Freedom of Information Act yet to come down with the rain

After spending money installing numerous sensors in canals and large drains during the last few years, we’ve now discovered that they were “overwhelmed” by last Sunday’s (5 June 2011) storm.  The sensors were designed to monitor levels every two minutes in wet weather. When the water level in a drainage system hits 75 percent of capacity, the sensors are supposed to send an alert to subscriber buildings. They do this again when the level hits 90 percent of capacity.

Last Sunday however, the water rose so fast, the system failed. Alas, it wasn’t clear from the Straits Times why exactly it failed, though we may be able to make an educated guess.

Try to parse this report for yourself:

In a matter of minutes during Sunday’s heavy storm, Stamford Canal at Orchard Road was overwhelmed.

So was the PUB’s water level sensor software, which sends out SMS flood alerts to building owners in flood-prone areas.

The software was programmed to send out alerts when water levels in canals hit 75 per cent, indicating a ‘moderate’ risk of flooding; then at 90 to 99 per cent, indicating a ‘high’ risk. It was not programmed to send out text alerts if the water level went beyond 100 per cent.

However, at the Stamford Canal in front of Forum The Shopping Mall that day, water levels surged from 50 per cent to overflowing – or above 100 per cent – in two minutes or less.

The result? The building managements of Tanglin Mall and St Regis Residences and Hotel did not receive any alerts. Both the mall and St Regis Residences suffered serious flooding, although the hotel was spared that.

[snip]

However, on Sunday, water levels at that segment of Stamford Canal rose ‘three or four times faster than anything we have seen before’, said Mr Peng Kah Poh, director of PUB’s infocommunnications department.

‘It was so high that it bypassed the trigger points of 75 per cent and 90 per cent, so the SMS alert was not sent,’ he said.

[snip]

It will also look at shortening the current two-minute time interval for monitoring water levels during wet weather.

— Straits Times, 8 June 2011, Alert system overwhelmed by storm, by Grace Chua

A graphic accompanying the newspaper’s article revealed that at the 50-percent mark, the system notifies the Public Utilities Board (PUB). Then when it crosses 75 and 90 percent, alerts are sent to buildings that subscribe to the alert system. The report above indicates that after crossing 50 percent, the water rose to overflowing within two minutes; in other words, by the subsequent sampling, it was over 100 percent already. Perhaps the system then failed to send out an alert because alerts are only triggered when water levels are in prescribed ranges (75 – 89% for the first public alert and 90 – 99% for the next one), not triggered by measurements that return a True finding when tested for >75% or >90% levels. If you understand a little about software programming you will immediately see the difference and slap your head over the folly of using ranges instead of threshold tests!

I wished the reporter had the presence of mind to pin the official down on this.

* * * * *

Anyone who has ever struggled with software glitches in our government systems will tell you this is bloody typical of the Singapore bureaucracy that keeps on boasting how wired they are but actually is bedevilled by amateurish incompetence.

For example, I’m told by a friend that filing annual returns for a registered society is a pain. The system requires attachments (e.g. financial statements) to be scanned and linked to the filing, except that it imposes a maximum file size that effectively limits it to only two pages of attachments. Anybody who has handled audited financial statements will tell you that if they come in only two pages, they are very dodgy. So who in the Accounting and Corporate Regulatory Authority came up with software that only wants dodgy statements?

In 2007 I wrote about trying to lodge an application for a Professional Visit Pass for a foreign academic, Douglas Sanders. The online form that I had to fill allowed me a maximum of 25 characters (characters, not words) to declare ALL his previous employers and their addresses. In another box, where I had to declare ALL his previous residential addresses, I was also allowed the grand luxury of 25 characters.

Am I surprised that storm surge software is not able to handle a storm surge? No.

* * * * *

Was it such an impossible event last Saturday that the water rose ‘three or four times faster than anything we have seen before’, as Peng Kah Poh said (quoted above)?

A total of 124mm of rainfall was recorded for the central area, about 77 per cent of the average monthly rainfall for June.

About 65mm was recorded within 30 minutes yesterday compared with 100mm within two hours on June 16 last year, which had led to serious flooding of Orchard Road.

— Straits Times, 6 June 2011, Govt to review drainage after year’s worst flood, by Grace Chua

Already, those numbers suggest something troubling. Why is it that the data for the 16 June 2010 flood is based on a two-hour reading (100 mm), and that for 5 June 2011 based on a half-hour reading? Why not compare the 65 mm of 2011 with the worst half-hour reading from June 2010? Apples with apples? And if the PUB didn’t volunteer the information, why didn’t the reporter ask?

Is it possible that in fact we do not have data pertaining to half-hourly precipitation until recently? Because if we did, it would be the easiest thing in the world to compare like with like. And if we do not have historical data pertaining to half-hour spans, then how would the PUB know that the flow that was experienced was three or four times faster than anything we’ve seen so far?

It’s like this: Unless we have historical data about surges, we really can’t plan for them.

Consistent with my guess that we have not been collecting such data, I cannot easily find references to precipitation peaks in previous years on a half-hour or even one-hour basis. I see reports like this one,  mentioning rainfall on a 24-hour basis:

To date, the highest rainfall recorded in Singapore over 24 hours was 512 mm in 1978, the second highest was 467 mm in 1969, while the third highest amount of rainfall to drench the island was on 20 December 2006, when 366mm of rain fell over a 24-hour period. Flash floods also occurred in various parts of the island republic in 2007 and 2009, raising the disturbing possibility that such floods are becoming a regular feature.

— Asia Sentinel, 17 June 2010, Hip Deep on Singapore’s Orchard Road. Link.

Every search I tried about the 2 December 1978 flood returned the same figure of 512 mm in a 24-hour period, such as this sunny story in The New Paper (click the thumbnail) titled Why Orchard Road will never flood. No other figure regarding peak rainfall for the worst one-hour or two-hour period could be found. I find it hard to believe that the data is not available (at least internally within the Meteorological Department) but hey, stranger things have happened.

So, might it be possible that planners simply used the figure of 512 mm over 24 hours for planning purposes? Dividing it by 24 hours, they would come up with 21.3 mm per hour, then make some adjustment on the fly and say to themselves: Let’s plan for 50 mm per hour. Only to be caught out when rain came down at a rate of 65 mm in 30 minutes.

That’s the danger with relying on averages when trying to make projections for surges. There is no short cut. One has to go out and collect data about surges themselves.

If this was what happened, it would not the first time. Remember the huge controversy a few years ago about crunches on trains, with the management of SMRT Corp issuing press release after press release giving average crowd densities, in effect denying that things were as bad as commuters reported them to be? I wondered then, how did they measure what they measured? When did they measure what they measured?

If only we had a Freedom of Information Act, I would be asking for historical data of previous rainfall peaks. I would also be asking what assumptions the PUB used in planning for drainage, not just in terms of precipitation, but also in terms of run-off from built-up areas. Then hydrological academics in our universities can crunch those numbers and tell us whether they are realistic.

If only. . . .

How do we monitor the quality of our government unless citizens have access to information?

50 Responses to “Freedom of Information Act yet to come down with the rain”


  1. 1 deathmule 9 June 2011 at 00:16

    PUB’s code of practice for surface water drainage can be downloaded from here: http://www.pub.gov.sg/general/Documents/HeadCOPFINALSep2006-1.doc

    • 2 yawningbread 9 June 2011 at 00:55

      Ah! Thank you. Most interesting.

      For readers who don’t get quite as excited as I do with technical stuff, let me describe what it is. It’s a set of guidelines for architects and engineers to help them calculate an appropriate drainage capacity for their projects. Page 22 is relevant. However, Page 23 says that
      “important installations” (which I would assume includes major canals and drainage tunnels) should have a designed capacity to cope with rain conditions expected to occur only once in 50 years. But how would one know what conditions need be expected only once in 50 years? It then refers professionals to a chart on page 37 (Appendix 2) which I have embedded into this comment so readers can scrutinise.

      Click the graph to enlarge it. The red and blue remarks (and dots) were added by me. If my interpretation of this graph is correct, the event that occurred on 16 June 2010 (100 mm over two hours) was a once-in-six-years event. That which occurred last Sunday, 5 June 2011 (65 mm over 30 minutes) was also a once-in-six-years event. In other words, going by PUB’s own charts, the two events had a similar likelihood of occurring, which is completely opposite to the impression (that last Sunday’s was unprecedented) that the PUB spokesman was trying to give. And both were far from being once-in-50-years events for which an “important installation” should be designed for.

      The Code of Practice was dated March 2000. The graph could be much older though exactly when it was drawn up cannot be ascertained from the face of it. Nor can we see the dataset behind the graph.

      • 3 Yixin 9 June 2011 at 10:57

        A few minor quibbles regarding your interpretation:

        1. In the Code of Practice, the stated examples of major installations (to be built to 50 year time periods) are “airports and tunnels”, while “outlet drains and secondary drainage facilities” are to be built to 5 year return periods. So I wouldn’t be as quick to assume that Stamford Canal falls in the former category.

        2. The vertical (rainfall intensity) axis is logarithmic, not linear. While it is apparent that the reported rainfall intensity is somewhere between the 5- and 10-year time periods, I think stating a 6-year time period is being unnecessarily precise.

        More broadly, return periods/frequency intervals (like any probability measure) are heavily dependent on the underlying datasets and modelling assumptions. From what I can make out (equally spaced frequency curves on a log axis), the underlying assumption is a binomial distribution, which is probably not conservative enough when dealing with extreme (or tail) events. I could be wrong (i.e. it could be log-normal with a low scale) – it’s tough to tell from the graph.

        But to your primary point, it would be very useful (and fascinating) to have access to the data – to see how many years of data were used, which weather stations, how the data was aggregated, and to work out the modelling assumptions.

      • 4 Resha 9 June 2011 at 12:26

        Your reading of the graph is sound. If indeed this is the graph that PUB uses, they have a lot of explaining to do.

        Rainfall intensity aside, the catchment area and downstream discharge conditions are also important factors to consider in determining the capacity of the drainage system. The original design capacity may no longer suffice if the catchment area increases substantially. If they have designed for a 50yr event, then they may end up with only a capacity for say a 10yr event.

        But unless the flood can be attributed to reclaimed land, it is unlikely that the problem lies with the catchment area. Large canals are the final conduits in the system and it makes sense that they have already been designed with a capacity for most if not all of singapore’s catchment, regardless of whether an area is built-up or not. Afterall, Singapore has done very extensive master-planning very early.

        The key constraint at the discharge point would be the sea level. The higher the sea level, the wider the drain has to be for a given capacity. Theoretically, being able to control the water level at the discharge point is a tremendous advantage in terms of stormwater management, especially if we could keep that level low during high tide and discharge surges.

        But our Barrage has another priority. It needs turn our still blackish water into fresh water. To do that, it has to maintain the water level above the sea level so that the blackish water can be discharged and replaced with the stormwater from the drainage system. Effectively, the water level behind the Barrage will have to be raised most of the time, for as long as is needed to desalinate our rivers, above the sea level at low tide.

      • 5 Ben 12 June 2011 at 21:28

        It is possible to check the discharge capacity Qc of the Stamford Canal using Manning’s Formula in 7.2.2 against the peak discharge Qr using the Rational Formula in 7.1.1 if you can get the dimensions of the canal at the section where it got flooded, the topographical map to define the catchment area and the time of concentration to derive the rainfall intensity based on a return period of 50 yrs using the graph.

        Time of concentration is the sum of overland flow time and channel flow time. There are different methods to calculate the overland flow time. One university in Singapore uses t = 3.26(1.1 – C)L^0.5/S^(1/3)

        where,
        t = overland flow time (min)
        C = runoff coefficient
        L = overland travel distance (m)
        S = overland slope (%)

        The catchment area for Stamford Canal can be very big, consisting of densely built up areas, less densely built up areas, urban areas and fully built up areas. I do not know what runoff coefficient was used when Stamford Canal was built and whether PUB has forseen the intensity of development in the catchment area.

        The overland flow time could be shorten by a lot when runoff coefficient increases. This can shorten the time of concentration used for deriving the rainfall intensity, which in turn pushes the peak discharge Qr to a level beyond Qc.

    • 6 Tommy 9 June 2011 at 01:05

      They should just burn the code since it does not work.

      • 7 CafeSolvere 9 June 2011 at 23:28

        Please read bettersingapore.blogspot.com

        2011 05 05 with clear warning email sent to PUB and RSP (i saw RSP sign on the raising road work) on the raised Orchard Road will be flooding worse, because the raised do not increase the flow to the bay.

        2011 06 05 is the flood as forewarned.

        The code specifically disallowed drains to be covered, so that we can inspect, we can clear.

        Today, every officers, every construction and maintenance contractors is happy to have drains covered, just like data under-covered.

        The heaviest down pour was on Jan 18, Thursday 2007 1530 hours.
        Even Telok Blangah Hill Top HDB also flooded with water coming down stairs – the Stomp video said something Mount Faber also flood, is no longer available.

        There was a SMS warning at 1500hrs to PUB when sky was clearly blue.
        Within 10 minutes sky turn dark and rains punched the ceiling like machine guns.
        This is the first event, Singapore netizen actively sharing videos, photos.
        This rapid rain was never mention by PUB after that.

        Read PUB Yap Kheng Guan two page Straits times report, this was not mention, and previous 2007 2008 orchard road flood also not mentioned.

        Fellow citizens, go hunt down these report, blogs, stomp, utubes.
        I told Vivian i am coming to see him to help him out.

        There is some more important informations PUB yet to release.
        Singaporeans, please think, is the water level just to sms shops owner, or activate the pumps somewhere down streams ?

  2. 8 ~autolycus 9 June 2011 at 00:22

    What gets to me is incomplete information in the replies provided. Here are three examples:

    1) Minister is asked about 1st batch Integrated Programme results (http://www.moe.gov.sg/media/press/2002/pr31122002.htm). Note that he gives results for three schools and omits the fourth (http://www.moe.gov.sg/media/press/2002/pr31122002.htm).

    2) Minister is asked about whether results are satisfactory (http://www.moe.gov.sg/media/parliamentary-replies/2010/01/chinese-language-psle-results.php). Note how he avoids making that judgement.

    3) Minister is asked about sports in schools (http://www.moe.gov.sg/media/parliamentary-replies/2010/11/sustained-achievement-award.php). I couldn’t find the answers. Actually, it made me read all the Parliamentary Replies back to 2004. Amazing experience.

  3. 9 Tommy 9 June 2011 at 01:02

    Pointless. It is like having earthquake sensor that only gives you a 5 sec advance warning. The drainage systems have no buffer capacity which is why it can go from 50% to 100% in 2 minutes. That in itself shows that the drainage was overwhelmed. Sensors were pointless even if they worked.

    PUB started talking about sensors to show that they are doing something. But the big billion $ problem is that the drainage system cannot cope with all the recent property developments. Instead of water being absorbed by open fields, water is now channelled directly into the drains.

    • 10 yuen 9 June 2011 at 06:06

      by selling old carparks and grasslands in the downtown area to build expensive condos and shopping centres, they brought in big sums of money, but then could not keep the buildings from getting flooded in heavy rain

      http://singazen.com/complaints/more_floods

      Now they will need to spend heavily to add canals and reservoirs in the built up areas to cope with this; it is rather like Goliath being brought down by a stone… nature taking revenge on humans who become too clever…

  4. 11 YH@2 9 June 2011 at 03:03

    Hi Alex, your article hits home on so many areas that irritate the hell out of me as an engineer and a citizen.

    Software ‘glitch’ – as you pointed out in your example of a simple programming code improvement- was not a glitch but just poor project development,planning and management. I’m sure many others who know a little about programming had *facepalm* moments after reading about the “glitch”.

    a) PUB obviously did not test or design a robust test to dry-run their system which was key to preventing losses amounting to millions of dollars. Either way – Epic Fail!

    b.i) PUB engineers or whichever company they outsourced this software programming to obviously did not bother to properly think of various possible scenarios.
    b.ii) The e-filing/e-request forms you mention are another case in point of this lack of scenario planning which IMO is indicative of poor critical and creative thinking.

    I’m reminded of your points made in Kaya Toast Productivity 1&2 posts.
    Love your phrase “bedevilled by amateurish incompetence” – Indeed!

    I’m with you about the lack of data available to public scrutiny, the wishy-washy methodology and liberal (ab)use of averages!!!

    p/s the “chart on page 37 (Appendix 2) ” looks like the charts I see in technical textbooks printed in the 1960s.

  5. 12 ahhock24 9 June 2011 at 04:15

    The truth is, which PUB will never admit, marina barrage is the root cause of all flooding downtown. As we know marina barrage now serve as a reservoir retaining water from catchment nearby.

    Problem is with the retained water, marina barrage has created inland a ground condition which is always in a state of artificially created high tide. We know that when storm coincide with high tide, probability of flooding is high. Opening the gate, no matter how fast, to lower the water level by discharging water back to the sea when storm is looming does not help.

    Why? The groundwater, at most time, at high level don’t drop as fast. In fact, soil behaves like sponge soaking water even higher than the last water retention level. As a result, our soil has lost all its water-absorbing ability, all rainfall will now end up in the drain instead.

    The PUB is now planning to pass the problem upstream by revising building code requiring provision of retention pond for all building owner.

    • 13 yawningbread 9 June 2011 at 11:49

      I’ve wondered about that myself. The barrage’s purpose is to keep water levels as high as possible — that’s what a reservoir is supposed to do. This means the water table in our soil is now maintained at a high level too, so the water absorbability of our soil during rain is now reduced. Unfortunately, none of us has data, and the government by its insistent denial that the barrage contributes to the problem is not about to go collect data themselves.

      • 14 prettyplace 13 June 2011 at 13:49

        Yap, along the same line of thought. Further the built up area has increased over the years in the flodding areas with far less land to absorb rain water.
        There was another interesting point by Resha, with regards to re-claimed land, especially around our water fronts.

    • 15 ahhock24 15 June 2011 at 16:38

      Development built-up is progressive, while damming of Marine Channel is disruptive. It does not help that all flood cases start to surface when Marine Barrage is completed.

      At tip of iceberg, the Botanical Garden and Fort Canning (equivalent to 2000 times an olympic size pool) are pouring water into Orchard drain instead of adsorbing it. We can only imagine much worst scenario if all pocket of green area are as impervious as well-paved ones.

      As our land is no longer capable to serve us as retention pond, the natural progression for us is to start creating man-made ones.

  6. 16 Once Every 50 Years - Out of Touch Elites 9 June 2011 at 05:28

    Alex
    Again, a great article.

    If you, a lesser mortal (by THEIR standards) can think and ask these critical questions.
    I don’t see why our scholar-elites cannot likewise apply themselves.

    My guess is outsourcing.

    The scholar-elites think that out-sourcing the job is the same as actually solving the problem.

    We need to sack a few scholar-elites to remind them that they cannot outsource the responsibility and accountability.

    I have personal knowledge of how IT projects are outsourced in a local bank.
    The IT vendor is manned by inexperienced personnel from overseas.
    The bank’s employees managing the vendor are inexperienced local fresh graduates.
    The experienced staff (over 40 years old) have long been quietly forced out through unfair treatment.
    Amateur hour prevails.

  7. 17 DetachedObserver 9 June 2011 at 07:20

    As a programmer who earns quite a decent living via requirements analysis, project management and actually writing code, I can tell you why the water level sensor failed. With the given description of how it worked, it was probably never designed to detect a sudden surge.

    An actual flood detection system would be real-time and probably cost much more. Probably anathema to penny wise pound foolish management drones of the PUB.

  8. 18 Lee Chee Wai 9 June 2011 at 07:25

    http://app2.nea.gov.sg/localclimate_climateinfo.aspx

    I’m tempted to buy this (the relevant hourly information probably costs between $100 to $200), but it seems fairly useless with respect to the problem of flooding if we do not also have some rudimentary record of quantified/quantifiable flood incidents over the same time period. So, I’m pretty mixed about this.

    I am interested to figure out (through correlation) if the data will agree with the hypothesis that the Barrage is responsible for the recent spate of floods. Further evidence, if available, can come in the form of quantifying any permanent reduction in drainage capacity and then determining if that reduction coincides with the construction of the Barrage.

  9. 19 Wa Lau Tan 9 June 2011 at 07:32

    Both singtel and ICA also have size limit on attachment for documents and photograph online. The programmer ought to be fired.

  10. 20 Lee Chee Wai 9 June 2011 at 07:37

    Sigh … never mind … rainfall data is $27.50 per station per month. There are 51 stations and data from Jan 1990 to Apr 2011 (52 months). Getting the whole hourly data set would set me back $72,930 …

    I sometimes really wonder where our tax money really end up … I thought these activities were conducted in the service of our nation through our taxes. Sure, I’d cough up a small administrative fee (a few hundred, maybe) for information, but $73k?

  11. 21 budak 9 June 2011 at 09:19

    Australia’s meteorologist answer the question: Why do 100 year events happen so often? http://reg.bom.gov.au/water/designRainfalls/rainfallEvents/why100years.shtml

  12. 22 Slava 9 June 2011 at 10:51

    I remember that it is possible to pay a fee to buy a copy of the weather figures (such as temperature and other things) from what was then the Meteorological Service in Singapore. But I’m not entirely sure if they will have the necessary data such as rate of rainfall, which is ultimately more important than the absolute water levels itself as seen in this most recent episode.

    And on a completely facetious note I’d imagine PUB’s code to be:

    %PUB’s flood warning code:

    while (yes_rain == TRUE)
    check_meter( );
    if (water_level == 75%) %yippee, we is genii! we put in conditionals!
    send_SMS (“Flood alert!”);
    elseif (water_level == 90%)
    send_SMS (“Flood alert!”);
    else
    do_nothing ( );
    end
    end

    %Got ppl say genius plural is geniuses, but plural of fungus is fungi what.%
    %What about rate of increase har? Surely that one more serious than absolute water level?%
    %From Team Leader to above remarker: hullo, the SOP says 75% and 90% so just follow, ok?%

    • 23 reservist_cpl 10 June 2011 at 01:00

      You forgot (in the loop),

      Sleep(120000); //assuming milliseconds

      Probably the staff also slept along with the machine. The staff in the loop.

  13. 24 yawningbread 9 June 2011 at 11:42

    Yixin (9 June 10:57h) – Page 23 of the Code of Practice says major rivers should be designed to handle weather events that occur once in 50 – 100 years. Surely canals draining a large catchment area as Stamford Canal does is equivalent to a major river?

    The “outlet drains” that the Code speaks of are literally that: the drains that architects and engineers have to design for the property (condos or shopping centres) they are building. The amount of water these handle is quite a different order of magnitude from what the Stamford Canal has to handle.

    • 25 Yixin 9 June 2011 at 12:08

      I’ve e-mailed PUB to ask for clarification, better than speculating.

    • 26 Yixin 1 July 2011 at 21:01

      Alex – I have also forwarded the reply from PUB. The explanatory portion of the email reads:

      “The Stamford Canal is not classified as a major river as its catchment is relatively small. It also existed more than two decades before revisions to size major rivers to cater for higher intensity storms were made to the Code of Practice on Surface Water Drainage in 1991. In view of the densely built up nature of the Stamford Canal catchment, options to increase its capacity have to be considered within the constraints of limited land and cost effectiveness.”

      I would rather not comment on areas where I’m not an expert (i.e. classification of canals), but I do want to add that my experience with government agencies is that they will respond to requests for information, albeit not always in haste or in full.

  14. 27 InMyOpinion 9 June 2011 at 12:01

    It is always easy to write about “If only I know this…” and “If only I know that…”. I wonder what are the assumptions and basis you will use when you are in that position to effectively monitor the drainage level and at the same time to minimize the number of false positives that will create strain on available resources. There is always this balance we have to maintain in any implementation of a monitoring system, isn’t it?

    I guess what we can fault them on is really how frequent they review these parameters and thresholds or even ranges used.

  15. 28 We Don't Care - Just Fix It 9 June 2011 at 13:09

    Bottom-line.
    We are paying these scholar-elites world-beating salaries to solve our problems.

    So that ordinary citizens like Alex Au don’t have to spend time thinking about cause-solutions.

    My message to Video Balakrishnan:
    “I don’t care. Just fix the problem. Or we vote you out.”

  16. 29 Anonymous 9 June 2011 at 14:01

    SPH is a biased media company. Hence their publications like Strait Times are subject to numerous filters.

  17. 30 Dismal Soyanz 9 June 2011 at 16:18

    @SPH

    Napolean is attributed with the following: “Never ascribe to malice, that which can be explained by incompetence.”

  18. 31 nitegazer 9 June 2011 at 16:26

    Your article leads me to wonder how these flood sensors are implemented in hardware. Are there physically two sensors at the 75% and 90% mark in the drains/canals, or is the water level somehow being inferred by another metric, such as rate of flow in a part of the channel that has a sensor?

    Apart from the above, from an engineering perspective, the failure of the early warning system raises many questions as to the principles behind the design in the first place:
    1. Why the threshold timing of 2 minutes was selected for the polling interval – probably a misguided assumption from poor use of statistics as pointed out by Alex. 2 minutes was most likely deemed to be a ‘safe’ threshold that would only be surpassed ‘once every 50 years’.

    2. Why it was implemented as a polling system in the first place. Early warning systems, by their nature, are time-critical and should have been implemented in such a way that the sensor triggers when the condition is satisfied rather than each time it is checked every two minutes. How would you like it if the fire alarm waited two minutes before going off?

    3. What the sensor that alerts the PUB at 50% actually does, besides simply for logging purposes. If the intention is to provide an effective flood protection/warning system, perhaps this 50% sensor could be used to put the entire sytem into an ‘alert mode’ and alert a human operator which can then manually issue warnings or execute flood control actions in the case that the automated systems fail.

    • 32 yawningbread 9 June 2011 at 16:35

      From the Straits Times, it appears that there is only one the sensor which is placed virtually at the bottom of the canal. The sensor measures pressure — the theory being that the more water there is above the sensor, the greater the hydraulic pressure exerted on it. This begs another set of questions – e.g. while it is easy to calculate theoretically the correlation between pressure and depth of water under calm conditions, what if the water is turbulent, as would be the case in the midst of a heavy storm? What if mud is sitting on top of the sensor? Has the PUB tested the correlation between sensor thresholds and actual water depth under storm conditions?

      • 33 Slava 9 June 2011 at 18:09

        Well, what I have to say next is purely speculation because I don’t know what sort of sensors they deploy. The water is obviously flowing and pressure P goes as the negative of velocity squared (i.e P is proportional to -v^2) so the faster the flow (and it goes as a quadratic!), the lower the pressure. Now if, on their data collection centre, their code does not adequately account for this and they use the simple formula of pressure being related only to the amount of fluid above it, then we have a big problem because we can assume that the greater the rate of rainfall, the faster the discharge is into the storm drains and the greater the error can get in estimating the level of water. If this were the case, we can see how the system could get overwhelmed because of consistent under-reporting by the sensors and eventually when the drainage system got so overwhelmed, the velocity slowed and the sensors showed a huge jump in a matter of minutes. As I said, this really is just speculation because I don’t know for sure their implementation and really, this is taught at undergraduate levels and qualitatively in the A level’s so their mechanical/civil/hydraulic engineers implementing the system should know it for sure.

        That being said, I would think the more likely culprit is the accumulation of mud and debris over the sensor.

  19. 34 stngiam 9 June 2011 at 16:38

    Sigh. There are a lot of things you can blame the govt for, but inadequate historical rainfall intensity charts are not one of them. How do you think rainfall data was collected up till 10 or 20 years ago ? By some guy going out to a weather station and measuring the amount of water that was collected in a rain gauge ! How often do you think he would check ? Once a day ? Twice a day ? Twelve times a day ? Sure, you can do all these things electronically today, but how far back do you think high-resolution rainfall data goes ?

    The real problem is that humans think they can control nature. And in Singapore’s case, we even transfer that omnipotence to the government. The simple fact is that we can’t. We can learn to live with nature and co-exist with it, but we will never be able to bend it to our will.

    • 35 Resha 9 June 2011 at 23:26

      stngiam, I think you’re missing the point. What Alex has pointed out is that based on the chart, historical data tells us the 2 recent rainfalls which resulted in floods are 6 yearly events, not 50 or even 25 yearly events as asserted by the govt.

  20. 36 Daft Singaporean 9 June 2011 at 17:48

    Fat hope to have a Freedom of Information Act in Singapore so long as PAP is ruling. PAP knows very well “information is power”. By not giving or giving little information to the public is to ensure no one can check on them. This is precisely the outcome of a single party dominant political system. This is also the obscurantist policy to ensure easy governance and control. Instead of freedom of information, PAP government has this Official Secret Act. This Act is basically abused to withhold information, regardless of its nature, from the public so that the public has no or incomplete data to scrutinize the government. This is not an accountable and transparent government. It was not in the past 50 years and it will not be now and future. Only alternative party government may be accountable and transparent when people have a say.

  21. 37 Anonymous 9 June 2011 at 21:58

    If Marina Barrage was designed to alleviate flooding in its catchment areas, and the Stamford canal which runs through Tanglin and Orchard, why weren’t the pumps activated to rapidly lower the water level along the canal? The cumulative effect of 7 pumps at the barrage are touted to be able to pump the volume of 1 Olympic sized pool in 9 seconds. Surely that could have helped somewhat?

  22. 38 Alan Wong 9 June 2011 at 22:55

    Right from the start, engineering students are taught to design for the worst case scenario with a certain safety margin factorised in. Never on averages.

    Proper drainage design is based on so many factors : proper drain size/type; good distribution network/gradient of flow; correct start/end invert levels relative to existing ground levels, correct historical rainfall data, proper maintenance, no clogging, adequate catchment/discharge areas including necessary use of mechanical pumps if required, etc.

    So when LKY tried to explain that we have to end up with big canals all over Singapore if we were to based our designs on the worst case scenario, he may not be exactly honest with that kind of answer. What is rather more vital is that there must be a well designed and properly maintained distribution network to drain away the excess water as fast as it is collected.

    Any serious flooding incident should be viewed as an engineering failure. No need to blame god or the weather.

  23. 39 thetwophilo 9 June 2011 at 23:34

    Alex,
    I wrote a post when the first flooding occurred.
    In it, among others, I suggested that video cameras strategically mounted should be used to ‘eyeball’ the water levels at important points and waterways. My reasoning was that it would be faster and more accurate to monitor the water level in this manner.

    Months later, I read in the news that the PUB would be doing precisely that. I don’t know whether it was as a result of my input or something they came up with on their own.

    In any even, I am surprised that in this latest flooding, there was no mention of the cameras which the PUB had said would be mounted.

    It is also my suspicion that the barrage has a part to play because the flooding did coincide with the commencement of operation of the barrage. The barrage would effectively impound a lot of water as it is in fact the means by which the Marina reservoir is created. So on an ordinary ‘rainless’ you can expect the water level within the reservoir to be well ABOVE sea level. We do not have the data from the PUB of how much above sea level is the height of the impounded water. But common sense say it must be quite significant, in order for the Marina reservoir to function as one.

    Now one can well imagine the impact of this impounding on the water level throughout the entire network of streams, rivers, canals, storm drains that feeds this reservoir. The water level throughout would be effectively raised without a doubt to varying heights – depending on the actual elevation of these ‘waterways’ above/below sea level. This actually would represents an actual REDUCTION in carrying and holding capacity ‘at start’ of the network of a river, canal, etc if no parallel works had been done to increase its overall by deepening, widening, etc It is then small wonder that they would fill up quickly during a storm, much faster than previously before the operation of the barrage.

    One would imagine that whatever drainage network built by developers of buildings, condos etc in their properties, the water would have to eventually flow into the drainage network of the PUB. So you can imagine what would inevitably happen and has actually happened!

  24. 40 yawningbread 10 June 2011 at 00:59

    Grace Chua (reporter from Straits Times) emailed me. She provided additional information which I believe ought to be placed here on record, in order to be fair to her:

    QUOTE

    1. “Perhaps the system then failed to send out an alert because alerts are only triggered when water levels are in prescribed ranges (75 – 89% for the first public alert and 90 – 99% for the next one), not triggered by measurements that return a True finding when tested for >75% or >90% levels.”

    Yes, that’s correct. The PUB assumed (this was in the story) that sending out an alert above 100% would be redundant.
    We pointed out this was a rather obvious design flaw.
    They said (this was not in the story): “From our previous experience, it’s not necessary to flag 100% to people. All our previous data said that you would get a 75% or 90% reading within 2 minutes” (you try parsing that! What he meant was that they assumed testing water levels every 2 minutes would be enough to catch one of those conditions.)
    Again, we pointed out that this was an obvious design flaw…

    2. “Was it such an impossible event last Saturday that the water rose ‘three or four times faster than anything we have seen before’, as Peng Kah Poh said (quoted above)?”

    Certainly not impossible. BUT – sensors have been in drains only since 2007. (There was no reply from PUB on this that made enough sense to quote.)

    3. “Why not compare the 65 mm of 2011 with the worst half-hour reading from June 2010? Apples with apples? And if the PUB didn’t volunteer the information, why didn’t the reporter ask?”

    I am asking, but given the pressures of the daily news cycle, had to make do with what I had at the time. It is taking two days (or more) to get the rainfall information from the government. It doesn’t help that statistics are cherry-picked. If you can, PLEASE ask the NEA for relevant rainfall information. They may respond faster to the public. I agree that a Freedom of Information Act would be nice. At least one would get the data one actually asked for – eventually.
    ENDQUOTE

  25. 41 yawningbread 10 June 2011 at 01:01

    As suggested by Grace Chua, I have emailed the National Environmental Agency thus:

    QUOTE
    Dear NEA,

    I write on subjects of public interest for my website Yawning Bread
    (yawningbread.wordpress.com). I seek information from NEA about
    unusual rainfall events.

    After the flooding at Tanglin Mall on Sunday, the Straits Times report
    of 6 June 2011 said, “About 65mm was recorded within 30 minutes
    yesterday compared with 100mm within two hours on June 16 last year,
    which had led to serious flooding of Orchard Road.” I believe this
    information was given by the PUB to Straits Times, nonetheless I
    believe my questions are best directed to the NEA since the
    Meteorological Service is part of your organisation.

    I find it unsatisfactory that a half-hour rainfall reading for 5 June
    2011 was being compared with a two-hour rainfall reading from 16 June
    2010. I therefore have the following questions:

    1. Does the NEA/Met Service routinely monitor rainfall by 30-minute
    time periods?

    1.1 If No to Q1, then how was it that a half-hour reading was
    available for 5 June 2011?

    1.2 If Yes to Q1, when did routine monitoring of half-hour readings
    begin (i.e. when did such records begin?)

    1.3 If Yes to Q1, what are the ten most intense-rain half-hours on
    record (date / time / weather station)?

    1.4. If Yes to Q1, what was the highest half-hour reading on 16 June 2010?

    2. Does the NEA/Met Service routinely monitor rainfall by 60-minute
    time periods?

    2.1 If Yes to Q2, when did routine monitoring of 60-minute readings
    begin (i.e. when did such records begin?)

    2.2 If Yes to Q1 or Q2, what are the ten most intense-rain 60-minute
    periods on record (date / time / weather station)?

    2.3. If Yes to Q1 or Q2, what was the highest 60-minute reading on 16 June 2010?

    2.4 If Yes to Q1 or Q2, what was the highest 60-minute reading on 5 June 2011?

    3. Does the NEA/Met Service routinely monitor rainfall by 120-minute
    time periods?

    3.1 If No to Q1, Q2 and Q3, then how was it that a 120-minute reading
    was available for 16 June 2010?

    3.2 If Yes to Q3, when did routine monitoring of 120-minute readings
    begin (i.e. when did such records begin?)

    3.3 If Yes to Q1, Q2 or Q3, what are the ten most intense-rain
    120-minute periods on record (date / time / weather station)?

    3.4. If Yes to Q1, Q2 or Q3, what was the highest 120-minute reading
    on 5 June 2011?

    Thank you.

    Alex Au.
    ENDQUOTE

  26. 42 Senang Diri 10 June 2011 at 04:54

    Hi Alex

    You are amazing!

    Your articles of ” subjects of public interest ” are truly enlightening and a joy to read.

    Btw, PUB has a poor track record as far as IT is concerned.I believe that in the late 1990’s, they screwed up their SAP implementation.If I am not wrong, they recruited Indian IT outsourcing personnel who just could not do the job🙂.

    The cock up was mentioned in the ST but not about the recruitment of poor quality foreign IT professionals.

  27. 43 stanley 10 June 2011 at 12:05

    PUB is talking about building rentention pond to alleviate flooding in the Tanglin Mall area.

    What I dun understand is since Stanford Canal drains into Marina Barrage, why can’t they just pump the water out?

    • 44 Gard 10 June 2011 at 15:26

      “We move 13 million gallons of water a day when it’s not raining”
      – Why the Subways Flood, 8 Aug 2007.
      Source: http://cityroom.blogs.nytimes.com/2007/08/08/why-do-the-subways-flood/

      The above links refer to New York City. The Big Apple is no stranger to floods.

      “Ow Huay Ping, manager of Station Operations, SMRT said: “Within our tunnels, we have a pump system, so in the event that water does enter our tunnels, the pumps will be activated to discharge the water immediately. ”
      – ChannelNewsAsia, 19 Jun 2010

      This site has compared aerial photos of Orchard Road now and then:
      http://www.nowhere.per.sg/?p=1494

      Parks and gardens have no need to keep their ‘basement’ dry.

      If these buildings have their pumps to keep their basements dry, this is merely a matter of whose pump is more powerful.

  28. 45 Anon 10 June 2011 at 13:35

    stanley,

    My speculation is that the entire concept of the barrage was half-baked. Someone in PUB failed to take into account the time required for impounded water throughout the entire drainage network feeding the reservoir would take to respond to the opening of the ‘flood gates’ of the barrage.

    Maybe some one can compute the time it would take, say, water in the Bukit Timah Canal to fall (and at what rate) after the gates were opened at the barrage for the impounded water to flow into the sea. I wouldn’t be surprised that it would be the region of hours as there are multifactors involved that would determine the flow speed.

    My question is whether the PUB as built a proper model of the entire system using actual physical models or by computer simulation – in which case, accurate data collection would be crucial.

    Frankly speaking, I don’t have much faith in the PUB having done a thorough job of this given the political interference/influence – the old man’s personal interest – input that is bound to have taken place. Perhaps, it’s just too bad that the sloppy work is being exposed by Mother Nature herself! Putting the reaction of the NEA minister then and the statement made by LKY himself and the National Day Award to that high PUB official in charge and the timing of certain international ‘water’ event, it is hard not to consider this angle!

    • 46 Ben 12 June 2011 at 21:57

      It is practically possible for the Marina Barrage to keep the water level in the reservoir lower than the tide level by:

      1) lowering the tidal gates to pour any excess reservoir water which is above the tide level out into the sea
      2) pumping reservoir water out so that water level falls below the tide level outside.

      Therefore, the barrage should never be blamed unless those operating the barrage are asleep on the job.

      • 47 Brendan 21 June 2011 at 16:51

        Pehaps, there’s also a reason for keeping the tide high. Why ? Water sports such as dragon boating are held in the reservoir. Buy lowering the tide, can these sports take place ? As with other reservoirs, money is made from boat rentals by the hour. Another case of milking the cash cow at all costs again, huh ????

  29. 48 shornlock 10 June 2011 at 14:17

    I suspect that the perceived level of incompetence is linked to higher pay, if Daniel Pink is right: http://www.youtube.com/watch?v=u6XAPnuFjJc

  30. 49 ? 12 June 2011 at 08:55

    “However, on Sunday, water levels at that segment of Stamford Canal rose ‘three or four times faster than anything we have seen before’, said Mr Peng Kah Poh, director of PUB’s infocommunnications department.”

    “A total of 124mm of rainfall was recorded for the central area, about 77 per cent of the average monthly rainfall for June.

    About 65mm was recorded within 30 minutes yesterday compared with 100mm within two hours on June 16 last year, which had led to serious flooding of Orchard Road.”

    So we have seen 100mm/2h or an average 25mm/30mins. Three times of that is 75mm/30mins. Someone is blatantly lying..

  31. 50 S 13 June 2011 at 03:27

    I had an experience working with one of the PUB engineers,(manager) I can’t say it is indicative for all, but I stopped trying to get more work from PUB because of him. Frankly, I did not know where he got his master degree, the technicians were a better lots but they had no power. Throughout the project, he did not visit the installation sites once, and totally not interested in any engineering issues or implications or difficulties. Very clearly he was interested only in pushing his career according to the paper standard set by the PUB. The whole culture is just screwed up for any meaningful discussion of engineering planning and work. They just wanted fairy tale.

    The current CEO is getting all the heat now, I am not sure he is to be blamed as the culture was long formed before him.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s





%d bloggers like this: