Revising the Historic Monthly Mean Niagara River Flow at Buffalo to Reflect Better Estimates of Maid-of-the-Mist Pool Outflows

Transcription

1 Revising the Historic Monthly Mean Niagara River Flow at Buffalo to Reflect Better Estimates of Maid-of-the-Mist Pool Outflows Background The Niagara River flow at Buffalo is currently computed by summing the flow diverted for hydro-electric power generation at the high head plants at Queenston-Lewiston, the outflow from the Maid-of-the-Mist Pool (MOM), and the diversion through the New York State Barge Canal, and subtracting local inflows and the portion of Welland Canal Diversion returned to the river upstream of the Falls. The outflow from the Maid-of-the- Mist Pool is determined with rating equations using the water level recorded at the Ashland Avenue water level gauge. The equations presently being used were approved in These equations were revisions of a previous equation approved in Discharge measurements have been carried out routinely to verify the 1981 rating equations, and it was determined that the 1981 equations were underestimating the flow. In September 2008, the Niagara Board approved a revised rating equation. Prior to 1964, several equations were used to determine the MOM outflow. In 1965, the Coordinating Committee on Great Lakes Basic Hydraulic and Hydrologic Data revised past Niagara River flow estimates using the 1964 Ashland Avenue rating equation for MOM outflow. The Ashland Avenue gauge was installed in 1957, and replaced the Morrison Street gauge for computation of flow out of the MOM. The Coordinating Committee adapted the 1964 Ashland Avenue rating equation to use levels at Morrison Street gauge and re-derived the MOM and Niagara River flows from 1926 to The 1964 equation was derived using 556 discharge measurements from 1907 to The measurements before 1952 were made near the head of the river, above Grand Island. The measurements from were made in the lower river, below where the present OPG Beck plants discharge back into the river. The flow out of the MOM at the time of these measurements was determined, as appropriate, by subtracting the Beck plant diversions (or discharges for ), subtracting the New York State Barge Canal diversion and adding local inflows, to the measured total flow. During these measurements the MOM outflow ranged from 1416 to 6940 cubic meters per second (cms). Most of these measurements were made at MOM outflows well above the flows experienced since OPG s BeckII and NYPA s Robert Moses hydropower plant began pulling water from the river above the falls, in 1955 and 1961 respectively. NYPA s plant was fully operational by Prior to 1962 the monthly mean outflow from the MOM ranged from about 2940 to about 6700 cms. Since 1962 the range has been from about 1420 to about 3170 cms, only half of what it had been. Beginning in 1973, discharge measurements were made at a new measuring section in the lower river the Cableway Section. This section is upstream of the hydropower plants and measures only the flow out of the MOM pool. After several sets of measurements were made it was decided to revise the Ashland Avenue rating equation using only discharge measurements at the Cableway Section and water levels at Ashland Avenue. The resulting 1981 Ashland Avenue rating equations were based on 70 measurements 1

2 made between 1973 and When the 1981 equations were adopted, beginning in April 1981, the historic Niagara River flows prior to April 1981 were not corrected to reflect the revised MOM rating. Continued routine discharge measurements of the outflow from the MOM, particularly those using Acoustic Doppler Current Profilers (ADCP) indicated that the 1981 Ashland Avenue rating equations were underestimating the MOM outflow particularly in the higher flow ranges. The rating was re-evaluated and a new equation determined. With the acceptance of the September 2008 revised Ashland Avenue rating, it became necessary to consider revising historic flows to reflect better estimates of the MOM outflow. Because the 2008 rating equation, like the 1981 equations, was derived using only measured flows in the 1400 to 3700 cms range, it only reflects the MOM outflow since the NYPA plant came on-line. Therefore, only historic flows from 1961 to present will be considered for revision. In this report the term published refers to the historic flows published by the Coordinating Committee in its report, Lake Erie Outflow, , with Addendum , and to flows that continued to be coordinated between the office of the U.S. Army Corps of Engineers (USACE), Detroit and Environment Canada (EC), Burlington. Objective The objective of this study (being done for the International Upper Great Lakes Study (IUGLS)) is to determine revised historic monthly mean Niagara River flows that reflect better estimates of MOM outflow. To do this the difference between the 1981 or 1964 equation estimate and the 2008 equation estimate was determined and applied to the historic/published Niagara River flows from 1961 to Several methods to determine this difference were evaluated. The availability of historic water level data in digital format was a limiting factor. Analysis There are limits to the amount and kind of historic data that is readily available for recomputing the MOM outflow. Hourly Ashland Avenue water levels are available from the National Oceanic and Atmospheric Administration s (NOAA) website from 1970 to present. For hourly Ashland Avenue data was obtained directly from Cary Wong of NOAA s National Ocean Service. For 1961, only monthly mean levels were available digitally. Presently, the MOM outflow is determined by a computer program utilized by the USACE, which uses hourly Ashland Avenue water levels and considerations for the transitions in the pool when Falls flows are switched during the tourist season. This program also computes the Niagara River flow at Buffalo, the Lake Erie outflow and the Lake Ontario inflow. The program requires data in addition to Ashland Avenue water 2

3 levels, such as the hourly diversions for power generation, and much of the older data is not readily available digitally. In the hopes of simplifying the method for determining revised historic Niagara River flows, several alternative approaches to retrofitting the program and obtaining all the past data it requires were tested. They all involved determining the difference between monthly flows computed using the historic MOM outflow equations and those computed using the revised 2008 equation. These differences would be used as the adjustments to the historic monthly mean Niagara River flows. One alternative used hourly Ashland Avenue water level data, another just used monthly mean Ashland Avenue water levels in the rating equations. The goal was to see how well hourly or monthly mean levels could represent the difference between the old and new MOM outflow estimates. Since the best estimates of MOM outflows are obtained using the Lake Erie outflow program, it was used to create a basis of comparison for the other methods. For the current format of the program, data was only available from The current program was changed to use the revised 2008 Ashland Avenue equation and run for the period These results were compared to the program results using the 1981 rating equations. The differences between the MOM estimates using the program with the two different ratings became the basis of comparison. These differences were compared to using hourly or just monthly Ashland Avenue water levels and the 1981 and 2008 rating equations, without considering the flow transitions during the tourist season, which the program does. A summary of the results is shown in Table 1. It is important to keep in mind that until 1993 Niagara River flows were coordinated to the nearest thousand cubic feet per second (tcfs) and to the nearest ten cubic meters per second (10 cms), thereafter. Using hourly water levels gives essentially the same monthly average difference between the two MOM estimates (1981 and 2008 rating equations) as using the program. The difference between using the program and using monthly mean water levels is less than 10 cms, but rounds to 10 cms during the summer months. Since hourly Ashland Avenue water levels originally were not readily available before 1970, and at that time the rating equation used for historic flow computation was the 1964 equation, another comparison was performed. This one used both hourly and monthly mean Ashland Avenue water levels for the years to compute flow differences between the 1964 and 2008 equations, and compared the resulting differences. Table 2 summarizes these comparisons. There seems to be a fairly consistent pattern where the differences are near the same from November through March and about 10 cms different from April through October. Using hourly Ashland Avenue water levels consistently resulted in differences about 10 cms higher than using monthly mean water levels from April to October. The same pattern was seen for the period once hourly water levels become available. It seems advisable that, if using monthly mean levels to compute the difference between the MOM outflows using the 1964 and 2008 equations, 10 cms should be added to the difference for the months of April to October. This would compensate for having to use monthly mean levels when hourly levels are not available. This effects only the adjustments for 1961, because hourly data was eventually obtained for

4 Table 1 Differences (in cms) Between MOM Outflow Estimates Computed Using 1981 and 2008 Ashland Ave. Rating Equations 2005 to 2007 Based on Program Based on Hourly WL Based on Monthly WL Program Diff - Hourly Diff Rounded to 10 cms Program Diff - Monthly Diff Rounded to 10 cms 2005 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

5 Table 2 Differences (in cms) Between MOM Outflow Estimates Computed Using 1964 and 2008 Ashland Ave. Rating Equations 1970 to 1975 Year Month Based on Hourly WL Based on Monthly WL Hourly Diff - Monthly Diff Year Month Based on Hourly WL Based on Monthly WL Hourly Diff - Monthly Diff Jan Jan Feb Feb Mar Mar Apr Apr May May Jun Jun Jul Jul Aug Aug Sep Sep Oct Oct Nov Nov Dec Dec Jan Jan Feb Feb Mar Mar Apr Apr May May Jun Jun Jul Jul Aug Aug Sep Sep Oct Oct Nov Nov Dec Dec Jan Jan Feb Feb Mar Mar Apr Apr May May Jun Jun Jul Jul Aug Aug Sep Sep Oct Oct Nov Nov Dec Dec Hourly Diff - Monthly Diff Rounded to 10 cms Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

6 Methods for Determining MOM Outflow Adjustments, Prior to 1926, the Niagara River flow at Buffalo was not computed using the MOM outflow, so no adjustments are needed for the historical flows. For the period 1926 to 1960, the 1964 rating equation was the best representation of the high MOM outflows of this period. As noted previously, the Coordinating Committee used the 1964 equation to determine the published flows for their report. Only for the period 1961 to present, were adjustments determined and applied to the published Niagara River flows. These adjustments were derived by computing MOM outflow using the revised 2008 rating equation and subtracting the flow computed using the MOM outflow rating equation used in the determination of the published Niagara flow for that time (either the 1981 or 1964 equation). The resulting differences, rounded to the nearest 10 cms, will be applied to the published flows. For some periods, additional adjustments may have also been made, as described below. For the period 1962 to present, hourly Ashland Avenue water levels are available, and were used in the computations. Using hourly water levels and the rating equations to determine the difference between the two estimates of monthly MOM outflow and applying this difference to the published monthly Niagara River flow is about as good as using the Lake Erie Outflow program to determine a revised monthly Niagara River flow. For the period April 1981 to present adjustments were determined from MOM outflows computed using: 1) hourly Ashland Avenue water levels, meters, IGLD 1985; 2) the 1981 Ashland Avenue rating equations, in metric and IGLD 1985; 3) the revised September 2008 Ashland Avenue rating equation. For the period 1962 to March 1981 adjustments were determined from MOM outflows computed using: 4) hourly Ashland Avenue water levels, meters, IGLD 1985; 5) the 1964 Ashland Avenue rating equations, in metric and IGLD 1985; 6) the revised September 2008 Ashland Avenue rating equation. For 1961, hourly Ashland Avenue water levels could not found so monthly mean levels were used. An additional adjustment was applied, depending on the time of year. From the tests done for the period , and comparisons done once hourly levels became available for , it appears reasonable to add an additional 10cms to the rounded differences for the months April through October. To summarize, for 1961 the adjustments were determined from MOM outflows computed using: 1) monthly mean Ashland Avenue water levels, meters, IGLD 1985; 2) the 1964 Ashland Avenue rating equations, in metric and IGLD 1985; 3) the revised September 2008 Ashland Avenue rating equation; and 4) an additional 10 cms applied to the rounded adjustments for the months of April through October. 6

7 The Equations and Datum Conversions Revised 2008 Ashland Avenue Rating Equation The MOM outflow rating equation approved by the Niagara Board at its meeting in September 2008 is as follows, where Q is MOM outflow in cms and AA is water level at Ashland Avenue gauge in meters on IGLD Rating Equation: Q = * (AA ) Ashland Avenue Rating Equation The 1981 Ashland Avenue rating consists of two equations. Which equation to use depends on whether the flow is at or above 1416 cms or below 1416 cms. The original equations were developed in English Standard units, with Ashland Avenue water levels on IGLD 1955: MOM outflows at or above 50,000 cfs: Q = *(AA ) 2 MOM outflows less than 50,000 cfs: Q = *(AA ) The 1981 rating equations converted to metric units with water levels on IGLD 1985 are: Flows at or above 1416 m 3 /s: Q = *(AA ) 2 Flows less than 1416 m 3 /s: Q = *(AA ) Ashland Avenue Rating Equation As reported in Proposed Revision of the Ashland Avenue Gauge Rating Equation for Maid-of-the-Mist Pool Outflow, the 1964 rating equation is: Q= * ( AA ) 2, where Q is in cfs, and AA 55 is the Ashland Avenue level in feet on IGLD The 1964 rating equation using levels at Ashland Avenue converted to metric and IGLD 1985 is: Q = * (AA ) 2, where Q is flow in cms, and AA 85 is the level at Ashland Avenue in meters, IGLD85 7

8 Recommended Revised Niagara River Flow at Buffalo Based on the above detailed procedure, differences between the old and new Ashland Avenue rating equations were determined and used to adjust the published/historic Niagara River flow at Buffalo from 1961 to Table 3 lists the derived adjustments. Table 4 gives the recommended adjusted monthly mean Niagara River flows at Buffalo for , as well as the coordinated (USACE, Detroit and EC, Burlington) published flows from The adjustments to MOM outflow are about 4-5% of the total MOM outflow. This is similar to what is seen in recent discharge measurements compared to the 1981 rating equation. Applying the adjustments to the published Niagara River flows increases them about 2%. 8

9 Table 3 Differences Between MOM Outflow Estimates Computed Using 1964 and 2008 Ashland Ave. Rating Equations To Be Used to Adjust Published Historic Niagara River Flows at Buffalo (Cubic Meters per Second, rounded to nearest 10 cms) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Adjustments based on: 1961: 1964 & 2008 MOM rating eq. and monthly mean levels at Ashland Avenue gauge. An extra 10 cms added to April through October to account for difference in using monthly mean water levels instead of hourly Mar 1981: 1964 & 2008 MOM rating eq. and hourly levels at Ashland Avenue gauge. April : 1981 & 2008 rating eq. and hourly levels at Ashland Avenue gauge. 9

10 Table 4 Recommended Monthly Mean Niagara River Flows at Buffalo With Flows for Adjusted to Reflect Better Estimates of Outflow from the Maid-of-the-Mist Pool (Cubic Meters per Second, rounded to nearest 10 cms) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

11 Table 4, Continued Recommended Monthly Mean Niagara River Flows at Buffalo With Flows for Adjusted to Reflect Better Estimates of Outflow from the Maid-of-the-Mist Pool (Cubic Meters per Second, rounded to nearest 10 cms) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

12 Table 4, Continued Recommended Monthly Mean Niagara River Flows at Buffalo With Flows for Adjusted to Reflect Better Estimates of Outflow from the Maid-of-the-Mist Pool (Cubic Meters per Second, rounded to nearest 10 cms) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Adjustments based on: : NO adjustments made 1961: Based on 1964 & 2008 MOM outflow rating eq. and monthly mean levels at Ashland Avenue gauge. An extra 10 cms added to April through October to account for difference in using monthly mean water levels instead of hourly Mar 1981: Based on 1964 & 2008 MOM outflow rating eq. and hourly levels at Ashland Avenue gauge. April : Based on 1981 & 2008 MOM outflow rating eq. and hourly levels at Ashland Avenue gauge. 12

13 Comparing the Adjusted Niagara River flows to Flows Computed using the 2001 Buffalo Rating Equation This comparison was only done for May and November, which are usually considered to be open water months. Figure 1 shows a comparison between the Niagara River flows computed using the 2001 Buffalo rating equation and the published Niagara River flows. Figure 2 is the same comparison but with the adjusted published flows. The period of comparison is because the 2001 Buffalo rating equation is for river conditions since the 1973 Directive changed the operation of the International Control Structure in the Chippawa-Grass Island Pool and consequently the hydraulic slope in the river. Figure 1 Comparison Published & Computed (2001Eq) Niagara River Flow Published Flow, cms May Nov Computed Flow, cms 13

14 Figure 2 Comparison ADJUSTED Published & Computed (2001Eq) Niagara River Flow Published Flow, cms May Nov Computed Flow, cms 14