YARDLEY TOWPATH (RIVER & CREEK FLOODING)
Flood Control "Pennsylvania Ave & Vicinity"
By Larry Hale - February 22, 2007 (Rev 5-7-07) with Elevations Supplement (Rev 5-7-07)

This article is focused on "Pennsylvania Ave & Vicinity" a hydraulically unique area alongside the canal and located on the upriver side of the Lower Makefield (Belmondo) line, a short walking distance along the canal from Macclesfield Park. "Flood Proofing" and "Flood Control" issues are presented for consideration once work is completed in the high priority "Morgan Ave & Vicinity" where houses are being elevated.

How does the canal fit into the flooding picture ?

Is it the canal that causes flooding ?

What are the facts ?

The perception by some that the canal causes flooding is only partly true and only in an indirect and limited way as described above.

Four (4) Highest Floods in Yardley (Last 100-Years)
[ Appendix A: Ten (10) Historic Flood Crests at Trenton ]

• River flooding in August 1955
  [ Appendix C: August 1955 Flood ]
  
  
• Creek flooding in June 1996
  [ Ref: Article Titled "Yardley Flooding" posted March 10, 2005 ]
  
  
• River flooding in April 2005
  [ Ref: Article Titled"APRIL 3-4, 2005 - Delaware River Flood"posted May 14, 2005 ]
  
  
• River flooding in June 2006
  Duplicate of the April 2005 flood except it was 6-in lower.

Floodwaters from each of these major floods entered the canal, looked for a low elevation weak spot, and caused a breach, which flooded Pennsylvania Avenue. Creek flooding presents a different set of problems but can also cause a canal breach as in June 1996. There are also ongoing lesser problems caused by creek flooding.

Creek Flooding - duration measured in hours:
The canal system below Lock #6 down to the railroad does not properly handle "creek-flooding" as originally intended because the various canal components that control the inflow and outflow have not been maintained, resized, or added as needed to keep up with the ever increasing storm runoff from local watersheds, as would have been done when the canal was fully operational. Some of the component inadequacies also apply to the "river-flooding" problem.

River Flooding - duration measured in days:
"Pennsylvania Ave & Vicinity" already remains dry for some river floods that inundate other parts of the Borough, and as indicated below has a number of inherent advantages that if exploited might keep it dry for even higher floods.

• Railroad Bypass
• Aqueduct Upgrade
• Canal Towpath Upgrade
• Overland Flow Study
• Letchworth Avenue Pipe Outlet
• Low Impact Development
• Upriver Flood Reduction
• L.M.T. Area 1 Study
• Coordination & Synergy

All these factors taken together suggest that "Pennsylvania Ave & Vicinity" not subjected to deep flooding in the first place, although this needs to be studied on a 2-ft contour map, may in the foreseeable future (10 years or more) be dried-up for flood levels up to the Apr 4, 2005 level. Proposed projects mentioned above will also reduce the frequency of creek flooding. Gradual adoption of Category III "Low Impact Development" to existing built-out will take longer [20-years or more] and it remains to be seen if the desirable goal of treating surface water as a resource instead of a waste product can be achieved. The towpath will always have some potential for river topping towpath because it was built low hoping to obtain free water from the Delaware River.

What's happening at the railroad?

Below Yardley a very long and wide swamp occupies a natural floodway and drains to Potato Creek near the water treatment plant, but it has been blocked just below Yardley by the railroad embankment. Floodwaters build up behind the embankment, flooding Letchworth Avenue and eventually rising high enough to bypass around both ends of the railroad embankment, as follows:

• Left End – flood is diverted to river as follows: 1) Drains through the man-made Belmondo retention basin (drains slowly) and 2) Overflows over the Belmondo entrance/exit road (both 1 and 2 bypassing the Macclesfield swamp and going straight to the river.

• Right End – The larger part of the flood scours out the towpath below Lock 5 and discharges through the railroad arch to the swamp behind Macclesfield. In the case of creek flooding there is additional large outflow from College Ave to Brock Creek because the ground slopes in the upriver direction.

FLOOD LEVEL DIFFERENCE BETWEEN CANAL AND RIVER
The April 2005 flood level in the "Pennsylvania Ave & Vicinity" basin was about 2-ft higher than the flood surface of the river at College Avenue and Belmondo. There are a number of variables, which might account for this. In August 1955 it was an additional 3-ft higher in the immediate vicinity of the canal but this did not carry over to the high ground along the river. One of the possible variables in 1955 was the Yardley Wagon Bridge. The possible effect of the bridge became strikingly evident when I read the following account:

Those upstream from the Yardley-Wilburtha Bridge can plainly see the mass of debris that has collected against it. The bank of wreckage has been building all night, with the majority of it being eddied over to the eastern bank. The pressure created by its water resistance threatens to collapse the bridge. The water now washed fully over its deck, and all parts of the bridge ---railings, spans, and piers--- are taking a tremendous beating."

"At about eight a.m. [Saturday, August 20, 1955], a bungalow that has been swept down stream from one of the summer colonies slams into the bridge. It's the last straw for the beleaguered structure. It loses more than three spans [on the Yardley side], in excess of half its 902-foot length [in 1955]  [The 902-ft length included one span over the Trenton Power Canal now Route 29. The river is 800-ft wide today.]
"Devastation on the Delaware" by Mary A. Shafer, 2005.

YARDLEY WAGON BRIDGE
What appears to have happened is that the backup caused by flood debris buildup at the Yardley Wagon Bridge, combined with other factors to raise the flood level on Pennsylvania Ave.

•  Debris build-up at the Wilburtha Bridge

•  Ground level obstructions as trapped floodwater tries to get to the river by the shortest route

•  The canal "stilling well" effect transmits flood level almost instantly toward Letchworth Ave.

•  Debris build-up immediately ahead of Lock #5 at the natural spillway

•  Force of river is to the right as river bends to left

It should be noted there's a chance that the flood crest behind the church might be somewhat lower than in August 1955 if that same flood occurred today, since the bridge no longer exists.

CANAL BASICS

It is apparent that the canal must have the same average slope as the natural slope of the river bottom toward Trenton. This is achieved by a series of locks inserted between the long flat segments. The water surface elevation between Lock #6 and Lock #5 is set at the same elevation as Silver Creek so as to easily supply water to the canal, and the water level is held more or less constant level by the overflow weirs at Brock Creek. Three variables are involved in studying river flooding in relation to the canal:

A) The varying elevation of the land along the canal
B) The varying height of the towpath above the land
C) The elevation (and downward slope) of the river surface during a flood

As the river flood surface rises its downward slope toward Trenton increases. The 2005 flood dropped 5.6-ft across Yardley from Lock 6 to the Railroad Bridge.

As the flood level rises it eventually intersects the towpath just below Lock #6. As the flood continues to rise the point of intersection of the river gradient gradually shifts further south along the canal.

Some distance above the theoretical point of transition (above Brown Street) the river was higher than the towpath. Some distance below this theoretical point of transition (near Letchworth Ave) the towpath was higher than the river.

The towpath height varies up and down with many low spots 6-inches height or less. In addition to greatly reducing the capacity of the overflow weir this facilitates overflow into the canal above Afton Ave and overflow out of the canal at several critical low spots below College Ave. As a general rule we don't want river floodwaters entering the canal at all, because it overflows the towpath at low points causing a breach doing great damage at locations where the canal sits high above the adjacent ground. Under "Pennsylvania Law" suburban runoff was allowed to increase almost without limit and yet the Commonwealth did not install double floodgates at Brock Creek as had been done at many creek locations when the canal was operational.

TOWPATH HEIGHT ANALYSIS (12-Inch versus 18-Inch)

The towpath between College Avenue and Maple is critically low. This is the part of the problem that has undoubtedly been with us since 1832. Apparently it was only in Yardley where this mistake was never corrected. There is a general river flooding problem all along the canal, due to the initial attempt to feed the canal by gravity. Edgewater (above Afton) Ave is more critical because the flood surface is higher, while S. Edgewater (below Afton) is less critical because the river is sloped downward in that direction. The height of the towpath between Afton Avenue and Maple also relates to the existing house elevations... the question is how do you connect driveways to the raised towpath needs to be addressed. This is where a 1-ft lower river flood can make a tremendous difference when considered in combination with other corrective actions. A 12 or 15-inch towpath may require greater than a 1.5-ft reduction in the river level whereas an 18-in towpath may only require a 1-ft reduction. The two overflow weirs at the aqueduct set at 36.34-ft need to be studied, although only small part of canal inflow from river over broad front above Brown Street.

Pennsylvania Ave & Vicinity Basin
Detailed description of April 4, 2005 Flood Inside Subject Area

Here's what happened...
The floodwater in Belmondo arrived entirely from the "PA Basin" splitting into two separate flows just above the Belmondo line, the left part of it flowing through a cyclone fence into the center of Belmondo, and the right part bypassing around the rear of Belmondo with dry ground separating the two. There are no basements at Belmondo and (as relates to the left part of the flow) the entire area volume of the site functions as part of the retention basin, isolated from and with real solid ground separating this basin from the flooded area along the canal up to the level of stage 25.3-ft at Trenton. Floodwater bypassing the right side of the railroad scoured out the canal towpath with tremendous force due to a 3-ft elevation advantage ---estimated 37.5-ft at Letchworth versus 34.5-ft flood surface elevation under the railroad bridge at the crest. This elevation was maintained for several hundred feet along the swamp where the backed-up floodwater presumably from canal overflow and breaches both further down was perhaps 3-ft lower than the floodlevel in the canal. The apartments are at a higher elevation and well protected by two spillways to the river, one consisting of the entrance/exit road cut through the ridgeline and one over the ground surface along the railroad, although that one was not observed during the flood.

The ridgeline along N. Bell and South River Rd is higher than the river all the way up to Trenton stage 25.3-ft. In fact the entire ridgeline from Ferry Street down past the Macclesfield housing development was dry in April 2005, except for the following overflows from the land to the river: 1) PA Basin overflow at College, 2) Belmondo overflow at Entrance/Exit Road, and 3) overflow to the river along Longshore starting at College tapering down to zero at the upper edge of the Metz apartments, beyond which is where the ridgeline meets the river and bends toward Morrisville. Floodwater inside the front part of Macclesfield entered through the two impoundment drainage pipes. Land to the east of the ridgeline in the triangle that includes "Steaks & Things" is obviously not protected by the ridgeline. Floodwater into the PA Basin arrives from 1) canal overflow + breaches, and 2) overland from the low point on E. Afton. The ratio between items 1 and 2 is unknown. My current guess is that the overland flow toward College may represent 1/3 the total flow into the PA Basin, the remaining 2/3 arriving at high velocity through the canal once it breaches.

Courier Times newspaper provided a full color half page photo of the April 4, 2005 Belmondo flood surface showing 1) brown muddy water over the parking lots, 2) blue water at mid-river, and 3) green grass along the various high ground. There are a lot of possible points of comparison between the photo and the L.M.T. 2-ft contour maps. The edge of the flood on the rear parking lot shows indicates flood level at 36.2-ft with water not reaching the 2-in curb above which the land was dry and sloping up another 6-in to elevation 37.3-ft on the contour map. Flood clearly did not rise to 39-ft elevation around buildings, and bear in mind the "High Stage" flood said to be 39-ft was 0.3-ft below the April 4, 2005 flood at Trenton. The photograph was discussed with photographers at both the Courier Times and the Intelligencer and both indicated said these pictures are typically taken early afternoon. The picture was taken on Monday, April 4, 2005 and appeared in the Tuesday morning Courier Times. It should be possible to find out the exact time.

The printout from the Trenton Delaware River level gauge shows that the peak occurred around 3:00 p.m. and there was very little change for 1 1/2-hrs on each side of the peak. Even if the photo had been take anywhere during daylight hours the variation it would be 25-ft +- 0.3-ft stage at Trenton.

A line of levels was established at the April 4, 2005 flood level, said to be higher than in 2006. The April 2005 flood level was identified by several Belmondo residents and by office personnel. This flood level was carefully transferred to the highest point along the entrance/exit road and by this method the flood level was measured to be 37.2-ft elevation. Taking photo results into account the estimate is rounded off at 37.0-ft.

Flood Level Verification  (Par. Added July 7, 2007):  Photograph of April 2005 Flood Line, found on Internet.

Article #4 being revised here is intended to suggest in broad outline how a detailed analysis of a specific sub-area containing perhaps half the floodplain population in Yardley Borough, and where flooding may not be deep enough to justify elevating the homes, may lead to lower flood levels and to suggest that this can be best achieved by equally focusing on individual sub-areas side-by-side with the overall 3-state area. All the data and ideas expressed in this Article are tentative, and should not be used for design purposes.

The section on Yardley Elevations previously attached to the end of this Article #4 has been revised and used to serve as the beginning of a new Article #5, which may ultimately be augmented to include additional flood level information. The term Yardley Elevations is mainly intended to refer to flood level elevations in feet above sea level, but is also indirectly related to House Elevations.