New Model To Predict Floods Around Thailand

[webfact]

New model to predict floods

Pongphon Sarnsamak

The Nation

To simulate situations for week after getting weather data

BANGKOK: -- A team of researchers from the Hydro and Agro Infomatics Institute (HAII) is developing a new hydrological model to predict flooding around the Kingdom over any seven day period.

The new model is a combination of weather forecasting and hydrological modelling which will help researchers get a more accurate picture on a flood situation.

"We could simulate a flood situation for the next seven days within four-and-a-half hours of receiving weather and data from hydrologic telemetry devices that we will install along all channels to fetch actual water levels across the country," the head of HAII's hydro modelling section, Surajate Boonya-Aroonnet, told The Nation in a special interview.

Surajate, who led a team of researchers, said the output from this model will serve as key information for the Water, Flood, and Disaster Management Centre to make a plan and decide on how to cope with the flood situation.

"We want this model to be a centre providing reliable and accurate information to predict a flood situation in a short period," he added.

To date, in Thailand, there are only two models used to predict flood situations - a weather forecast model, used by the Thai Meteorological Department, and a hydrological model, used by the Royal Irrigation Department.

These two models work separately in predicting flood situations.

Now for the first time, two models can work together to predict flood situations more accurately.

In the near future, this model will be able provide all information about water and weather such as runoff, water discharged from dams, the levels of humidity and temperature, and the tidal level affecting water drainage systems from the Chao Phraya river basin to the Gulf of Thailand," he said.

Information about infrastructures in each area, such as the level of dykes, will be also put into this model to predict the direction of floodwater.

"We will know the exact direction of floodwaters after we get details about the exceeded volume of runoff water in a river. This model will take about 12 hours to calculate the direction of floodwater," he said.

"It could provide 7-day flood predictions, meaning people will have six days to prepare themselves," he added.

Surajate said his team expects this model will strengthen the flood prediction system and make more accurate forecasts during the upcoming rainy season.

-- The Nation 2012-04-17

[clockman]

Better to ask a fortune teller, and cheaper!

[looping]

"New model..."

For a moment I thought they meant something else, maybe T's daughter

[ResX]

You are making a progress. But then please understand that such model is worth third rank defence as far as flood control is concern. Without having first and second line of defence this third rank defence can cause more harm than good. The model is more useful for power generation than flood control since it can be used to optimize power output of the generating units.

Back here in my country we can do away with such model. So far our record to curb extreme floods for major flood control dams are better than Thailand. I'm not saying ithe model is useless but don't rely too much on it (for flood mitigation & control) just because such model is very advanced thing.

[kdc1899]

Is this the new model? (Pinky)

[steenasger]

I want to see it before I believe it!....that is the timely and accurate forecast results.

Many forecast models are already available, also in Thailand, but here at least two main problems remain:

(1): the maintenance of the telemetric systems for providing timely data input to the models

(2): the dessimination of timely, relevant and correct forecast results to the local people affected by the floods

[Lifer]

Is this the new model? (Pinky)

More ridiculous comments by TV posters.

[Lifer]

"New model..."

For a moment I thought they meant something else, maybe T's daughter

More ridiculous replies by TV posters!

[animatic]

There should have been similar simulation models in place

and continuously updated for the last 10 years here.

Only NOW do they think to tie rainfall and water levels

and flow of water in different watersheds together in a unified model???

No communication between meteorology and the water conservation departments til now?

Edited April 17, 2012 by animatic

[jsgatse]

Instead of spending a ton of money on being able to accurately tell us how much we are going to get screwed by rising flood waters, why don't they invest in fixing the ROOT issues that are causing the problem?

[DoctorD]

HAII is (relatively) independent of the government and runs its own network of telemetry stations around the country. They are very much on top of the technology as well as the scientific principles of modeling.

Flood modeling on a national basis is an extremely complex objective and of course will never be totally accurate. (The nature of a model is that it serves as an abstraction of reality.) However, there's no reason to bash this attempt. Integrating multiple domains usually makes models more accurate (at the cost of increasing their complexity).

As for spending money on eliminating the "root causes" of flooding, this is of course important, but the growing number of extreme weather events means that prevention is always going to be behind the curve. If this model proves to be accurate, it can contribute significantly to the country's flooding problems by providing at least short term quantitative predictions.

[ResX]

Okay. Now the model predicts Bhumibol and Sirkit dam will be getting 7billion cubic meter of flood waters over 7 days duration beginning from next Monday. What the dams operators can do? Please do not release 7billion cubic meter of waters in 7 days prior to the coming Monday. Otherwise at best they just make the flood to come even earlier. At worst, the model has predicted wrongly but the flood has occurred by mistake.

Don't get me wrong. The model is useful as long as the strategy to use it is correct. But don't over estimate its capability.

Edited April 17, 2012 by ResX

[khaowong1]

Is this the new model? (Pinky)

Now you tell me who wouldn't listen to her about any subject you can name? :

[otherstuff1957]

There are plenty of competent scientists and engineers in Thailand, the problem is that they don't run the country, the politicians do. I have no doubt that the HAII can do a fairly good job at tracking a predicting water movement. However, what happens after they give this information to the government? Hopefully it will be acted upon in an intelligent manner, but I'm not optimistic.

[PoodMaiDai]

Why don't they just use the GT200 bomb detectors? I'd imagine those are just as good at predicting floods as whatever contraption they spent millions of baht on this time.

[PoodMaiDai]

There are plenty of competent scientists and engineers in Thailand, the problem is that they don't run the country, the politicians do. I have no doubt that the HAII can do a fairly good job at tracking a predicting water movement. However, what happens after they give this information to the government? Hopefully it will be acted upon in an intelligent manner, but I'm not optimistic.

I strongly believe, the best of the best leave Thailand for better pay and more opportunity.

If I was a competent scientist and had the opportunity to work abroad where my work may make a difference, people may listen to me, and the pay would be better, I'd be outta here.

But you are correct, those who do remain are ignored by politicians.

[philw]

There should have been similar simulation models in place

and continuously updated for the last 10 years here.

Only NOW do they think to tie rainfall and water levels

and flow of water in different watersheds together in a unified model???

No communication between meteorology and the water conservation departments til now?

Ahhhhhhh.

But the NOW is the hopeless, Burberry wearing, useless, incapable of independent or coherent thought and speech, you guessed it, Yingluck government.

Maybe Thailand would be a better place if she had taken up politics 10 yrs ago.......

Just saying, like.

[lungmi]

Okay. Now the model predicts Bhumibol and Sirkit dam will be getting 7billion cubic meter of flood waters over 7 days duration beginning from next Monday. What the dams operators can do? Please do not release 7billion cubic meter of waters in 7 days prior to the coming Monday. Otherwise at best they just make the flood to come even earlier. At worst, the model has predicted wrongly but the flood has occurred by mistake.

Don't get me wrong. The model is useful as long as the strategy to use it is correct. But don't over estimate its capability.

In my family I have one retired and one still active engineers of the Bhumipon Dam.

"No one listens to us, no one follows the instructions of our venerated KING Bhumiphon. This is Thailand."

[hellodolly]

There should have been similar simulation models in place

and continuously updated for the last 10 years here.

Only NOW do they think to tie rainfall and water levels

and flow of water in different watersheds together in a unified model???

No communication between meteorology and the water conservation departments til now?

I think your question mark was a whole bunch to small.

I was just thinking lately

1 We were told they were going to eliminate traffic deaths during Songkhran

2 We were told we had a good Tsunami warning system that warned us about a Tsunami that didn't happen and

3 We had a good earthquake warning system that didn't warn us of three of them.

And now we are going to have flood predicting machines.

I feel so munch safer now.

[hellodolly]

HAII is (relatively) independent of the government and runs its own network of telemetry stations around the country. They are very much on top of the technology as well as the scientific principles of modeling.

Flood modeling on a national basis is an extremely complex objective and of course will never be totally accurate. (The nature of a model is that it serves as an abstraction of reality.) However, there's no reason to bash this attempt. Integrating multiple domains usually makes models more accurate (at the cost of increasing their complexity).

As for spending money on eliminating the "root causes" of flooding, this is of course important, but the growing number of extreme weather events means that prevention is always going to be behind the curve. If this model proves to be accurate, it can contribute significantly to the country's flooding problems by providing at least short term quantitative predictions.

Well I will except what you say as being fact,

What I don't understand is how a 7 day warning is going to help munch if 358 days of working to avoid the problem don't help munch.

Yes I know 7 days will give the people time to move out of their homes and hopefully save what is in them. But the damage to the homes will still be there. And if handled like the last flood many of them will not have a job to go to.

[ResX]

Okay. Now the model predicts Bhumibol and Sirkit dam will be getting 7billion cubic meter of flood waters over 7 days duration beginning from next Monday. What the dams operators can do? Please do not release 7billion cubic meter of waters in 7 days prior to the coming Monday. Otherwise at best they just make the flood to come even earlier. At worst, the model has predicted wrongly but the flood has occurred by mistake.

Don't get me wrong. The model is useful as long as the strategy to use it is correct. But don't over estimate its capability.

In my family I have one retired and one still active engineers of the Bhumipon Dam.

"No one listens to us, no one follows the instructions of our venerated KING Bhumiphon. This is Thailand."

Thank you. I know that. "Politicians like to get hold the problem they understand rather than the solution they don't as long as they get votes".

[ResX]

HAII is (relatively) independent of the government and runs its own network of telemetry stations around the country. They are very much on top of the technology as well as the scientific principles of modeling.

Flood modeling on a national basis is an extremely complex objective and of course will never be totally accurate. (The nature of a model is that it serves as an abstraction of reality.) However, there's no reason to bash this attempt. Integrating multiple domains usually makes models more accurate (at the cost of increasing their complexity).

As for spending money on eliminating the "root causes" of flooding, this is of course important, but the growing number of extreme weather events means that prevention is always going to be behind the curve. If this model proves to be accurate, it can contribute significantly to the country's flooding problems by providing at least short term quantitative predictions.

Well I will except what you say as being fact,

What I don't understand is how a 7 day warning is going to help munch if 358 days of working to avoid the problem don't help munch.

Yes I know 7 days will give the people time to move out of their homes and hopefully save what is in them. But the damage to the homes will still be there. And if handled like the last flood many of them will not have a job to go to.

You can see the real problem right? As I mentioned above, such model cannot be used just like that. The way that it can be used to add value is to provide the likelyhood for peak flood flows over the next one week so that the releases from all dams can be averaged out over a longer duration, By doing so the peak flood flow can be trimmed down. . For example, without the model we maybe forced to release 7 billion cubic meter of waters from all dams only when they come. With the model, that can predict 7 days ahead we may want to discharge 7 billion cubic meter in 14 days. This will cut down the peak discharge by 50% and it can be made with minimum negative implications.

Edited April 17, 2012 by ResX

[surayu]

communication between meteorology and the water conservation departments

I think they were having some intense communication times between departments with the old model of predictions...

[sirchai]

"These two models work separately in predicting flood situations".

"Now for the first time, two models can work together to predict flood situations more accurately"

Please see attached files for your deepest consideration Miss Ynglcku......

Edited April 17, 2012 by sirchai

[animatic]

There should have been similar simulation models in place

and continuously updated for the last 10 years here.

Only NOW do they think to tie rainfall and water levels

and flow of water in different watersheds together in a unified model???

No communication between meteorology and the water conservation departments til now?

Ahhhhhhh.

But the NOW is the hopeless, Burberry wearing, useless, incapable of independent or coherent thought and speech, you guessed it, Yingluck government.

Maybe Thailand would be a better place if she had taken up politics 10 yrs ago.......

Just saying, like.

So you imagine the Yingluck and co. proposed this idea.

Not on your tintype, this has academic written all over it.

The same ones proposing it for the last 20 years.

It could have been put in place by her blubbering Burberry wearing Brother Thaksin 10 years ago,

during his allegedly golden era for Thailand, but he ignored it blithely when it was proposed then.

This is just more swamp gas camouflage to hide the fact that one of their own

ordered the dams to hold back water to save his friends rice crops.

Cynically, or greedily saving the crop for the rice pledging schemes political payback model.

More smoke and mirrors, doing too little too late.

It's a good idea 10 years late, if it isn't charade to save face and distract the populace.

Edited April 18, 2012 by animatic