Voodoo Hydrology 1

An interesting article from Stormwater Magazine. Good fodder for discussion.

http://www.stormh2o.com/sw_0607_voodoo.html

 

[queque]

As a 40 year experienced Hydrologist, this article illustrates the following--
Hydrology is voodoo simply because no one wants to collect the correct information to solve the problem.
Example:
If you want to understand how much water runs off from a small urban watershed, measure the precipitation and the runoff from that rainfall event.
Does that sound simple?
Try this -
place two coffee cans 15 ft apart and measure the rainfall.
You should get the same amount in each can within 0.2 ft.

That seems to provide the total rinfall. The problem is that if you place a series of cans, 15 ft apart, they will all meassure rainfall within 0.2 ft. However, the amounts and storm duration (intensity) will be different.

NOW , WITH THAT INFORMATION, YOU CAN MODEL THE 8,000 SQ FT LOT (WATERSHED).
NOW REPEAT THE PROCESS AND METHODOLOGY FOR A 2,000 ACRE WATERSHED, AND MODEL THAT EVENT.

Then repeat that effort for your area of study, and you will wind up with 40 years experience chasing the unknown data that is needed.
There is no voodoo in Hydrology, only lack of effort to collect what you need to solve the problem.

 

[francesca]

There is no voodoo in Hydrology, only lack of effort to collect what you need to solve the problem.

I guess some people just don't plan 25 years ahead and start collecting rainfall data when they're born so that they can use it when they're 25 and a hydrologist and need to calculate a 100-year storm hydrograph.

 

[DMcGrath]

Collecting the data as queque states is great, until you have to build/change something in your watershed. Then what? I doubt you would find a reviewing agency to accept "We'll design the stormwater collection system after the subdivision is built and we have had time to collect rainfall and runoff data for a year or so"

I also doubt any clients would be willing to pay for such studies when the "accepted method" is already out there. I think the thrust of the article is there are plenty of "acceptable" methods out there, even variable applications of the same method, and they are all technically correct. So which one is "right"?

 

[blueoak]

Thanks for the article DMcGrath. Very well done. A really fun article on Mockus and the SCS method is at:

http://mockus.sdsu.edu

Unlike queque I don't think it is as much a data item as using good judgement. I think that this is best summed up in that the SCS republished their hydrology manual not because of errors or major advances but because it was being used improperly. I think changes were like, "do not do A .." changed to "do NOT!!! do A, and we really mean it". I listened to a stomrwater "guru" attack TR-55. For some reason he thought it should work for storm events that occur 5 to 10 times a year.

One of the best hydrology projects I was part of was where the the main h&h guy actually went out to the field to calibrate the model to actual field marks of a recent 1% storm.

I think the method is tied to what are you doing and what happens if you are wrong. I always run a regression equation even it is a rule of thumb just as a check, and it seems like you can normally find local clues as to what has happened in the past. And then again, as my colleague likes to say "If we had 6 fingers on each hand we would protect for the 144 yr event."

 

[LHA]

Entertaining; a couple of things in response:

1. I've seen many SWM Reports which claim the associated project is proposing SWM facilities for all sorts of profoundly esoteric reasons: protecting things from floods, cleaning runoff, recharging GW, etc. They are all incorrect. The only reason we do SWM design is because the governing Ordinance requires it. If in doubt, answer two questions: 1) has anyone ever paid you to do one where none was required? 2) have you ever done a design for free (besides pro-bono work which was required)? For this reason, my narratives always say "SWM design is per xx Ord.", period...not to protect the children from the ravages of flood waters or bituminous concrete, because I've never done a design for either reason. So, if the reason for doing this work is an Ord., and that Ord. specifies a methodology, the validity of said methodology is rather irrelevant. No, actually it is totally irrelevant.

2. Recently at a Planning Commission meeting, the traffic engineer who was with me was really grilled by one of the members about the broad and general assumptions used in his TIS. Why, it was asked, didn't he count every car, scooter and manure spreader at every intersection within an obnoxiously vast area, and why didn't he consider dozens of possible future (as yet unproposed) land use scenarios and roadway alteration? His response? He uses a reasonable amount of admittedly limited field data, and augments that with historical data and currently approved scenarios. He then applies reasonable and conservative assumptions, and got a result which he KNOWS GOING IN ARE NOT 100% ACCURATE. He further stated that he is confident that his conclusions meet or exceed due diligence and the requirement for mitigation, and reminded the members that he did seal both the study and the plan set containing the proposed mitigations. It was only then that I realized how very similar SWM design and traffic mitigation really is. By the way, we have another meeting for that project Monday night, and I expect conditional approval recommendation for his traffic proposals...my storm is still being beaten to death by a 25+/- yr old overly-eager EIT. No doubt he is dreadfully concerned about the validity of some Ia/P or ARC constant or Distribution S-curve or other such minutiae. After he's had his fun, we will - I am certain, from 10 years of experience - end up with the same pipes, basins and infiltrators in the same places doing the same things with by-and-large the same results I proposed two months ago. Oh, how very similar SWM design and traffic mitigation really is.


Engineering is the practice of the art of science - Steve

 

[RWF7437]

Thanks McGrath for this posting. Excellent article and should be required reading for all engineers, and the professors teaching them. Calibration is the key, I have long believed. Simply following an Ordinance is not engineering.

Thanks again for stimulating this discussion.

 

[env21tech]

Voodoo has apparently prevented drainage engineers from looking up in the sky. Rainclouds are huge - a 5-minimum duration raincloud moving at 12 mph is one mile long!! IDF curves provide the average rainfall intensity for these rainclouds.
Engineers are creating site-specific micro-clouds that are completely out of touch with reality.

 

[LHA]

env21tech:
Once again, if the reason for doing this work is an Ord., and that Ord. specifies a methodology, the validity of said methodology is totally irrelevant.

I need a bldg permit to build a toolshed, much less a Walmart or warehouse or other such structure which will house jobs, domestically-produced product and tax revenue. I need an approved SWM Report to get a bldg permit. So, I will continue to produce SWM Reports based on "...site-specific micro-clouds that are completely out of touch with reality." for as long as is required to get a a bldg permit.

Engineering is the practice of the art of science - Steve