of the total February snowfall over the Ramapo River basin is from 10 to 20 inches. The liquid water equivalent of the total February snowfall recorded at the Woodcliff Lake NJ USGS rain gage is about 4.1 inches. Comparison to Other Historic Events Gage height and discharge data at two USGS gages, Ramapo River at Pompton Lakes NJ and Pompton River at Pompton Plains NJ were examined for the recent March nor easter flood, as shown below in Figure 1, and compared to the same data for the recent past floods of the April 15-17, 2007 nor easter flood, the flood of April 4-5, 1984, Tropical Storm Floyd (September 15-17, 1999), April 3, 2005, and October 8-9, The comparison is provided in Table 1. Figure 1. Discharge Hydrographs Ramapo and Pompton Rivers. For the USGS gage, Ramapo River at Pompton Lakes NJ, the 40 year design flood of the recently completed Ramapo River at Oakland NJ Flood Protection Project was also examined to see how it compared to the recent March nor easter storm and flood as well as these earlier storms and floods. 4

6 The storms that produced these floods were also compared at two NOAA rain gages, Charlotteburg Reservoir NJ and Caldwell NJ Essex County Airport. Data gathered included total two week antecedent rainfall (inches), total storm rainfall in inches, and maximum storm rainfall intensity in inches per hour. The data is summarized in Table 1. For the two USGS stream gages, and the aforesaid seven floods, data consists of: Peak discharge, cfs, and peak gage height, feet Rank by size among floods in table, based on peak discharge and gage height Initial base flow in cfs and initial gage height in feet before each flood, as an indicator of drought to saturated initial condition The average rate of rise, in feet per hour, between the initial and peak gage height. The following conclusions can be drawn from the data presented in Table 1 The greatest average rate of rise observed among the six historical floods in Table 1 for the two USGS stream gages for which data is presented, is 0.40 feet per hour at the Ramapo River at Pompton Lakes NJ gage, for the April 1984 event, which was surpassed only by the hypothetical 40 year design flood. The average rate of rise of the March 2010 nor easter flood, 0.13 in/hr, was surpassed by the average rate of rise of all other floods in the table. A similar result is obtained at the Pompton River at Pompton Plains NJ, USGS stream gage, about a mile downstream of the Ramapo River at Pompton Lakes NJ gage. The average rate of rise of the March 2010 nor easter flood, 0.14 in/hr, was the lowest average rate of rise of all 6 historic floods in the table. In terms of pre-flood saturation, the March nor easter flood ranked third most saturated (third highest initial base flow) among the floods presented in Table 1, at both the Ramapo River at Pompton Lakes NJ, and the Pompton River at Pompton Plains NJ, USGS stream gages. 5

7 Table 1 March Nor'easter Storm and Flood - compared to five earlier recent storms and floods and the 40 year design flood (Ramapo River at Pompton Lakes only) GAGE PARAMETER : Apr- 84 FLOYD FLOOD : Apr- 05 Oct- 05 Apr- 07 Mar YR DESIGN Ramapo Peak discharge cfs 15,400 14,000 6,930 6,600 9,930 11,000 15,700 River Peak gage height ft At Rank among floods in table Pompton Initial flow, cfs Lakes Initial gage height, ft New Jersey Average rate of rise, feet per hour between initial and peak gage height : Pompton Peak discharge cfs 25,400 16,400 15,300 10,600 18,000 18,800 NONE River Peak gage height ft " at Pompton Rank among floods in Plains table " New Jersey Initial flow, cfs " Initial gage height, ft " Average rate of rise, feet per hour " between initial and peak gage height : Charlotteburg Total two-week antecedent rain, inches Reservoir Total storm rainfall, inches New Jersey Maximum hourly rain inches Caldwell New Total two-week Not Jersey antecedent rain, inches available Total storm rainfall, Not Essex County Airport Notes : inches Maximum hourly rain inches available Not available Ramapo River at Pompton Lakes NJ gage height is, for all floods in Table One, at the new gage downstream of the Pompton Lakes Dam. 2. The 40-year design storm is a 40-year hypothetical storm with no antecedent storm precipitation. One inch initial loss was assumed for the watershed to be fully saturated before the maximum hour of precipitation. 3. The maximum hourly rain in inches for Charlotteburg Reservoir NJ rain gage, 0.46 and 0.31 inch per hour, are estimated values. 6

8 Discharges jumped up abruptly from 1110 cfs to 1490 cfs at 1800 hrs Friday March 12, 2010, and again from 1510 cfs to 1940 cfs at about 0545 hrs on Saturday March 13, 2010 at the Ramapo River at Pompton Lakes NJ USGS stream gage, due to gate operation. Bank full discharge downstream is about 3500 cfs. There were, however, no abrupt upward jumps due to gate operation at or near the peak discharge of 11,000 cfs which occurred at noon on Sunday March 14, 2010 See Figure 1. The following comparisons can be made in terms of storm rainfall : At the Charlotteburg Reservoir NJ rain gage (in the Pequannock River Basin, to the west of the Ramapo River Basin : The estimated total antecedent precipitation, the snowfall of February 2010, with a water equivalent of 5.31 inches, for the March nor easter storm, is the largest of all the storms in the table. The estimated March storm total of 5.0 inches is exceeded by that of the April 4-5, 1984 storm (6.1 inches), Floyd (8.1 inches), October 7-9, 2005 (9.2 inches), April 15-17, 2007 (5.7 inches) and the 40 year design storm (5.8 inches). The March storm estimated maximum hourly intensity of 0.31 inch per hour is exceeded by that of all the other storms for which data is presented in Table 1 at this rain gage, ranging from 0.46 inch per hour for the April 15-17, 2007 storm to 1.77 inches per hour for the 40 year design storm. At the Caldwell NJ Essex County Airport rain gage just south of Wayne NJ and the Ramapo and Pompton River Basins: The estimated total antecedent precipitation, the snowfall of February 2010, with a water equivalent of 4.1 inches, for the March 12-15, 2010 nor easter storm, is the largest of all the storms in the table. However, it is only 4.1 % larger than the second largest 3.9 inches antecedent rainfall for the April 2005 storm. 7

9 The March storm total of 3.82 inches is exceeded by that of Floyd (10.82 inches), the October storm, 6.63 inches, and the April 2007 storm, 5.82 inches. Data was unavailable at this rain gage for the April storm. The March storm maximum hourly intensity of 0.24 inch per hour was exceeded by 1.77 inches per hour for Floyd, 1.06 inch per hour for the October 7-9, 2005 storm, 0.47 inch per hour for the April 2007 storm, and 0.26 inch per hour for the April 2005 storm. Data was unavailable at this rain gage for the April 4-5, 1984 storm. ANALYSIS OF THE EVENT Development of the Inflow Hydrograph Since the purpose of this report is to determine if the operation of the gates contributed to flooding downstream of Pompton Lake Dam as compared to the pre-project condition, a precise estimate of the inflow hydrograph is needed. Once the inflow hydrograph is developed, the flood event can be routed through the pre-project conditions structure (the 300 ft fixed spillway without gates). The pre-project outflow hydrograph would then be compared to the post project outflow hydrograph as measured by the USGS Gage Ramapo River at Pompton Lake, 400 ft downstream of the dam. For this analysis, the inflow hydrograph was developed from the outflow gage data recorded by USGS Gage Ramapo River at Pompton Lake, 400 ft downstream of the dam (Figure 2). Since the lake level at the beginning and end of the event are roughly equal, (Figure 3), the volume of inflow to the gates is assumed to be equal to the outflow volume. In other words, the area under the inflow hydrograph is equal to the area under the outflow hydrograph. Previous simulations during the design phase (Ramapo River at Oakland, New Jersey Flood Protection Project, Phase II General Design Memorandum (GDM), May 1994) showed that the outflow hydrograph closely follows the inflow hydrograph, with the inflow hydrograph essentially splitting the steps of the outflow hydrograph associated with each gate opening (Attachment 1). 8

10 Figure 2. Discharge Hydrograph Ramapo River Downstream of Pompton Lake Dam Figure3. Pompton Lake Water Surface Elevation 9

11 The design analysis in the GDM used the future improved discharges at hydrologic node R17G located at Pompton Lake Dam as inflow to the gate routings. This hydrograph takes into account the increase in discharge due to any loss in storage resulting from all project improvements. Comparisons to the pre-project conditions discharge at the dam and immediately upstream of the lake indicate that the storage effects of the lake increased the 10 year peak discharge by only 40 cfs, or 0.42% of the peak flow. The upstream channel work increased the 10 year peak discharge upstream of the lake by an additional 20 cfs, or 0.21 % of the peak flow. The detailed development of the inflow hydrograph is based upon the reservoir routing concepts that Average Inflow - Average Outflow = Change in Storage. The change in storage is determined from the elevation of the lake, assumed to be a flat pool. This procedure was employed utilizing the 15 minute data from the USGS gages. The USGS Gage Ramapo River at Pompton Lake Dam provided outflow data, and the USGS Gage Ramapo River above the Dam at Pompton Lakes NJ provided storage data (converting stage to storage volume using the Pompton Lake stage storage curve (Figure 4). Figure 4. Pompton Lake Stage vs Storage The results, as shown in Figure 5a,b below for both the entire hydrograph and just the peak, show a sawtooth shaped inflow hydrograph due to minor variations in the lake level. However, 10

12 certain trends can be implied. The general shape of the inflow hydrograph closely resembles the outflow hydrograph, with the peak inflow just slightly below the outflow. Figures 5 a,b. Pompton Lake Dam Outflow and Estimated Inflow The outflow hydrograph was also examined step by step after gate operations. As shown on Figure 6, the initial increase in discharge resulting from a gate opening was followed by a decrease in discharge for a few time steps. Discharges leveled off then increased. Figure 6. Estimated Inflow and Outflow Hydrographs 11

13 These changes in discharge are a direct reflection of the lake level, since the discharge is a function of the head over the spillway and gate. So, a reduction in discharge indicated a lowering of the lake level, meaning the outflow was greater than the inflow. Conversely, as the shape of the curve starts to rise, the lake level is increasing, and the inflow is greater than the outflow. At the local minimum point in the curve, the lake level is stable, and the inflow is exactly equal to the outflow. The same general concept was applied to the falling limb of the hydrograph. By analyzing each of these steps, and connecting the points of known inflow (where it is equal to the outflow), the general shape of the inflow hydrograph can be determined. Based on these parameters, a determination of the inflow hydrograph was made. The results of this analysis indicate that the peak inflow at the gates was approximately cfs. Based upon the peak discharge vs. frequency in the GDM, this event was approximately equivalent to the 17 year future improved hypothetical event, As a check, a comparison was made between the volumes of the estimated inflow hydrograph and the measured outflow hydrograph. The volume of the discharge hydrograph, allowing for base flow, was 43,162 ac ft. The volume of the estimated inflow hydrograph was within 0.01%. The inflow and outflow hydrographs are plotted together in Figure 7. Figure 7. Inflow and Outflow Hydrographs. 12

14 Comparison of Inflow and Outflow Hydrographs A comparison of the inflow and outflow hydrographs to the gates indicates that the peak outflow of 11,000 cfs, as determined from downstream gage data, was 200 cfs greater than the peak inflow of cfs. This peak outflow occurred when the gate opened 0.25 ft to 7.75 ft, the maximum opening for this event. The gates remained at this opening for 3.25 hours, at which time the gates were closed 0.25 ft to 7.5 ft. The timing of the peaks was such that the peak outflow occurred 2.25 hours prior to the estimated peak inflow. Gate Operation The Pompton Lake reservoir is regulated to provide flood damage reduction benefits to the reach immediately upstream of the lake as well as to satisfy the existing water supply operation. Reservoir regulation to prevent upstream damages is a unique situation; the majority of flood damage reduction reservoirs are regulated to prevent damages downstream of the reservoir by retaining peak discharges. As would be expected, the operation of the Pompton Lake Dam gates to release flow is quite different than most gate operations. After the April 07 flood event, and subsequent gate operations during the first year the gates were on line, it was apparent that operations of the gates could be modified for a more efficient operation and to reduce downstream impacts. The initial gate opening lake level of resulted in frequent operations of the gates, sometimes with every rainfall event on saturated ground conditions. It also did not allow the lake level to rise high enough to pass floating debris, and as a result, the dam crest was lined with tree limbs, etc. Also, the minimum lake level was 0.5 ft below the dam crest, which was too low to permit safe operation of the NJDWSC Ramapo Pumping Station. Another issue of concern was the minimum incremental gate opening of 0.5 ft. The initial gate opening of 0.5 ft resulted in an increase in discharge of about 850 cfs over a base flow of about the same. The resulting sudden increase in downstream water levels was alarming. As a result, the following modifications to the gate operation were programmed, and were in effect during the March 2010 event: 13

15 The initial gate opening lake elevation was raised to The minimum lake level was set to 201. The minimum gate opening was reduced to 0.25 ft. The gate operation is a function of the difference between the lake elevation and the target elevation. If the lake level is more than 0.25 ft above the target elevation, the gates are to be opened 0.25 ft; if the lake level is more than 0.5 ft above the target elevation, the gates are to be opened 0.5 ft; if the lake level is more than 1.0 ft above the target elevation, the gates are to be opened 1.0 ft. Similarly, if the lake level is more than 0.25 ft below the target elevation, the gates are to be closed 0.25 ft; if the lake level is more than 0.5 ft below the target elevation, the gates are to be closed 0.5 ft; if the lake level is more than 1.0 ft below the target elevation, the gates are to be closed 1.0 ft. The gates can be adjusted every 15 minutes to avoid any large fluctuations in lake pool and provide outflow rates essentially the same as inflow rates for a broad range of runoff hydrographs, and to minimize sudden increases in discharge that would represent a threat to life or property. A full description of the gate operations can be found in Appendix B of the OMRR&R Manual, and is attached for reference. Automatic operations initiated at 1800 hours on 12 March 2010 when the lake level reached an elevation of or 0.25 ft above the target elevation resulting in an initial gate opening of 0.25 ft. The gates continued to open gradually, until a maximum gate opening of 7.75 feet was reached on 0910 hours on 14 March. Automatic operation of the gates continued, and the program closed the gates on 0928 hours on 18 March. However, as the gates began to close from 0.75 feet to 0.5 feet, a log which had became lodged underneath Gate No. 1, prevented the gate from closing beyond 0.57 feet open. Gate No. 1 was disengaged from automatic mode and remained at 0.57 feet open. NJDWSC, in consultation with NJDEP, attempted to remove the log. The removal was complicated and required multiple attempts. The log was successfully removed on March 26. Gate No. 2 was unaffected by this event and continued to operate in automatic mode without incident. In general, the gates opened in 0.25 ft increments. However, between 1039 hours on 13 March and 0139 hours on 14 March, when the lake was more than 0.5 ft above the target elevation, the 14

16 gates opened in 0.5 ft increments from 3.0 to 5.5 ft. As the lake level continued to rise, the gates opened in 0.25 ft increments until they reached the maximum opening of 7.75 ft, and remained at that opening through the peak of the inflow hydrograph. As the inflow decreased, the lake level fell to 0.25 ft below the target elevation and the gates began to close in 0.25 ft increment. A complete history of the gate opening for the March event is shown below in Figure 8. Supervisory Control and Data Acquisition (SCADA) data is recorded every few seconds throughout the gate operation. A review of the SCADA data indicates that, in general, the gates operated in accordance with the revised Gate Operation Rule Curve. The initial gate operation occurred when the lake level exceeded as recorded in the SCADA data. The minor difference between the recorded SCADA lake level and the USGS gage data (approx. 0.2ft) had no impact on downstream levels. Figure 8. Gate Opening Time Series (in ft) Gages The gages initially installed during construction were susceptible to drifting, whereby the readings would vary increasingly over time from actual levels. At that time, USGS, being the experts in the field of stream gages, was tasked with providing gages similar to those used throughout the region. They also provided a secondary back up gage for redundancy. Prior to the March 2010 event, lightning strikes in the area caused spikes in the readings which affected gate operations, opening the gates unexpectedly, and eventually taking those gages offline 15

17 temporarily. At the time of the March 2010 event, repairs in the vicinity of the USGS gages station transferred gages for the gate operation back to the original sensors at the dam. A comparison of the lake level data between the SCADA data from the gate operating system recorded during the event to the official USGS data for Gage , indicates that the gate operating sensors read approximately 0.2 ft higher than actual levels, which caused the gates to open at a lake level of about rather than This theoretically would open the gates a few minutes sooner and result in slightly lower discharges due to the lower head. However, for this event, such deviations would be negligible. There were also issues with other gages on the Ramapo River in the vicinity of Pompton Lake Dam during the March 2010 event. The USGS Gage , Ramapo River at Dawes Highway at Pompton NJ, provides river stage data. Even though the gage was functioning properly, the incorrect gage datum listed on the website resulted in an erroneous conversion of the data which indicated river stages higher than actual. The USGS gage , Ramapo River at Railroad Bridge at Oakland was installed to provide an indication of the inflow to Pompton Lake. However, during this event, debris carried by the swift moving waters dislodged the gage from its mount, resulting in erroneous data readings. It should be noted that both of these gages were installed as part of the Ramapo Oakland NJ Flood Protection Project for information purposes only and not to be used for the gate operation. Although these gages malfunctioned, they did not effect the operation of the gates. As of this report, all gages have been repaired and are functioning correctly. Downstream Impacts of Gate Operations In order to get a true evaluation of the impacts of the gate operations, an estimate of the preproject downstream discharges had to be developed. This would be accomplished by a simulated routing of the inflow hydrograph over the pre-project 300 ft fixed spillway. However, before that could be done, a simulation of the as-built project was needed to confirm the accuracy of the inflow hydrograph. That simulation yielded a peak outflow of cfs, with a maximum gate 16

18 opening of 7.5 ft, closely reproducing the actual outflow of cfs and gate opening of 7.75 cfs, and therefore verifying the accuracy of the inflow hydrograph. That same inflow hydrograph was then routed over the fixed, pre-project spillway. The resulting peak discharge was cfs, with a maximum lake level of ft. The 214 cfs increase in downstream discharge (11,000-19,786 cfs), which can be attributed to the gate operation, resulted in an increase in water level of 0.25 ft at the downstream gage and less than 0.1 ft at the Dawes Ave gage 3000 ft downstream. The results of these simulations can be seen below in Figure 9. A summary of the simulations is shown in Attachments 2 and 3. As previously stated, the maximum lake elevation for the pre-project condition would have been The lake level immediately prior to the event was approximately From Figure 4. Pompton Lake Stage vs Storage, the storage lost as a result of the gate operations is approximately 856 ac-ft, which is less than 2% of the total volume of runoff of 43,162 ac-ft for the event. The results of the simulations also indicated a shift in the timing of the hydrograph. While the peak of the pre-project discharge hydrograph was delayed about an hour from the peak of the inflow, the peak of the gate operation occurred 2.25 hours sooner that the peak inflow, occurring at the time of the last gate opening. Throughout the rising limb of the hydrograph, the shift between the pre and post project hydrographs was approximately 1.5 hours. In general the slopes of the rising limb of all 3 hydrographs (inflow, pre-project and post-project outflow), as shown in Attachments 2 and 3 are very similar, indicating a consistent rate of rise over time. 17

19 Figure 9. Simulated Discharge Hydrographs The Ramapo, Pequannock and Wanaque Rivers all flow together into the Pompton River. Fig. 10 shows the discharge hydrographs for all 4 rivers for the March event. As can be seen, the Ramapo River at Pompton Lake peaked approximately 5 hours before the Pompton River at Pompton Plains. Under pre-project conditions, the Ramapo would have peaked 3 hours later, and only 2 hours before the Pompton River. To estimate the impact of the gate operations on the peak discharge of the Pompton River at Pompton Plains, the simulated with project Ramapo River hydrograph (Qpeak=10994) was subtracted, and the without project hydrograph (Qpeak=10786) was added to the Pompton River hydrograph. The results show that the shift in the timing of the Ramapo hydrographs essentially cancelled out the increase in discharge resulting from the gates. The operation of the gates had no impact on peak stages in Pompton Plains. 18

20 Figure 10. Ramapo, Wanaque, Pompton and Pequannock Rivers Hydrographs A review of the USGS Gage , Ramapo River at Dawes Highway shows several instances where the river stage increased rapidly, in some cases over 3 inches in 15 minutes. These increases coincide with the gate openings at Pompton Lake Dam. The highest increases of 0.27 ft occurred early in the rising limb of the hydrograph when the river stage was within bank, and was followed by a period of several hours of stable, or falling river stages. The largest out of bank increase in river stage (0.25 ft in 15 minutes) occurred the first time the gates opened 0.5 ft, during the steepest part of the hydrographs when the inflow was increasing at its fastest rate. The next 4 gate openings of 0.5 ft, all of which occurred within a 3 hour period, resulted in an average increase of 0.14 ft in 15 minutes. The consistency in the slopes of the outflow hydrographs for pre and post project conditions (Attachments 2,3) would infer that the average rate of rise in the river stage at Dawes Hwy would be similar. The Ramapo River at Dawes Hwy stage hydrograph is shown in Figure 11. Also shown are the Ramapo River stage hydrographs upstream (lake level) and downstream of the Dam for comparison. 19

21 Figure 11. Ramapo River Stage Hydrographs At Dawes Hwy, approximately 3000 ft downstream of the gates, the peak flow was assumed to be the same as at the dam. This is a reasonable assumption since any increase in discharge due to the additional contributing drainage area could be offset by additional attenuation in the flood plain. Utilizing the rating curve immediately downstream of the Dawes Ave Bridge (Figure 12.), an increase of 214 cfs would result in an increase in stage of less than 0.10 ft. It should be noted that the estimate of a 214 cfs increase in discharge at Dawes Hwy is conservative, since the volume of flow in the flood plain would result in an attenuation of the peak caused by the gate opening. 20

22 Figure 12. Rating Curve on Ramapo River at Dawes Hwy Neither the inability of Gate No. 1 to close completely or the actions to remove the log increased downstream flooding. The incident had no negative impacts downstream because the incident occurred when flooding along the Ramapo River had ended and the river was well within the stream bank. CONCLUSION and RECOMMENDATIONS The March nor easter storm, and the flooding it produced on the Ramapo and Pompton Rivers, downstream of the recently completed Ramapo River at Oakland NJ flood damage reduction project, though severe and extensive, was surpassed by the April 1984 and Floyd floods on the Ramapo River and by the April 1984 flood on the Pompton River, in terms of peak discharge and gage height. Without the effect of gate operation, it was surpassed in terms of average rate of rise by all the floods it is compared to in Table 1, at the Ramapo River at Pompton Lakes NJ USGS stream gage. Further downstream, at the Pompton River at Pompton Plains NJ USGS stream gage, it was surpassed in terms of average rate of rise by all the floods it is compared to in Table 1. 21

23 In general, the gates operated in accordance with the revised Gate Operation Rule Curve, aside from the minor difference between the SCADA data and the USGS gage (approx. 0.2ft). The discharge hydrograph, lake levels and gate openings were in close agreement with the simulations of the event. These findings indicate that the changes in the timing and flow due to the operation of the flood gates during the March flood had no significant impact downstream. The 214 cfs increase in discharge attributed to the operation of the gates resulted in a minor (0.25 ft) increase in peak stage in the pool immediately downstream of the gates. The impact on the Ramapo River at Dawes Highway was negligible (less than 0.10 ft) and is not within the accuracy of any modeling effort. Differences of this amount are considered to represent a no-change condition. There was no impact on peak stages at the Pompton River. The effect on shift in the timing of the event, from 1.5 to 3.25 hours in an event that lasts several days is not considered to be an increased hazard caused by the gate operation. The issues arise with perception and expectations the system was expected to react as it has in the past, and it reacted differently. Once it is understood how the rivers will respond to the gate operations, then appropriate actions can be taken. Significant improvements to the gate operation program were made following the April 07 event. In an effort to further reduce downstream impacts, the following options can be studied: Reduce the minimum gate opening Reduce the time increments for gate operations Incremental gate openings Adjustments to the Target Elevation. Combinations of the above The debris that accumulated around the floodgate facility during the event is concerning. Such debris accumulation increases the likelihood of a tree/log becoming lodged beneath one or both floodgates during a future event. Ways to minimize and/or control the accumulation of debris around the floodgates should be studied. In the interim, the OMRR&R should be updated to include procedures for removal of debris from beneath a floodgate. 22

24 23

25 24

26 25

27 Appendix B GATE OPERATING RULE CURVE OMRR and R Manual March 2008 Pompton Lake Dam, NJ Appendix B-Gate Operating Rule Curve

28 POMPTON LAKE DAM GATE OPERATION RULES Current Gate Opening <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< <Lake Elevation> >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 0 Below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above and below and above ACTION - LAKE LEVEL FALLING>>> Close 1 ft. Close 0.5 ft Close 0.25 ft No Action No Action No Action No Action ACTION - LAKE LEVEL RISING>>>> No Action No Action No Action No Action Open 0.25 ft Open 0.5 ft Open 1 ft.

29 RAMAPO OAKLAND, NEW JERSEY GATE OPERATION The gates are operated to maintain a set point pool elevation in Pompton Lake. The set point elevations are defined as a function of the gate opening and lake level, and vary from at a gate opening of 0 ft to at gate openings equal to or exceeding 4.5 ft as defined in the Gate Operation Rule Curve. Elevations are referenced to the National Geodetic Vertical Datum (N.G.V.D.). The gate operation is a function of the difference between the lake elevation and the set point elevation. If the lake level is more than 0.25 ft above the set point elevation, the gates are to be opened 0.25 ft; if the lake level is more than 0.5 ft above the set point elevation, the gates are to be opened 0.5 ft; if the lake level is more than 1.0 ft above the set point elevation, the gates are to be opened 1.0 ft. Similarly, if the lake level is more than 0.25 ft below the set point elevation, the gates are to be closed 0.25 ft; if the lake level is more than 0.5 ft below the set point elevation, the gates are to be closed 0.5 ft; if the lake level is more than 1.0 ft below the set point elevation, the gates are to be closed 1.0 ft. The only exception to these rules is that if the lake level is falling, then the gates would not be opened. Similarly, if the lake level is rising, then the gates would not be closed. These exceptions will minimize lake level oscillations. Under normal conditions, the gates remain closed and all outflow passes over the fixed spillway portion of the dam. The gates are initially opened 0.25 ft when the lake elevation reaches 202.5, or 0.25 ft above the target elevation of for a gate opening of 0. Initial gate opening will be controlled by the PLC/RTU automatic control equipment if the controls are in SCADA AUTO, or can be remotely controlled from either of the two SCADA PC work stations, or locally controlled from the gate control console cabinets. Every 15 minutes thereafter, the lake level and stream flow gauges are checked, and the gates operated automatically if necessary. The two gates can be opened simultaneously, or sequentially, but they should maintain the same opening. OMRR and R Manual March 2008 B-1 Pompton Lake Dam, NJ Appendix B-Gate Operating Rule Curve

30 Defining the gate operation as a function of gate opening and lake level allows the gates to be operated manually, without the benefit of computers and gauge data, should those systems become inoperable for any reason. Under such a situation, the gate operator need only check the gate opening and staff gauge at the dam site, and then plot that data on the rule curve to determine the required gate operation. Should one gate fail to operate, the gate operation would follow the same rules as the normal two gate operation. The one gate will simply open faster since the rules will make that gate open more frequently. OMRR and R Manual March 2008 B-2 Pompton Lake Dam, NJ Appendix B-Gate Operating Rule Curve

31 OMRR and R Manual March 2008 B-3 Pompton Lake Dam, NJ Appendix B-Gate Operating Rule Curve

32 SEQUENCE OF OPERATIONS TAINTER GATES A. INITIATION OF GATE OPERATION 1. Lake level reaches 202.2, alarm sounds at Water Commission HQ in Wanaque. 2. Lake level reaches (approx. 30 min after alarm sounded) a. A warning is sounded. b. Gate operation is initiated automatically or manually. 3. At 15 minute intervals: a. Gate operation continues automatically or, b. Operator can manually override gate operation as necessary. B. RETURN TO NORMAL OPERATION 1. Normal gate operation closes gates. 2. Operator dispatched to site to inspect and lubricate all equipment. OMRR and R Manual March 2008 B-4 Pompton Lake Dam, NJ Appendix B-Gate Operating Rule Curve