Don’t Let Heavy Rain Turn the City Upside Down

Story of Zhaoyang

This happened on August 7, 2015, in Beijing.

That day, my wife and I went to discuss our wedding banquet.

On the way home, it started to hail—and then it poured.

We walked about four kilometers through the storm.

When we finally reached home, the entire area around our building was flooded.

Let’s turn back the clock.

On July 21, 2012, there was another massive rainstorm in Beijing.

It poured for more than ten hours, dropping over 170 millimeters of rain.

To give you a sense of that, Beijing’s total annual rainfall is about 600 millimeters.

That storm was the strongest in sixty-one years.

The whole city’s traffic came to a halt.

Afterward, Beijing built eighteen retention basins and two pumping stations between the Second Ring Road and the Third Ring Road.

Then, on July 18, 2018, another storm flooded the Huilongguan area.

And in 2007, in Jinan, one rainstorm dumped 151 millimeters of water in just an hour or two.

The streets became rushing rivers; many cars were swept away, and some people lost their lives.

That’s our reality:

Every summer, we wonder, Will the city flood again?

Every year, we face the same problem.

But besides the floods we see, rainwater causes many other issues.

In one of our projects, for example, there was a farmers’ market above the drainage pipes.

When it rained, all the filth from the market washed into the pipes,

and the next time it rained, it all came back out again.

This is one reason why our rivers remain polluted—

rainwater itself becomes a carrier of pollution.

Take the Nanming River in Guiyang, where my teacher took some photos.

The river used to be crystal clear, the water quality excellent.

People fished, people swam.

But after one heavy rain, all the sewage and runoff poured out from one outfall,

and the whole river turned foul.

I’ve worked in urban drainage for about ten years.

My name is Zhao Yang, and today I want to talk about what’s gone wrong with drainage in our cities.

First, I want to explain how the public sees drainage,

and how we engineers see it—because they’re quite different.

When most people think of drainage, they picture those huge underground pipes.

In our project diagrams, every yellow or blue line represents a pipe segment,

and every dot marks an outlet.

Then there’s the thing everyone recognizes:

the grated inlets on the street, like mouths that collect rainwater.

Water enters through those grates, flows into the manholes,

and from there into the network of pipes.

In Japan, you can still find open channel systems—

but in major Chinese cities, those are already rare.

And of course, rivers themselves are part of the drainage system.

Eventually, all the water from pipes and channels flows into rivers,

then rivers into larger rivers, and finally into the sea.

There are also facilities people often overlook,

or don’t realize belong to urban drainage.

For example, in Berlin, in one public square,

the pavement looks ordinary—but between each brick are raised narrow strips,

so the bricks never fit tightly together.

When it rains, water seeps through the gaps into the soil below,

maximizing infiltration.

That’s also an important type of rainwater facility.

In the Netherlands, parking lots are designed with sunken strips along the lanes.

They deliberately slope the pavement

so that runoff gathers in those shallow depressions and then enters inlets.

There’s another type that looks like a garden.

If the garden in front of your house floods,

you might think of calling property management to complain.

But in fact, it’s designed to flood.

We call it a rain garden or grassed swale—

a lowered patch of greenery that collects runoff before draining it away.

Now let’s talk about how we classify rainfall scientifically.

We use what sounds like a very “professional” system:

light rain, moderate rain, heavy rain, and torrential rain.

This classification actually makes sense,

because of an important concept: the return period.

We record every rain event over time.

If a storm occurs on average once every year,

we call its return period “1.”

If it happens once every three years, its return period is “3.”

If once every ten or twenty years, then “10” or “20.”

All those facilities we just discussed are designed to handle different return periods.

Flooding problems usually occur during heavy or torrential rains,

rarely during small showers.

If your street floods during light rain, that’s when you should call to complain.

But for most cities, flooding only happens in the big storms.

Now let’s talk specifically about urban waterlogging—

and how it differs in China compared with other countries.

In 1987, we had a national Outdoor Drainage Standard.

At that time, the minimum allowed design standard

was just 0.3 years of return period—

meaning the pipes under your house might not even handle a one-year storm.

That was mainly because the country had limited funds,

and building larger pipes costs money.

Today, we’ve upgraded to two- to five-year return periods.

So a two- or five-year rain can be fully drained through the system,

and key areas may be designed for ten- or twenty-year events.

That’s roughly on par with Western standards.

So, if our design is comparable,

why do our cities still flood so often?

The real issue lies in maintenance, management, and operation.

Back to that “romantic” night when my wife and I walked home in the storm:

When we reached our building, sewage was overflowing, trash everywhere,

and the wind was howling.

The next day, the sanitation workers cleaned it up—

and all that garbage had come from just one tiny rainwater inlet.

No matter how large my pipe design was,

if half of it was blocked by garbage, how well could it work?

We even sent a small robot into the pipes to check inside.

In a one-meter-wide pipe,

the camera could barely move—the debris blocked it so badly.

Half the cross-section was filled with sludge.

When it rains, either that muck gets flushed downstream,

or the whole pipe simply stops working.

Under that farmers’ market I mentioned earlier,

the same thing happened:

sediment and blockage shrank wide inlets into tiny holes.

Concrete pipes corroded like blood vessels.

Construction waste pierced through and damaged the walls.

With pipes in such a condition,

no matter how high the design standard is,

it only performs at that level the day it’s built.

That’s why maintenance—steady, detailed maintenance—matters most.

Let’s look abroad.

First, the United States.

In general, their management is about the same as ours—

they also have clogged inlets.

But because their land area is larger and population sparser,

they can manage more carefully.

One simple method they use is to cover the inlets with fabric perforated by small holes—

trash stays on top, water passes through.

Now, our neighbor Japan.

In Osaka’s Namba district, there’s a large shopping complex

with an award-winning rooftop green space.

It’s filled with plants,

and every time the leaves are swept up,

where do they go when it rains?

Along the edge of the roof runs a narrow gutter.

Water drains through it, while the grate outside traps the leaves.

Below that is a fine grid filter,

and workers clean it regularly—

just like the drain in your bathroom.

The leaves and debris are stopped outside;

only clean water enters the system.

We also visited Okayama Korakuen, a famous classical garden.

Its pathways are paved with gravel and sand.

We wondered—when it rains, how do they keep all those stones clean?

Inside each drain is an iron bucket with holes at the bottom.

The stones stay inside the bucket.

When it’s full, they lift it out, dump the stones, wash them,

and put it back—simple and efficient.

In Kyoto, we saw many open drainage channels along the streets—

common in Japan, rare now in China.

Street water flows into the inlet and then into these shallow ditches.

A simple barrier board holds trash in place until it’s cleared.

At points where the street crosses the channel,

metal grates are installed,

and the workers clean them slowly and methodically.

Even the wooden sticks that look like trash

are actually used to guide the flow of water.

Now, about manhole covers.

People often admire how beautiful Russian or Japanese covers are,

but during our trip to Tokyo, we noticed they carry crucial information.

Each cover is marked with numbers identifying the specific manhole.

The last digits show the year of installation—say, 1973.

Every maintenance or repair job is logged in their database.

After each replacement, they verify and record the data again.

Seeing that, we felt we had much to learn.

Unfortunately, no one in our company speaks Japanese,

so we bought illustrated manuals about how to clean sewer pipes.

They explain exactly how to maintain pipes of every diameter—

400 millimeters, 800, 2000 —

and even where to block the upstream manhole before cleaning,

so the inflowing water doesn’t ruin your work.

People often say “the sewer system is the conscience of a city.”

Many governments are eager to spend on new sewers.

But what’s most important now is not building blindly,

but evaluating which segments need repair or upkeep.

Our basic job is to open manholes section by section,

send in machines to inspect, diagnose the problems,

then decide whether to fix, replace, or rebuild that stretch entirely.

So far, we’ve talked about storms with return periods of one to ten years—

the events the pipe network mainly handles.

Now let’s look at extreme storms,

those that happen once in twenty to a hundred years.

In terms of infrastructure, China has made progress,

but for these massive storms,

the gap with international practice remains clear.

Here’s one case:

After a typhoon, a landscape firm called SASAKI redesigned Boston’s master plan.

They projected that within a hundred years,

sea levels would rise and storms would grow more intense.

The city, under the twin pressure of tides and heavy rain,

could no longer function as it does now.

So they mapped which areas would stay safe,

and which would need relocation.

Their model covered two time frames—2050 and 2100—

overlaying storm surges and rainfall

to produce a flood-risk map.

Blue zones show where flooding will occur.

Critical facilities—power plants, transport hubs—

must not be built there.

Populations will gradually shift elsewhere.

They manage their cities using these maps—

deciding where to build, where not to.

We’re very familiar with this logic:

high ground rarely floods.

Every flood season, social media fills with headlines like

“The Forbidden City Hasn’t Flooded in 600 Years”

or

“Qingdao’s German Sewers Show True Craftsmanship.”

But we need to be objective.

The main reason the Forbidden City never floods

is simply that it was built on higher ground.

Its water drains into the moat.

That moat was dug precisely for that purpose.

Even the nearby Shichahai Parks and canals are interconnected

to carry the water away.

People marvel at the dragon-shaped spouts on the palace walls,

but the real secret is the multiple surface flow paths that guide rain away.

Today’s Beijing is far larger.

Where can all that water go?

As the city expanded, rivers narrowed;

we lost much of that natural space.

Does that mean we should stop building?

Of course not.

In Denver, USA, for instance,

they base their planning on flood-risk maps too.

Gray zones mark areas submerged only during extreme storms,

pink zones show those hit by twenty- or fifty-year events,

and blue lines are the protected drainage corridors

that must remain open for water to pass.

In the middle of one such map is a sports stadium—

and yes, it’s meant to be flooded during heavy rain.

The flood-flow paths correspond almost exactly with the city’s green spaces.

Where they diverge—in developed districts—

the planners maintain slopes that still allow water to pass through.

Flood-prone gray zones aren’t left unused.

They’re developed as industrial areas, wetlands, or parks—

places that can safely get wet.

That same stadium’s parking lot, for example,

is designed to flood temporarily,

while the stadium’s main structure sits well above ground level.

Between those green zones,

Denver roads are tilted toward the center

and dotted with holes,

giving runoff an easy route to travel.

Japan, on the other hand, is densely built and lacks open space.

So they carefully map flood depths—

which areas might see two meters of water,

which only half a meter—

and plan evacuation and emergency routes accordingly.

They often leave the ground floor of buildings open

as temporary storage for floodwater.

Along Myōshōji River, for instance,

every few hundred meters stands a large retention basin.

When water rushes in,

these basins prevent too much from reaching downstream Tokyo.

Normally, the ground level above them serves as recreation space—

kids playing soccer, people gathering—

but during a storm, it floods completely.

Such designs are common in Japan.

Back in China, more than a decade ago,

my teacher worked on a similar multi-function retention park in Yizhuang.

The area had once been a quarry,

leaving a giant pit.

We turned it into a park.

All surrounding runoff drained into it.

The lowest layer held regular water;

as rainfall increased, the flooded area expanded.

A fifty- or hundred-year storm could all be stored there

without affecting downstream neighborhoods.

We did the best we could at the time,

but if we’d made it an ecological wetland

where children could play and birds could nest,

it would have been even better.

Still, during the July 21, 2012 storm,

the park flooded completely—

yet the surrounding areas stayed safe.

That’s the key to handling heavy rain:

facilities for rare fifty-year events

should also serve daily functions,

so they remain a sensible investment.

Now, after flood control, let’s talk about runoff pollution.

It comes mainly from two sources.

First, the rain itself.

Rain seems clean, but the ground isn’t—

so once water flows across pavement,

it picks up trash, gravel, leaves, oil,

and rushes straight into rivers.

The total annual pollution load is enormous.

In some Beijing rivers,

runoff pollution exceeds even that from leaking sewage.

Another concept is the combined sewer system.

In the past, when budgets were tight,

we built single large sewers

to carry both rainwater and sewage.

During dry weather, the small flow went to treatment plants.

But in heavy rain, the overflow discharged directly to rivers—

causing huge spikes of pollution.

How do we control runoff then?

The simplest way is to keep rain out of the pipes in the first place.

We build green spaces—like inverted lids—

so water first filters through soil and vegetation before draining away.

It’s a basic and effective approach,

though it depends on having available land.

In cities with little space—like Guangzhou, where we are now—

there isn’t enough greenery to lower.

So we build retention tanks,

holding the water temporarily

before sending it slowly to treatment plants.

For years we assumed separate systems were always better,

so we tried to replace the old combined ones in many cities—

Nanjing, Shenzhen, and others.

But in truth, no city is completely without combined sewers.

We have to acknowledge that reality.

In 2008, New York still had 65 percent combined sewer areas;

Tokyo had 82 percent.

Once we accept that,

the solution is either storage basins or surface green infrastructure.

We mustrecognize that fully separating stormwater and sewage is not easy.

It takes space, money, and years of coordinated construction.

Now let’s talk about small and moderate rainfall,

which, surprisingly, has been the biggest blind spot in our drainage thinking.

We tend to believe that if the city can handle a heavy storm, everything’s fine.

But in fact, most of the total annual pollution load

comes from small and moderate rains—

they account for 80 to 90 percent of all rainfall events,

while truly large storms are rare.

China didn’t begin seriously addressing this category until 2014,

when it was finally brought into the field of drainage engineering.

The United States had started this work back in 1972.

At first, their water-quality regulations

required control measures only for developments

larger than five acres.

Later, they tightened the rule to one acre.

And now, in New Zealand,

even the construction of two parking spaces

requires runoff-control facilities.

Small, yes—but extremely important.

In the past, both the U.S. and China relied mainly on large centralized ponds,

like the multi-purpose retention park I mentioned in Yizhuang.

But later, the U.S. began to decentralize—

building small collection spaces throughout every street and neighborhood,

letting water infiltrate right where it falls.

We’re now doing the same.

This is a resettlement community we first visited in its original state:

paved over, with no green infiltration areas at all.

We decided it needed to change.

After redesign, we added a sunken central area—a rain garden—

so all the runoff flowed there.

We replaced the surface paving,

planted flowers and grass,

and turned the space into a living landscape.

The next year, though, we discovered something funny.

Our residents had started using the rain garden to hang laundry.

The trees we planted were spaced just right

for pulling a rope between them to dry blankets.

Some even cleaned fish next to the drains—very convenient, they said.

At first, we thought this was discouraging.

But by the third year, we saw something else:

the rain gardens built in the wrong locations had vanished,

while the ones placed correctly had stayed.

People actually liked using them.

Gradually, the whole neighborhood changed.

It wasn’t the ideal picture we’d imagined,

but it worked,

and our facilities and the community slowly merged into one.

After talking about flooding and pollution,

I want to stress something essential:

these two problems are actually the same at their core.

Our urban development has broken the natural hydrological cycle.

As more and more surfaces are hardened with concrete and asphalt,

less water can infiltrate into the ground.

The city loses its natural absorption capacity,

and runoff simply rushes away.

The ground conditions change,

and so, inevitably, does the entire water cycle.

I’m in my thirties now,

and I’ve been working in drainage for about ten years.

There’s no secret to it—

just one pipe at a time, one neighborhood at a time,

step by steady step.

Through these years, I’ve learned that we must face facts honestly:

acknowledge our real conditions, our gaps, our strengths and weaknesses.

We must admit that many areas still use combined sewers,

that operation and maintenance matter more than grand designs,

and that we still have much to improve.

Only when we accept these realities

can we stop heavy rain from turning our cities upside down.

Thank you.