Saturday, December 31, 2011

Back in the US of A

It has been a whirlwind last couple of weeks, visiting family in the US. We had some trepidation, not having been back home since moving to Australia three years ago. We first visited San Francisco, then Colorado, and Connecticut. 

We had a lot of catching up to do so we dove right in with some junk food in Colorado...

Sometimes called pigs in a blanket
The nutritional menu at Burger King (known as Hungry Jacks in Australia). For reference, 1230 Calories  is about 5 Mega-Joules. 82 grams of fat would fill one of those small travel shampoo bottles. 
And some guns... We drove by the famous Gunsmoke Gun shop in Wheat Ridge, Colorado.

Assault weapons are legal. Really you can buy anything you want.  
... even a hot-pink handgun...
We set off from Denver, west and south (through Fairplay, Buena Vista, Salida, Canon City, Pueblo and Colorado Springs... 425 km/265 miles in all) to visit family. Along the way we got to take in some hydrologic tourist hot-spots. 

We stopped by the Roberts Tunnel, one of the many tunnels that brings water from the west side of the Rocky Mountains down to the east side. 
Stern warnings outside the Roberts Tunnel
Roberts Tunnel outlet
After a minute of taking pictures, a police cruiser pulled up and the officer started asking questions. It's not a good idea to be taking pictures of important infrastructure like this, "after, you know, 9-11 and all." After listening to my story and reading my business card (which describes me as a "forecasting enthusiast"), he decided I was probably harmless enough and suggested I move along.

Colorado is full of immense landscapes
Later on that day Kitty's mother showed me the Hutchinson Homestead Ditch, just walking distance from where she lived. 
Hutchinson Ditch, Colorado


The gate splits the river flow in two, left or right or both, depending on how the gates are cranked down.

We saw a few streamgages along the way, such as this snowbound stilling well near a frozen river (I believe it was the South Fork of the South Platte). 


But the real delight for me was seeing the streamgage for Chalk Creek at Nathrop.

Streamgage for Chalk Creek at Nathrop
I had been a forecaster for this location for a number of years and had never been to it in person even though I had driven near it several times. The gauge has existed for over 60 years and is relatively free flowing. It drains 14,000+ foot (4,300 m) mountains, just a few miles away. The average slope, before the river hits the valley floor is about 20% (as mountains go, that's steep!)

We tarried much too long though and ended up driving our tiny rental car in white-out conditions on the highway clear past midnight. It was a feat considering that we hadn't seen snow in years and hadn't driven in months. We arrived safely and managed to have a White Christmas with parents, however.

To keep warm, I wore three pairs of pants and every shirt I own. 

Tuesday, December 20, 2011

The Empty Military Dam of South Korea

Speaking of items in the news (North Korea's leader recently passed away), another unique hydrologic feature is a reservoir that waits empty in case North Korea ever released a wave of water against Seoul. The Peace Dam can store a volume about one tenth the size of the lake behind Hoover Dam, but it nearly matches the volume behind Geumgang Mountain Dam in North Korea. Ever since the 1980s there has been concern that a North Korean dam could either purposefully (in the case of war) or accidentally (in the case of neglect) cause a disaster downstream. Rumor has it that North Korea attacked South Korea three times with floods in the last ten years, with six people dying two years ago. If used, the Peace Dam might even back up and flood some areas in North Korea, so the threat goes both ways. 

A wide shot of Peace Dam from Panoramio
The dots in the center (people) give a sense of the scale of the dam (from Daniel Plummer's travel blog) 
The location of the Peace Dam


My blog here has a number of other links to hydrologic oddities, found by searching the archives, or looking through the "best of" page.

Saturday, December 17, 2011

Typhoon Flooding in Philippines Kills at Least 600

Earlier this year we visited the Philippines flood warning center in Manila as two major typhoons struck, back to back (those last two links point to this blog's stories about that). Last night another typhoon (Sendong/Washi) tracked across the southern islands, passed back over the ocean and then hit land again a few hours later near Puerto Princesa (we also visited Puerto Princesa's Underground River and wrote a three part series about it).

According to the weather and hydrology agency, PAGASA, 180 mm (7 inches) of rain fell at Hinatua causing flash floods. Six bridges were destroyed. Many stories put the death toll at 400 although numbers like this quickly rise (a story out in the last hour says 600 dead).

Village washed away by flash flood
Some quotes from the news

Iligan City Mayor Lawrence Cruz said many people were caught by surprise when water rose one meter (three feet) high in less than an hour, forcing people onto roofs. "Most of them were already sleeping when flood waters entered their homes. This is the worst flooding our city has experienced in years."...then....

“I can’t explain how these things happened, entire villages were swept to the sea by flash floods,” Galon said. “I have not seen anything like this before. This could be worse than Ondoy,” he said, referring to a 2009 storm that inundated the capital, Manila, killing hundreds of people.


Another story suggests some difference in the magnitude of the rainfall forecasts between agencies (although technically TRMM is not a forecast but rather a measure of what rain is already following). That same story expressed concerns about the storm warning signal being based only on windspeed and not volume of rainfall, a concern echoed by hydrologists because rainfall is what causes floods and is most relevant.

TRRM [sic] analysis of data gathered by the satellite's instruments showed that Sendong was already carrying "very heavy rainfall of over 50mm/hr (~2 inches).” In comparison, Ondoy dumped 56.83 mm/hr of rainfall on Metro Manila in September 2009. Around the same day, Philippine state weather bureau PAGASA and the disaster management agency NDRRMC predicted only a 10-25 mm per hour rainfall amount for Sendong. The highest storm warning signal raised over areas affected by Sendong was only signal number 2.


There are some videos of the floods

Video 1
Video 2

The latest videos are likely available across the header of the ABS-CBN website.

Wednesday, December 14, 2011

Pashupatinath Temple in Kathmandu

Always on the lookout for hydrologic oddities, I usually do a lot of neck-craning driving over bridges. Every time we came anywhere near the airport in Kathmandu, however, the driver would wave upstream and mention "That place, very holy, very famous". That place was the Pashupatinath Temple, the most sacred site dedicated to Shiva in the world, built before 600 AD. It is a UNESCO world heritage site and the complex has close to 500 temples. 

Too bad for us, but only Hindus are allowed inside the actual complex. Kitty and I don't exactly blend in, especially with me on crutches, and we only got about 15 meters in before a cane-yielding guard accosted us and turned us around. Non-Hindus can go around the edge of the grounds to where cremations take place on the bank of the river, however. If the pictures look small and distant it was because I was trying to be respectful; the area does have a solemn air to it.  
Cremation area along river

Visitors stay on the far bank, and mourners stay on the other. 

The corpses are wrapped in cloth and brought to the Bagmati River's edge where they are blessed. 
A few bodies arrived in shipping crates and caskets, suggesting to us they were from overseas
A funeral pyre of wood and grasses is prepared and then the bodies are set alight. After the cremation is finished, the ashes are washed into the river. The ceremony is an all-day affair; we were there for three hours and only witnessed some of the process. There are many pyres going on in parallel along the river banks.  
Cremation stations along the far bank

Ashes are swept into the river
The river itself is not clean, aside from the obvious things going in it. I'm in no position to judge, but it is always a little disorienting to see someone lay chrysanthemums and other items in ceremony on their loved ones and then immediately toss the plastic bags in the river. Gangs of children would fish through the river to collect coins left with the bodies. Occasionally a cane-yielding guard would come and half-heartedly chase them off. 
Wild monkeys abound, wandering around the river's edge
In a nearby park there are more monkeys than you can shake a stick at. 
In terms of water supply and drought, there is a story that in a drought in 1964, "the people of kathmandu pleased Pashupatinath by filling the temple with the last bits of water until the Shivalingam drowned in the water. At that very moment dark clouds arrived and heavy rainfalls made an end to the drought."

Tuesday, December 13, 2011

Hunting for the Oldest Raingauge in Nepal

[We're going to take a few days off from preaching about "The Forecaster's Way" to catch up with a backlog of stories about some of my travel experiences in Nepal]

During my visit to Kathmandu, I had chance to meet Keshav Sharma, the Director General of the Department of Hydrology and Meteorology. Note that Nepal is the only instance I can think of where hydrology comes before meteorology in a national weather service's name. Nearly always it is the other way around, if hydrology is mentioned at all. Regardless, he was extremely gracious and generous with his time. I'm not just saying that. Here is a photo of him coming to meet me for an interview on the morning of Tihar (Diwali), the most important Hindu festival of the year. It is the crescendo of a month of holidays in Nepal. In the US, it would be the equivalent of NWS head Jack Hayes coming to meet you on New Year's morning. 
Keshav Sharma (head of Nepal weather service) and I on the Hindu holiday Tihar
Dr Sharma gave me a copy of his latest book "Climate Change Trends and Instances of Socio-Economic Effects in Nepal" which is similar to a PDF online "Climate Change Trends and Impacts on Livelihood of People". It contained this passage about precipitation measurements in Nepal:

Precipitation recording in Kathmandu was started on a regular basis in 1921 in the premises of the Indian Embassy at Lainchaur.... The station was gradually upgraded into a climate monitoring station through the years. The lone station of Lainchaur got a network character in the Kathmandu valley with two additional precipitation stations in Sundarijal established in 1940.

When I was coming back from meeting Anup Phaiju from Practical Action, a charitable organization that helps with flood early warning systems in Nepal, I realized that I would be walking by the above mentioned Indian Embassy in Kathmandu. Wouldn't that be great material for the blog, to get a photo of a 90 year old raingauge, the oldest in Nepal?   

Well, lets just say, "One does not simply walk into Mordor". In retrospect, I'm not quite sure I knew what I was thinking. There's no peeking over the fence here. The embassy grounds are expansive and are ringed by 30 foot tall walls, barbed wire, floodlights and so on. On the other side I pictured there were guard dogs.... or bees. Or guard dogs that barked bees. 
The outer wall of the Indian Embassy in Kathmandu
It was not a total loss. While walking around the edges I did stumble upon a Maoist rally, which was a first for me. There were many classic red hammer and sickle flags, but they also had some with crossed forks and spoons instead. Politics aside, I think the dinner theme resonates more with the common man than industrial tools do. But I digress...

I gave up the hunt until a few weeks later when we were turned away from a flight because we hadn't secured the proper visas to travel through India. Not travel to India, but travel through on a layover flight to Sri Lanka. They sent us off to the embassy to get a transit visa. 

Unluckily for us, we got there on a Friday as they were closing and there would be no hope* until Monday to get a visa and there was some rumor that Monday was a holiday ("again?!?"). With little left to lose, I decided to press my luck to see if I could find that famous measurement station. 

*And by "no hope" I mean that a travel agent (standing on the street listening to us get rejected by the embassy) offered to get us an "after-hours" visa for obscenely large sum of money. "No problem, what do you want?" he said, "They call me the problem killer". 
Hello? We were wondering if you had any raingages? 
I was able to pass Dr Sharma's book through the gate to the guards to see if they could help. Their English was not very strong (as was my Nepalese) and the incident was escalated up through several layers of security. Kitty sat at a nearby cafe and was surprised to watch me eventually pass through the small blue metal door in the gate. She thought it would be the last time we would see each other and started working out her own Plan B. Actually, this was only the outer gate. I'm not sure if I'm allowed to say, but there is another guard station inside, including more gates, cameras, metal detectors, etc.. No bees, are far as I could tell, however.

It was difficult to "explain the nature of my request" but eventually I met the director of security (a tall well dressed man in a suit) who told me that the Indian Embassy didn't exist in 1921 because India was not yet a country. The Indians inherited the embassy from the British in 1954 and he said I might have better luck asking at the British embassy, where they would have the records. "Where's that?" "Across the street". Sure enough, literally across the street was the UK embassy. It has its own signs saying that it didn't accept visitors, that visa inquiries should go to another office across town. The heavy metal "klong" of the gate behind me suggested we were done for the day.

Sadly, I now know that the Indian Embassy gauge is long gone, last recording in the 1970s. However, it does appear that there are records even earlier than 1921. Researchers at Columbia University tried to create a long record of temperature and precipitation in Nepal by blending tree ring data with station measurements. From their 2003 journal article,

To do this, we located the original published data sources to obtain all of the monthly climate data. These sources included the ‘Climatological Records of Nepal’ (DIHM, 1977) for 1921–75 data, the ‘India Meteorological Memoirs’ (Eliot, 1902) for 1851–1900 data, and the Annual Summaries of the Monthly Weather Reviews for 1901–20. The first two sources provided the identical monthly data that was previously [available], still with the 1901–20 data gap. The last data source, found in the British Meteorological Office Hadley Centre Library, contained the missing 20 years of data.

They thought some of the data was lost. From the researcher's webpage "Dr. Jones, searching through a unique archive of the late-British colonial period (1901-1920), found a rare copy of the Kathmandu Meteorological records". It is a good thing he did, if you look at the original data and their extended reconstructions, that previously lost period contained some of the driest years on record.

Data rescue deserves its own discussion, but I will say that if you're still looking for ideas for holiday gifts, consider giving the gift of data at the International Environmental Data Rescue Organization.

Sunday, December 11, 2011

All I Really Need to Know I Learned From Forecasting

Following on a post from earlier this week with warnings about not getting too literal with science analogies for daily life, here are some more lessons from "The Forecaster's Way" that I do think are useful.

These are less centered around forecasting as a science, but rather forecasting as an operational job in which deadlines are tight and pressure can be high. "Operational" means basically that someone is relying on the forecasters' products and expecting them to be there when the products are needed.

Perfect is the Enemy of Good (a quote from Voltaire)

See also, "perfect" is the enemy of "finished." Hydrologists need to triage and prioritize to get their jobs done. Don't we all have to do that though? Forecasting deadlines are hard and timeliness matters. Remember the scene in the movie "Trading Places" where the commodity trading pits go into a shouting frenzy when the results of latest orange juice crop survey are announced? The official didn't even finish "The cold winter has apparently not affected the orange harvest..." before the traders erupted into hair pulling and eye gouging.


It can be a little like that in the trading pits when Western US water supply forecasts are released because they affect the prices of hydropower and natural gas. I remember that the forecast release schedule was announced months ahead of time so that any individual trader didn't get an unfair advantage by being the first to know the forecasts were out. 


Therefore, the hydrologist spending all his time polishing a given forecast to perfection is probably not giving enough attention to all the other forecasts that have to go out the door. Pace yourself and plan ahead, but most of all, learn when to let it go. Furthermore, when trying to build new forecasting systems and adopt new technologies, someone may want to wait around for, say, the perfect model, but in the meantime he may be passing up opportunities to use perfectly adequate models.


Another way of looking at it is "Forecasts are never finished, only abandoned" (apologies to da Vinci). There are many variants on this, "Dissertations are never finished...", "Reports are never finished...". As an aside, problems do start when my idea of "good enough/done enough" is different from yours. Indeed, I seemed to have this problem every time I had someone do remodeling or house repair. I can't tell you how often I have been left with abandoned jobs.

Periodically Test Your Backup Plan 

Reliability and resilience are highly valued qualities in operational systems. When things are at their worst, people with operational responsibilities have to be at their best. Users do not usually accept hardware malfunctions as an acceptable excuse for forecasts being late or not being issued. It is your job to make sure that either your computer does not crash or that you do not lose all your work when it does. Naturally, there are tradeoffs- for example, US government agencies' stringent computer security policies are enforced as if by a stern martinet. I know examples where this has stifled innovation.

This plan should have been tested. 
This is not to say what your backup plan should be, but rather that you should test that it actually works. Few things are more frightening than pulling the emergency release handle and accidentally having it come off in your hand. Maybe you bought an external hard drive to backup your computer files? Have you ever actually tried to restore a file, in a non-emergency situation, just as practice to see if it works? You should! When backup plans fail is when problems turn into fiascoes.

This checking takes discipline. As Andrijanto, an Indonesian dam operator, said to me once, "The most important thing in water management is discipline."

Tuesday, December 6, 2011

People Are Not Particles

A few posts ago I mentioned "The Forecaster's Way" in the hopes that the philosophy of what makes a good forecaster could also be applied to daily life. During my time at Vipassana meditation I remembered a few other "tenets" of what might make a good personal philosophy (more on that tomorrow).

Before I go much further however, it would be good to recognize the danger of applying science analogies to daily life. A theme running through the meditation center's lectures was that quantum physicists have finally come around to recognize the truth in some of Buddhisms teachings. This includes ideas like matter is not solid, that instead it is made up of mere "subtle vibrations". Also, everything is impermanent, matter flicks in and out of existence trillions of times a second, etc.

Nothing rankles physicists quite like when non-scientists try to apply physics laws to their daily lives. The radio show This American Life has a fantastic and funny episode called "Family Physics" that discusses this; I highly recommend it. As some context, I have a physics undergraduate degree and spent a summer as a student working with the particle accelerators at Brookhaven National Labs.

Honestly, you might be surprised at just how far quantum physics is outside the realm of your ordinary daily life. It is entirely likely that you could survive in the woods for a long time without knowing anything about modern physics. This is because the laws of motion figured out over 300 years ago work (and will continue to work) just fine for human- or breadbox-sized objects.

Dilbert.com
Dilbert discussing the project uncertainty principle
In comparison, there is no survival challenge where you would need to apply the well-known Heisenberg Uncertainty Principle. That is the one where the better you want to measure where something is, the less able you'll be able to know how fast it is going.

This is probably the most abused and misinterpreted physics principle, commandeered into cringe-worthy chains of logic like, "nothing can be measured exactly...therefore there is no universal reality." Granted, it may be true that there is no universal reality, but the proof to you is not going to be how electrons like to pass through a hole in a wall.... (As an aside, ponder that paradox! Can the statement "there is no universal reality" be itself a universal reality?... Far out!... But I digress...).

When you actually dig into the equations, if you want to measure something like a baseball toss or cricket ball pitch speed to within about 10%, the Uncertainty Principle limits your ability to measure the ball's location to within one trillionth of a trillionth of a trillionth of a millimeter, give or take. My GPS struggles to know where I am within 100 meters, so there are lots of other issues to resolve before quantum physics gets involved.

In the end, probably the most useful idea I got from my physics degree was that "People Are Not Particles." In the vacuum of outer space, "a body in motion tends to stay in motion", true. However, once your mother-in-law gets it in her head that the whole family is going to vacation in Greece this year, even though we already had plans to go to Bali, does that mean there's no changing her mind? Not necessarily. Or that "for every force there's an equal and opposite force" implies that Karma is going to catch up with you eventually? Actually, as Mark Twain wrote, sometimes the bad little boy doesn't get punished, and the good little boy doesn't get rewarded.

A quantum physicist should probably not be your first stop on the quest to discover the meaning of consciousness or to learn the behavior of the mind. It is probably better to ask a psychologist because they specialize in that kind of thing. Maybe you would even want to consult a positive psychologist if you are concerned about how therapists mostly focus on how broken you are. Maybe even consult a happiness engineer if that is what you're really after?

My advice is, "Don't Panic"

Sunday, December 4, 2011

Icelandic volcano could cause massive flooding

This item in the news recently. The link goes to a video that has neat pictures.

Icelandic volcano could cause severe flooding scientist warns

Hundreds of metres under one of Iceland's largest glaciers there are signs of an imminent volcanic eruption that could be one of the most powerful the country has seen in almost a century.
Mighty Katla, with its 10km-wide crater, has the potential to cause catastrophic flooding as it melts the frozen surface of its caldera and sends billions of gallons of water surging through the east coast and into the Atlantic ocean.
...The last major eruption occurred in 1918 and caused such a large glacier meltdown that icebergs were swept by the resulting floods into the ocean. The volume of water produced in a 1755 eruption equaled that of the world's largest rivers combined.

Saturday, December 3, 2011

Modelling the Monks' Mess Hall (Part 2)

[Yesterday we discussed how to model monks going in and out of a cafeteria and how it is like a hydrological model, with rainfall going in and streamflow going out. This is based on my experience with a 10-day silent meditation program in Kathmandu, Nepal. Today we talk about how, even though simple models can tell most of the story, problems can still happen.]

So where does it all fall down? Why are there dozens upon dozens of different models out there? And why are they such a challenge to use for real-world forecasting? 

Are we capturing all the processes?

Besides queuing and eating there are actually a fair number of other processes going on. When a server's station is empty (e.g., we're out of the brown bean slop), how long is the line stopped while the station gets refilled from the kitchen? What happens to the monks' plates when they are finished- do they have to wash dishes themselves and if so, what kind of delay does that cause? Maybe some monks show up, see too long of a line and decide to come back some other time or skip the meal? Occasionally, a wild papadum-eating monkey wanders on the scene. What effect does that have?

Sometimes these things are not an issue (e.g., there is always enough brown bean slop in Kathmandu) and so they do not have to be part of the model. For example, in Australia there is so little snow that nearly all streamflow models in use there do not bother modeling snow. However, you would not be able to get away with that in northern Alaska where snow rules the water cycle. 

Maybe a process cannot be ignored, but its effect could be lumped in with some other process. Say it takes on average 15 minutes to eat and 3 minutes to wash dishes, a complex model might include both processes separately. A simpler model might combine them into one "eating and washing dishes" task with an average time of 15+3=18 minutes. Most importantly, if all you can do is watch monks go in and out and never know what is inside, it can be hard to distinguish all the individual internal processes from eachother. Similarly, I know how long it takes for food to pass through my body, but I would be stumped to tell you how long it spends in my stomach versus my lower gut. 

In hydrology, for example, water leaves bare soil or lake surfaces to the sky through evaporation. Transpiration is the same thing, except it is done by plants. Often times these are lumped together under a single notion of "evapo-transpiration" and this is good enough for most purposes. 

Modelers might also argue about the configuration of what is going on inside the cafeteria, e.g. are there many different lines or just one? Are there lines to wait for a seat? There might be special models needed for banquets, family meals, and so on. 

Go to war with the army you have

In the case of the Vipassana meditation center, I can say that breakfast was relatively short compared to lunch. One modeler might say that this was so because the monks wanted to finish quickly and race back to bed for some extra sleep (true story!). In that case, the rate of eating would depend on the time of day. Let's call the proponents of this theory "time-of-day-ists" (and its staunchest supporters are a professor from Miskatonic University and his clan of students).

Another school of thought might say that the mess hall was cold in the morning and that monks wanted to race back to bed where it was warm (also a true story!). These "temperature-ists" hail from Medfield College. Occasionally they meet at scientific conferences and get along for the most part, but they both hold strong convictions about how they believe the cafeteria behaves. 

One test of who is correct would be to see if the length of lunchtime varied based on how cold it was, relative to other lunchtimes. An extreme case might be that strange day where it was colder during lunch than at breakfast. If the monks showed up to a cold lunch and decided to skip it altogether, the International Journal of Cafeteria Processes might be floodeded with articles from the "temperature-ists" politely lambasting the "time-of-day-ists" as misguided empiricist nitwits.

This is nice and all, but even if meal length depends more on temperature than time of day, sometimes time of day is going to have to be good enough because you do not actually have temperature measurements at your cafeteria (but everyone has a clock). As Donald Rumsfeld said, "you go to war with the army you have, not the army you might want or wish to have at a later time." Temperature may be the "real" cause for how meal length varies, but if you do not have temperature data to give to the model, it is a moot point. Instead, you would need to find something that varies like temperature does, to approximate the effect (such as time of day).

In hydrology the parallel would be snow models. The rate that snow melts partly depends on how warm it is, but really it is a combination of radiation, windspeed, humidity and a whole host of factors. Some modelers recoil in horror at the idea of using a temperature index to figure out how much snow has melted but the problem is that nearly nobody measures all those other variables in the mountains. However there are lots of measurements of temperature in snowy areas going back dozens of years. So, operational forecasters have to resort to using temperature-based models. Their results might not be as good as they could be, but commonly they do not do too bad of a job. It is good enough for government work, as they say.

Lets throw another spanner in the works. Dinner at the Vipassana center was the shortest of all the meals. This was because returning students are only allowed hot lemon water for dinner while new students are allowed a banana and a small bowl of something like toasted savory rice krispies. Similarly, less food is typically served at breakfast (e.g. 3 kinds of gloop) than lunch (e.g. 5 kinds of gloop). 

What was served varied from meal to meal (image from Full Thangka blog). This looks like breakfast. 
Right, so now the "time-of-day-ists" can return the volley from the "temperature-ists" because they can point to how it is typically just as cold at dinner as it is at lunch, even though the time to eat is often radically different... ("respectfully, who's the nitwit now?")

The reality is a mix of all these factors, both temperature and the kind of food being served at various meals. In a way, all the modelers are a little bit right and a little bit wrong. Time of day is not a direct factor but it is a good proxy for some other factors. Therefore, even models that include "unphysical" processes can still be effective. In the meantime, however, models multiply, dissertations are published and the trench warfare among modelers continues.

Turning the general into the specific: Parameters and calibration

A cafeteria model might include a general description of how a serving queue works. What it does not know straight off is the speed of the queue at a specific cafeteria (e.g. the one for students at Kathmandu's meditation center, as opposed to the cafeteria for staff at the Children's Hospital in Boston). Modelers tend to leave things like that as a "parameter". For example, serving lines serve X customers per minute. For the Kathmandu meditation center X = 5 customers per minute. Sometimes parameters are observable but sometimes they have to be inferred from historical data.

To model the meditation center, a cafeteria-ologist might observe people going in and out for a while and then back-calculate what he thinks the speed of the queue is. Maybe this historical data is not available and has to be guessed at by some other means or directly measured. In hydrology, things like the size of a watershed are readily measurable from maps. However, some things are not, such as the rate that water drains from the soils. That parameter would have to be found by trial and error and/or using human expertise.

How you come up with these parameters is a major subject of debate in the hydrological research and operations communities. I won't go into the details now because the issue is technical and complex enough to deserve its own post. However, the heart of the debate goes to how much humans should or should not be involved with the process.

When things aren't what they used to be 

This too is another complex issue and so I won't go into detail, but in essence, what happens to your model when something about the cafeteria itself changes? What do you do if the server changed and the serving line is now noticeably slower or faster than it was before? (In the case of hydrology, this might be like after a major wildfire has swept through an area and many trees are destroyed).

What do you do if your cafeteria started with one serving line and a second line was added? This doesn't even necessarily have to happen in reality, there can be a case where the modeller is just asking a "what if?" scenario. Or what would happen if the number of monks that visit the cafeteria doubled from what had ever been seen before? Would some unprecedented behavior arise, like how there could suddenly be a shortage of tables and monks would have to wait for a seat? (Again, in the case of hydrology, this could be the case of trying to guess how the river will react to climate change or some very heavy rainfall that had never occurred before).

If your model had an accurate description of how all the cafeteria processes behaved, it might give the correct answer when it is pushed outside the envelope of what has happened before. But if the model was just set up to fit to some historical data and all the bits inside were just guesswork, then trouble can happen.

A cafeteria is a fairly simple system and even then we can run into all these troubles when trying to model one. Apply this to a river running through an entire landscape that is complex and varied (and is varying in the future), and you can see how there is the major potential for things to go pear-shaped in a hurry. However, we still need models to give us guidance and they are still the primary tool for forecasting. As the famous saying from statistician George Box goes, "all models are wrong, but some are useful."

Friday, December 2, 2011

Modelling the Monks' Mess Hall (Part 1)

I asked National Weather Service Hydrologist Andy Wood what he would ask people from other countries to understand the role of humans in their process of making river forecasts. He wanted to know how the humans interact with their models. Do the humans do a lot of intervention, adjustment and interpretation (like cooking over a campfire)? Is it more hands-off (like a microwave oven)?

Models come in many forms, but basically they are quantitative summaries of how people believe nature works. They are the vessels that scientists pour their knowledge in to. Models play a central role in nearly all forecasting (and for a good chunk of scientific research as well) so it seems that a discussion of them here is inevitable.

The classic analogy for hydrology models is that soils hold and release water like a series of leaky buckets. But instead, let's start by discussing how to model a cafeteria/mess hall/canteen for hungry meditating monks.

During my 10 days of silent meditation, the meals were a highlight and I looked forward to going to the cafeteria each day. When all you have to do for hours on end is concentrate on the breath whistling past your nostrils, you can entertain yourself by conjuring back-stories for that hairline crack in the wall. Lets just say I had a lot of time to ponder the how the cafeteria worked. I am also convinced that two different people painted the cafeteria's walls, one Swiss, the other a drunk.
A typical lunch at Vipassana. The food was quite good, really. Day 6's tofu dish still gives me cravings and attachment.   
Let us imagine that monks arrive at a building (like precipitation falls on a catchment). They can all arrive at once (a big downpour) or leisurely drizzle in over some period. The monks disappear into the building for a while (like rainfall goes into the soils). Later we see monks leaving the building (much like how water would drain out to a stream). The challenge is that we cannot go inside the building, we can only make a model based on, say, walking around the building (like measuring soil properties) and observing monks going in and out (the equivalent of measuring rainfall and streamflow).


The eating area for females in the meditation center
One cafeteria-ologist standing outside the building could hear the cacophany of slurping and utensil clanging, along with the occasional full-throated belch (true story!) and guess that there is eating going on inside. He might guess that eating probably involves a set of stages, primarily queuing in line to get served and sitting down to eat.
Major processes in the cafeteria
What might be a good model for serving? Maybe monks can get their dishes and utensils and work their way through the serving line at the rate of about 5 monks per minute. If 25 monks queued up at once, it would take 5 minutes before the line was empty again. The rate people are served does not really depend on how many people are in line.
Oh yes. I really did spreadsheet this up. Here is the chart of monks going in (orange) and out (blue) of the serving line versus time. Here the monks arrive in two bunches, one at the start of the hour, and one about 10 minutes later. The black line shows how many monks are waiting to be served. Monks are served at a rate of about 5 per minute. 25 minutes into the lunch service all the monks are now seated and eating. 
After each monk gets his meal, he takes a seat and starts eating. Personally I am a slow eater and was nearly always the last person to finish my meal (after about 25 minutes). If I had to guess, I would say the average time to eat was about 15 minutes, but I also saw some people literally throwing food into their mouth. There is probably a theoretical maximum speed that someone could inhale their food. I think I've witnessed close to it.
How long it takes monks to eat meals. Some are slower eaters (right side) some are faster (left side) but most people are somewhere in the middle. 
Putting it all together.  In this case, monks arrive at the cafeteria (brown) in two bunches. The first monks start finishing about 10-20 minutes later. The rate that monks leave the cafeteria is shown in green-blue. Substitute "monks arriving" with "precipitation falling" and "monks finishing" with "river flowing" and you're looking at the kind of charts that many hydrologists deal with all the time (called hyetographs and hydrographs). 
This fairly simple model probably captures the main things that influence how long people stay in the mess hall. Similarly, the core of nearly all hydrology models are the same or similar to eachother. Modelers call things like queuing and eating "processes", and hydrologists generally agree on what the major processes are (e.g. evaporation, infiltration, and so on).

So where does it all fall down? Why are there dozens upon dozens of different models out there? And why are they such a challenge to use for real-world forecasting?

[Tune in to part 2!]

Thursday, December 1, 2011

Unexpected impacts from drought

Here's an interesting story in today's news from National Public Radio


45,000 Told To Evacuate German Town Before Massive WWII Bomb Is Defused

This upcoming weekend, a team of German explosive experts and members of the army are scheduled to defuse an unexploded bomb found in the city of Koblenz in Germany. The bomb — with 3,000 pounds of explosives — is a remnant of World War II that emerged in the Rhine River because of low water levels. How serious is the situation? Authorities ordered half of the city's residents — 45,000 people — to leave, while they get the job done.

Original caption: On Saturday and Sunday, officials destroyed the chemical barrels. Some 1,500 nearby residents were evacuated first. The incendiary bomb was likewise exploded.
This follows an earlier article in Der Spiegel with more information about bombs and other things getting exposed as the water level drops to near historic lows with the drought. 

Sinking water levels have exposed much more of the river bed than normal, and it turns out the bottom of one of Europe's busiest waterways is still littered with World War II leftovers. Over the weekend, passers-by alerted the authorities to the presence of one incendiary bomb near Cologne, two barrels in Koblenz filled with a chemical used during the war to create smoke screens and one large bomb, also in Koblenz, that is still lying in 40 centimeters (16 inches) of water, waiting to be defused....

The munitions are now coming to light due to an unusually dry autumn in Germany, resulting in the lowest November water levels seen along the Rhine in decades. With no rain in sight, it is likely to continue to drop. Already, many barges that travel up and down the river have had to lighten their loads to avoid grounding. In Bonn, the river was just 1.17 meters (3.9 feet) deep last Friday. "That is really very low for this time of year," a spokesman for the Duisburg-Rhine shipping authority told the Bonn daily General-Anzeiger.

Nepali Early Warning Systems: Practical Lessons from the Field

An earlier post mentioned charitable organization Practical Action's Flood Early Warning Activities in Nepal. Given that the winter holiday season is coming up, maybe you'd consider giving the gift of water-related relief? They even have gifts as cheap as Lollipops for Cows.

Anyone interested in how much of a trial-and-error process working in real-world communities can be should read their report "Early Warning Saving Lives", discussing how they set up Flood Early Warning Systems here in Nepal. It gives a good assessment at each stage of the process of what worked and what didn't.

Local people were familiar with the idea of early warning for dangerous wild animals based on tall lookout-towers. They built on this by erecting flood watch towers that included sirens.

Flood watch and warning tower in Nepal.
As is typical, things don't always work as well as they do in theory. The sirens have an advertised range, say of a few kilometers. However, this isn't the actual range of how far the sound travels when the rain is coming down in buckets... Also, it is hard to hear the sirens when rain is falling noisily on metal roofs, as are common in Nepal.

But is the answer a bigger siren, requiring more electricity? If batteries can't give enough electricity, the sirens have to be connected to the power grid at a local house. Also electricity isn't necessarily reliable here, especially during flood disasters. Is it better to have many smaller sirens than fewer bigger ones?

How about hand-cranked sirens? But what happens when teenage boys get up on the tower and give the crank some turns out of boredom? It seems best to keep the handle separate from the siren itself. But you don't want to lose your only handle, so it's best to have a few handles around. And so on and so on...





A common challenge was that of knowing travel times of the river. Namely, another village 20 km upstream could call and warn that a flood was coming, but how long would it be until it arrived? In some countries there would be historical records of streamflow to base these calculations on. Instead, the community suggested throwing a soccer ball or bottle into the river upstream and see how long it takes to get downstream.

There is mention in Practical Action's document how a soccer ball would "spin" as it rolled along the surface of the river and so it would travel slower than the actual flood wave. Writing out the actual physics of how that works seems to me like it would be a good hydrology graduate course test question. Feel free to show your work in the comments section below. My solution would be to use a sealed, half empty bottle so it's partially weighted down.

Thanks to Anup Phaiju (right) for talking with us about Practical Action's projects
Another thing that impressed me was the cost of data in Nepal. The government pays people that measure rainfall 20 rupees ($0.25 USD) per observation. They typically measure three times a day and the community was trying to raise the funds to increase that to eight measurements per day during the monsoon. In comparison, automated computerized gauges cost about $2000 USD. Basically, it costs the same to install a single automated gauge as it does to pay a human to measure rainfall three times a day for thirty years.

Practical Action also has a lot of other useful and readable documents on their webpage related to climate change and disaster reduction in Nepal.