Tuesday, September 8, 2009

Floodgate Reports - The Bottom Line

I've had a couple of posts in the hopper while waiting on some last few pieces of information. I'm going to break into that info early, because I think it's important...

Two major levees-related news items today:

Analysis finds flaws in pumps at outfall canals. (Refers to M.W.I. Pumps)

Corps revises London Ave. Outfall Canal ICS operating procedures. The big news is that the safe water level has now been reduced so low (about 2.5 feet) that the pumps will now run during all major rainfall events.

What does this mean in the big picture?

The current setup was a temporary fix based off what the Corps could assemble as quickly as possible after the storm. (From Page 10 of Parson's Report)

The interim setup, even if every single pump operates flawlessly to its full design capacity, is only designed for a 10-year (UPDATE- "rain") event. We've heard many times how much of a joke the 100-year level is. The 10-year level is even weanier.

Pumping Station #6, which feeds the 17th Street Canal and is the main pumping station for the city of New Orleans, is capable of putting out a little over 10,000 cfs. That's an admitted shortfall of over 20% at that pumping station. In a big enough storm, water will begin to back up. In a short/lightish storm, the only thing that will probably happen is street flooding and maybe a few flooded cars. In a long-lasting/high-intensity storm, large parts of the city , especially Gentilly (drained by PS#3 & #4), Mid-City and Lakeview, will flood.

The parts of the floodwalls that failed were repaired quite thoroughly. The Corps rebuilt those parts with T-walls with ample foundations. But, the parts that didn't fail were left untouched. There's been a about what the safe water level for the Canals is. Each time the Corps has investigated, the walls have been derated. The walls SHOULD be rated to their top (a little over 10' elevation). Instead, they've been derated, especially on the London Avenue Canal, where they'll now close it at 2.5'**. There are currently no plans whatsoever to rebuild the walls or repair them in any way. The Corps says it would take too long and would take too much money and they'd rather get the permanent pump stations online.

If you were to sit the major engineering experts in the area down and ask them what the 3 most important projects are for New Orleans, I'll bet the lion's share would include the permanent pump stations at the mouths of the outfall canals. When the "10-year storm" standard was adopted (~ early 2006), the Corps was planning on having the permanent pump stations in place no later than 2010. Had they been replaced then, the 10-year storm standard may have been reasonable. Now, they're officially saying it will be complete by 2012, at the earliest. The Corps has picked up the pace, but it's looking more like it will be 2013 at the earliest before the stations are complete. The longer it takes for the permanent pump stations to be built, the more maintenance will be required and the more one should question the original design basis for the pumps.
NOLA Pumping Stations
Bottom line: The pumping capacity of the City of New Orleans is seriously constrained. Sections of the city, especially Gentilly, Lakeview, and Treme/7th Ward, WILL flood in storms significantly less intense than the minimal 100-year level we expect, even without a levee breach. The extent of the flooding will be proportional to the storm intensity and duration. The system will not be fixed for several years at the current rate. Major weakpoints in the system (the weakened walls) will still not be fixed by the time the system is complete.

I hope that sums things up without droning into too much detail.

________________________
* I'll leave the "contractual" issues for others to comment on, I'll stick with my expertise. From what I understand, the Corps basically called around to see who could supply the most cubic feet per second (cfs) by next hurricane season. That particular time, because of rising commodity prices, was a bad time to be buying heavy equipment, especially pumps, especially pumps that large. I'll give the Corps the benefit of the doubt and say that, among those they contacted, MWI had the most cfs with the quickest stated delivery.

** Safe water level used to be about 5' with closure at 4'. Now it's 4' with closure at 2.5'. The Corps has said they will take whatever measures necessary to stop the level from rising in that Canal, including shutting off all pumps feeding the canal until their pumps can catch up. It's already happened several times. The canal walls have been derated so much, I'd curious how they justified the decision to give them any rating whatsoever.

PS- Thanks again, Matt McBride. Fix the Pumps is a great reference site.

NOTE: Some minor edits after initial post.

UPDATE- See comments.

UPDATE 2- London Avenue floodgate closes today (9/12) due to heavy rainfall.

9 comments:

Tim said...

Clay,

I'm glad to see you moved this to your blog--discussion of such complex issues on Twitter was difficult. And I know you’re trying to keep it simple and short, but that is going to be very difficult to do given the complexity of the topic. In fact, I have to split this into multiple comments for it all to fit!

Your post brings up some significant issues that I will try to address, but you must keep in mind that what follows is just my personal opinion based on what I know about these projects. I speak only for myself here in the interest of sharing my understanding with a fellow engineer and the public who read your blog. I think you have an advanced appreciation for the design and operational issues on the outfall canals, but there are some important details you leave out and/or confuse.

Design Event: The 10-year, 24-hour rainfall event is the standard used throughout the area for rainfall runoff. The Corps did not select this standard; it matches what the NO S&WB, Jefferson Drainage Department and others use. If you are unhappy with this "weanie" standard, you should direct your complaints to the NO S&WB and other local drainage authorities.

You should not confuse this with the 100-year base flood elevation established by FEMA for the National Flood Insurance Program. Even if the city's pump system does not remove all rainfall runoff as it occurs, there is considerable storage capacity in the pipes, the streets, the parks, etc. If the city's pumps fail, for instance, or if we get a 20-year rainfall, homes and cars and businesses will not automatically be flooded just because the system is designed for a 10-year event. The 100-year base flood elevation takes into account the pumping capacity of the city and the storage capacity. The BFE changes throughout the city, not because of different pumping capacity so much as because of ground elevation.

--more--

Tim said...

--Part Two--

Pump Station Capacity: Some clarification of the numbers tossed about for the pumping prowess of PS#6, the most impressive structure of the S&WB's fleet of pump stations. If you walk through PS#6 and add up the capacities on the nameplates of all the pumps, you get 9,480 cfs. Add that to the two other pump stations that pump into 17th Street Canal (I-10 underpass = 860 and Canal Street Metairie = 160), and the total is 10,500 cfs. So that's how you get the often repeated capacity of a "little over 10,000 cfs."

This may be the total rated capacity, but does it really represent how much can or will be pumped during a storm?

First of all, 10,500 cfs is of course the rated capacity under optimal conditions. Optimal conditions are with water in the canal at El. 1. As an engineer you understand that as the water surface rises on the discharge side the pump efficiency decreases. In practice this means that if Lake Pontchartrain is at El. 1 the combined total pump capacity is 10,500 cfs. During hurricane events prior to construction of the Interim Closure Structures, the water elevation in the canals was much higher and actual pumping was reduced. I have not seen anything from NO S&WB addressing this issue, but I have seen one estimate that the actual pumping capacity is about 8,100 cfs when the lake is at El. 5.

(You may ask, why would the S&WB design their pump stations for such a low water elevation? Because they were intended to pump regularly recurring rainfall--service during hurricanes is an additional duty for which they were not expressly designed. This is also why there were no "safe houses" for operators and why many pumps rely upon the local electrical grid and few pumps have back-up power.)

The second thing to consider when determining the actual pumping capacity during a storm is the stormwater delivery system. A pump station can only pump the water present at the pump station intake. PS#6 has a gigantic intake basin to accommodate its impressive array of pumps, but it is limited by the pipes and canals that deliver water to it. The primary feeders are the Palmetto and the Monticello Canals. Again, the S&WB has not published information on the capacity of the city’s canals to deliver stormwater to PS#6, but indications are that it falls considerably short of 10,500 cfs. Plain and simple, you can't pump water that isn't there and there’s no reason to build a pump station to pump water above and beyond what the city’s system can deliver.

--more--

Tim said...

--Part Three--

Pump Shutdowns: You write, “The Corps has said they will take whatever measures necessary to stop the level from rising in that Canal, including shutting off all pumps feeding the canal until their pumps can catch up. It's already happened several times.”

I think you are completely wrong on this one. Pumps at the Interim Closure Structures are only turned on when the gates are closed. The gates have only been closed for tests and for Hurricanes Gustav and Ike. As I informed you via Twitter, during both hurricane events pumps on all three outfall canals performed the mission. There was NO problem with water rising too high in the canals and the S&WB pumps were NOT shut down.

I think you are confusing this with a few events when S&WB pumps have been shut down to keep water from rising above the safe water elevation when the gates were wide open and the pumps at the lake were not running. Again, the only time the pumps at the lake are operated is when the gates are closed. During a rain event, water in the canals starts out at whatever the lake water elevation is, and then rises depending on how much is pumped into the canals. Because the safe operating level of the London Avenue Canal is so low, pumping had to be curtailed on at least two occasions I am aware of (the T-P says it’s happened three times). I was still living in a FEMA travel trailer near London Canal during one of those events and I watched the street turn into a pond. But this has nothing to do with the adequacy of pumps at the lake because this happened when the gates were open.

This whole post and discussion was triggered by the press release announcing a new operational plan for London Avenue Canal. My understanding of the plan, which calls for more frequent closure of gates at the Interim Closure Structure and use of the pumps more often, including during non-tropical events, is that this will IMPROVE drainage. The problem has been that water in the canal rises too high because of the lake level and the pumping of stormwater. The simple solution then is to close the gates and pump the canal lower than the lake. This keeps the water surface in the canal at safe levels while the S&WB pumps run unrestricted.

The new plan does not appear to have been adopted to prevent wall failures; the operational plan already in place assured that goal. The new plan appears to be designed to allow MORE PUMPING from the S&WB. Based on that I think your conclusion that we’re more susceptible to flooding is incorrect. I agree with you that there is still considerable risk of flooding from rainfall just as occurred May 3, 1978 and May 8, 1995, long before installation of the Interim Closure Structures, and that we all need to be thinking of ways to reduce that risk.

The question you did not ask in your post was, “What does the NO S&WB think about this new operational plan?” Surely they were included in the discussion and decision to adopt a new plan. Surely the city’s engineers have an opinion on this change. Do they think it helps or hurts? It is curious that we haven’t heard from them on this. Oddly, there is no mention of the people who own and operate the city pump stations in the article by The Times-Picayune.

http://www.nola.com/news/index.ssf/2009/09/london_avenue_canal.html

Okay, that’s all I can write tonight. Hope this helps illuminate some of the issues. We’ll talk some more later.

Peace,

Tim

Clay said...

I enjoy the constructive back and forth. Also, I noticed the date on the comments. Hope I'm not keeping you up to late with this stuff.

Blogs are a good way to hash this all out. I write about this because it's what I know: pumps and piping. You can be very, very detailed, something twitter or a newspaper can't. I think the Dead Tree edition carried quotes from the S&WB that were cryptic, but sounded skeptical of the plan.

The issues on the nameplate capacities applies to both the ICS pumps and the S&WB pumps. To give just one example, the S&WB pumps can operate at their capacity for indefinite periods. The hydraulic pumps have known overheating problems and can't run at full capacity for extended periods.

You're correct the closure structures have only been closed during storms, but that still doesn't mean there's significant impediment to the S&WB pumps. For example, the 17th Street Canal has several restrictions put in place during the Corps repair work. There's a shelf at the bottom, a narrowed channel at the Bellaire Drive repair, and finally the closure structure itself. Restricted channels won't carry the same flow as open channels. The S&WB was already concerned the channel wasn't wide enough before Katrina (hence that Boh Brothers dredging barge).

Ike & Gustav were fairly dry storms, at least for New Orleans (not for other areas). I'd be curious how they'd handle a May 8th flood (but honestly, that's one curiosity that's better in theory than in practice, if you know what I mean). Keep in mind that the SELA Project greatly boosted the capacities feeding PS#6. I also thought that PS#6 is actually closer to 11,000 cfs, but I can't find documentation on that.

The closure method might turn ot to be a good workaround for the restrictions. There are still some issues with the MWI pumps, how to clear the "notch" for the gate (are they still using divers?), etc. but you're right, it could help pumping capacity.

All that being said, the walls should be strengthened or replaced along their entire length and the permanent pump stations project should be built ASAP. That's the real bottom line.

That's all for now.

Clay said...

Thinking over it more, the plan to turn the London Ave. Canal into a giant tank and drain it using the floodgate pumps as a workaround for restricted pumping capacity shows some promise. It didn't fully occur to me that that was the goal. If they're derating the canal even further, I guess they have to do it.

Turning a storm event into actual flowrate numbers to hand to your pump guy is not an easy thing. It takes calculating the drainage basin (in this case, basically all of Orleans west of the Industrial Canal), factoring in seepage, storage capacity, etc. New Orleans always gets screwed in those calculations because basically every single drop has to be pumped up and over the levees.

I still think that the pumping capacity is seriously constrained and that the Corps decision to leave the floodwalls intact after Katrina (indeed, there are still no plans to replace/repair those walls) is the bigger issue.

Also, the pumped capacities quoted by the Corps for the floodgate pumps: are those values based off flowmeter readings or are they just estimating based off pump runtimes?

Tim said...

Clay,

Every engineer in the world agrees that the pre-2005 floodwalls on all three outfall canals are inadequately designed. They do not pass the current required load combinations and factors of safety. That's why the safe operating level is many feet below the tops of those walls. The problem in resolving this issue is not the engineering. The problem is the cost. The official decision is that the pump stations and the canals can be operated safely without spending a lot of money to fix walls that do not pose a danger.

The largest complaint I hear is typically this one: if the new pumps at the lake fail, the canals will either overfill, the walls will fall and the city will flood, or, the city will stop the interior pumps to avoid overloading the walls and the city will flood. Bad scenarios both, but what are the other scenarios? If the walls are removed and possibly replaced, they would probably not be to the same height they are now; they would be constructed as high as required to safely operate the system. So if the new pumps at the lake fail, the canals fill with water and then overtop and the city floods, or, the canals fill with water and the interior pumping capacity is greatly diminished and the city floods. Not much difference in risk.

Another option being called for is the removal of PS#6 and construction of a canal at or nearer ground level. Under this scenario, if the new pumps at the lake fail, the canal fills with water, overflows and the city floods.

Another problem with repairing/replacing all the walls on all three outfall canals is constructibility. Any work undertaken would have to be accomplished in such a way that it does not reduce the drainage capacity of the city. That probably will require a by-pass of the canals and a lot of property will have to be taken. For most of the canals, houses would have to be demolished or at best homeowners would lose use of their back yards for 2 or 3 years. On Orleans Canal, this could be accomplished by routing through City Park, but I think we know how popular that would be.

So I understand how the decision to leave the walls, however defective, in place is the best decision. It's not a perfect solution but it's the best given the costs and constraints of the problem.

Peace,

Tim

Clay said...

No way in hell leaving those walls is the right thing to do. At an absolute minimum, the Corps could drive 50' sheet piles parallel to the walls on the unflooded side. This can be done with barges to minimize right-of-way issues (although, they'll still be some). Fill the gap between the top of the sheet pile (left at roughly the same elevation as the top of the floodwall) and you've got yourself a pretty strong patch. I know about that option, because it was offered to the Corps in the days after Katrina and they rejected it.

Leaving the walls in place is not the right call, even if the Corps is struggling to stay on budget.

Unknown said...
This comment has been removed by the author.
Unknown said...

We can never predict the God's plans and never will be prepared well enough to confront them. I thing floods are probably the worst natural disaster that leaves big consequences. That's why I think it is better to buy flood insurance policy