China: Highlights from Some of the Debris-Flow Workshop Presentations

The International Debris-Flow Workshop in Chengdu, China (August 2012) included about 40 speakers from China and several other countries presenting the results of their research.  Included were papers describing debris flows and their consequences, modeling studies, structural measures to control debris flows, and risk assessments.  Following are some items from 12 papers that I found particularly interesting:

Landslide Prevention in Hong Kong (Professor C.F. Lee, University of Hong Kong) 

As I pointed out in a recent blog post, much of the territory of Hong Kong is steeply sloping.  In fact, approximately 2/3 of Hong Kong is hilly or mountainous.  This limits the available area for new building construction for the Hong Kong Special Administrative Region with its population of 7 million.  In the past, Hong Kong was able to expand its area by pushing fill material into the harbour, but this is no longer allowed.

Thus, Hong Kong grows vertically replacing buildings of modest height with taller and taller structures.  Many high rises have been located adjacent to steep slopes.  Building codes may protect these buildings from the effects of earthquake shaking but not from large scale slope failures that frequently occur in response to heavy tropical rainfall (3000mm annually – nearly 120 inches).  These slope failures are complicated by the local geology.  Chemical weathering decomposes granite and rhyolite (tuff) bedrock into fine grained material.  The sheer strength of this material is greatly reduced by rainfall wetting, and the factor of safety for slopes decreases with multiple storms.       

When slope failures occur adjacent to developed areas, the results can be deadly.  Professor Lee discussed the Po Shan Road Landslide of June 1972 (during a period of heavy rains) when 20,000 cubic meters of material enveloped high rise buildings resulting in 67 fatalities. 

In 1994, a landslide was caused by a cracked water pipe which resulted in saturated soil.  As a result, Hong Kong has subsequently changed out old water pipes around steep slopes.

Over the past 20 years, slope stabilization has greatly reduced both fatalities and damage to structures. Hong Kong engineers have been able to successfully stabilize precarious slopes in populated areas using rock bolts on bedrock slopes, shotcrete to prevent rainfall infiltration, and grid retaining walls.  However, these techniques often produce visually unattractive results.  In recent years, soil nails are more commonly being used in both cut and fill slopes.  This method enables revegetation of slopes creating a more aesthetic appearance.



While visiting Hong Kong before the Debris Flow workshop, I saw some examples

of high rise apartments on steep slopes that Professor Lee spoke about.

Soil nails have been effective in stabilizing slopes in populated areas of Hong Kong.

Source: http://www.hbfleming.com/images/work/soil/nail_drawing.jpg

 

 

 

 

 

 

 

 

 

 

 

 

Structural Measures Against Debris Flows in Japan (Professor Takahisa Mizuyama, Kyoto University)

One of the more common structural measures to counter debris flows is the sabo dam, a large steel structure which is located in the center of a channel which is at risk of debris flows.  These are located perpendicular to the direction of the debris flow and have rock/concrete retaining walls on either side.  Typically, a series of sabo dams will be located along the channel.  The dams at the upper end will have large openings and will capture only large rocks and debris enabling water and smaller diameter rocks and debris to pass through.  Dams with successively smaller openings will be located downstream.  By the time the debris flow reaches the lowest dam, most of the larger materials will be filtered out and some of the silty water will have been retained behind each dam as the openings become clogged with large debris.  The volume and velocity of the water passing through the small openings in the final dam will have been considerably diminished.   

Example of a sabo dam photo from Professor Mizuyama’s presentation.

Following the debris flow, large material and sediment captured by each sabo dam must be cleaned out using heavy earthmoving equipment and teams of laborers.  

Another common structural measure in use in Japan is the wire ring net dam which consists of heavy steel cables woven together with large openings.  A shortcoming of these nets is the difficulty cleaning them out after a debris flow.  As a result, they typically must be replaced.

There is always the possibility that a large volume of water will be held behind sabo dam debris.  This can result in the sabo dam being overtopped by water or causing it to fail.  Dr. Mizuyama is proposing a modification of the sabo dam which he calls a debris-flow fence.  It allows water to more easily pass through the structure creating a more even flow downstream.

Equivalency principle: bed structures and bed load motion (Zhaoyin Wang, Tsinghua University)

Dr. Wang talked about the importance of the step-pool sequence in streams [sometimes called pools and riffles by fisheries biologists in the USA] to maintain stable habitats and beautiful landscapes.  Bed structures, whether natural or artificial, dissipate energy and reduce erosion.  They reduce bed load motion during floods.  Destruction of bed structures increases sediment transport of streams. 

Wang and his colleagues studied sections of several river systems in China.  In the Yalu Tsangpo (Grand Canyon) in Tibet, they found that bed structures consume most of the flow energy.  There is no bed load transport as a result.  

Artificial bed structures can mitigate the effects of debris flows by slowing their velocity.  In addition, placement of bank stones creates high resistance to flow and protects banks from failure. 

Construction of dams destroys the step-pool system at a cost of many times that of step-pool system construction.  Dams often fail under the tremendous stresses placed on them by debris flows and failure of unarmored banks adds to the debris flow volume.   

Professor Zhoayin Wang of Tsinghua University (China)
spoke about technical aspects of bed load motion in rivers.

Formation Mechanism and Risk Control of Catastrophic Debris Flow Hazards (Peng Cui, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences or CAS)
In 2010, Debris flow events in China were 10 times the normal average and resulted in 2909 deaths.  This large number was primarily the result of the Wenchuan Earthquake.  However, there are a number of factors contributing to the number, size, and destructiveness of debris flows:
- Climate change has resulted in more unusual weather events and intense rainfall.
- Glaciers in Tibet are melting therefore enlarging lakes and enlarging river flows.      
- A strong earthquake results in unstable slopes.
- Cascading dam failures enlarge debris flows by providing more water and materials. 
- Urban development occupies floodplains reducing the area available to accommodate floods.

Following this disastrous year of debris flows, the Institute of Mountain Hazards and Environment initiated a research project to look closely at the debris flows and flash floods that occurred.  They also performed risk assessments evaluating predicted locations of future events and their potential impacts.

Debris flows in China are serious business.  Note that one debris flow

in Gansu province in 2010 caused 1765 deaths. (slide from Dr. Peng Cui’s presentation)

Empirical Relationships for Post-quake Debris Flows in Wenchuan Earthquake Zone (Kaiheng Hu, Institute of Mountain Hazards and Environment, CAS)
Debris flows can last from a few minutes to several hours.  Their magnitudes relate to the size of the supply.  There are two types of debris flow basins:  rainfall-dominated and supply-dominated.

In the USA, debris flows are often associated with heavy rains after forest fires and are not earthquake related.

Slide from Dr. Kaiheng Hu’s presentation shows the two types of debris flow basins.

 

Crisis on Flash Flood and Landslide Disasters of Thailand in 2011, (Chinapatana Sukvibool, Land Development Department, Thailand)
Landslides in Thailand have increased sharply since 1970.  In 2011, disastrous landslides, debris-flows, and flash floods occurred in river basins which train south to Bangkok.  Huge floods in the Bangkok region inundated large areas and even flooded the airport.  A total of 4.4 million hectares (11 million acres) in Thailand were flooded in 2011 resulting in 65 of Thailand’s 77 provinces being declared disaster zones.

Thailand is adopting strategies for flash flood and landslide mitigation on both a macro and micro scale:

- conservation, reforestation, terracing, and check dams in uplands
- farm ponds, reservoirs, contour barriers, and vertical drainage ditches in lowlands
- use of BMPs (best management practices) such as gabions, wattles (erosion control logs), erosion control blankets, and vetiver grass

2370 Thai villages are at risk of flash floods, debris flows, and landslides but only 404 villages have early warning systems. 

Chinapatana Sukvibool of the Thai Land Development Department

discussed last year’s flash flood and landslide disasters in Thailand.

Debris Flow in a Metropolitan Area - 2011 Seoul Debris Flow (Chan Young Yune, Gangneung-Wonju National University, Korea)

South Korea experienced two large typhoons in the summer of 2011 with intervening large rain events and record precipitation for the year.

In July, following a day of 359mm rain (14 inches), a series of debris flows occurred simultaneously in all directions from Umyeon Mountain located on the south side of Seoul (South Korea’s capital and largest city).  It affected the city center and neighborhoods where many wealthy Koreans live. 

Flow velocities reached 28m/second (almost 60 mph) yet slopes averaged only 13 degrees.  These high velocities can be accounted for because of the lubricating effect of the wetted sediments.

Professor Yune and his colleagues have been conducting a real scale debris flow experiment in the field to better understand local debris flows and their predictability.



Professor Chan Young Yune spoke on debris flows

that occurred in the Seoul area in 2011.

 

Engineering Technology Against Large Debris Flow in Wenchuan Earthquake-hit Area (Xiaoqing Chen, Institute of Mountain Hazards and Environment, CAS)

Dr. Xiaoqing Chen pointed out there has been less research in engineering methods to prevent disasters than post-disaster studies.  He discussed and illustrated the different types of dams used to control debris-flows.  While dams may not stop debris flows, they can control their magnitude and prevent blockage of rivers and the subsequent flash floods when debris flow dams are breeched. 

Example of a frame check dam used to control debris flows.

(from Xiaoqing Chen’s presentation)

I was astounded to learn that one debris-flow had a measured thickness of 50 to 80 meters (160-260 feet)!  During our field trip a few days later, I saw evidence of a flow that reached the fourth floor of a building.

Dr. Xiaoqing Chen made a passing reference to “ecological methods” for debris-flow control.  Because most of the speakers had focused on structural methods, I asked him about “ecological methods” during the Q&A period following his presentation.  Did they plant local (endemic) species, I wondered.  He assured me they did.

Mitigation Measures Against Debris Flow in the Giampilieri Area after the Event of October 2009 (Marco Deana, Geobrugg Italia)

On October 1, 2009, 300mm (12 inches) of rain fell in five hours in the northeast corner of the Italian island of Sicily.  This followed another heavy rain a few days earlier.  The vulnerability of the steep slopes above the town of Giampilieri was compounded by thin soils over metamorphic bedrock.  In addition, orange groves used to cover the hillsides and helped protect the soil but the groves have been abandoned and many trees have died from neglect.     

The debris flow that occurred on October 1, killed 27 people and destroyed the town’s only access road.  Geogrugg Italia was called upon to recommend mitigation measures to protect against future debris-flows.   The firm executed a plan to protect against 100 year events (greater protection would have been prohibitively expensive).  The recommended measures included: 

-soil nailing
-superficial landslide barriers (steel netting over exposed area)
-debris flow barriers with dimensions based on testing

Each debris flow barriers only took one day to install.  During a subsequent smaller storm, one barrier worked well in containing a debris-flow.

Marco Deana described mitigation measures taken
in the wake of a deadly debris flow in Sicily.

 

High Mobility of Landslides in Volcanic Deposits Triggered by the Great East Japan Earthquake (Fei Cai, Gunma University, Japan)

The Great East Japan Earthquake of 2011 and resulting tsunami was reported extensively in the Western press.  At one location, the tsunami run-up height reached 43.3 meters!  Less well known are the landslides that were triggered by the earthquake particularly in areas of volcanic ash soils.  For example, the Hanokidaira Landslide killed 12 people and destroyed 10 houses below the base of a slope with a clay soil layer between ash layers.  With water concentrating in porous pumice or ash layers just above a low permeability clay layer, an ideal environment is provided for landslides. 

Another example of a catastrophic slide under similar conditions occurred in Las Colinas, El Salvador in 2001.  In this case, an earthquake triggered a landslide in volcanic tephra which killed 586 people.  

Dr. Cai’s team conducted tests on the sheer strength of pumice and clayey soil.  Pumice was found to have a low sheer strength while clay had a high sheer strength.  They concluded that there is a high mobility of landslides in volcanic deposits.  The apparent friction coefficient (ratio of height to the length of run out) is lower than for slides in other types of materials.  It is obviously important to remember this when planning for disaster prevention.

Some grim statistics from last year’s Great East Japan Earthquake.
(slide from Fei Cai’s presentation)

 Initiation Conditions for Rainfall-Introduced Debris Flows Generated by Surface Water Runoff in Landslide Deposits (Jiawen Zhou, Sichuan University)
This presentation looked at the circumstances leading to the destructive 2010 debris flows in the Wenjiagou gully in Szechuan Province and their characteristics. 

The bedrock in this location is mostly weathered limestone and dolomitic limestone. The debris-flows occurred in landslide deposits with thickness of as much as 150 meters.  These deposits were weakly consolidated with wide size distributions.    

Heavy rainfall of 227mm (9 inches) in a short period of time provided ideal conditions for the triggering of debris flows particularly where slopes exceeded 20 degrees in narrow gullies.  Fine particles in the landslide deposits were easily carried by surface water flow which developed into a high-mobility debris flow which was capable of carrying large boulders. 

Changes in Runout Distances of Debris Flows over Time in the Wenchuan Earthquake Zone (Shuai Zhang, Hong Kong University of Science & Technology)

This study looked at methods for quantifying debris flow hazards.  It involved use of remote sensing images with a GIS platform plus field investigations.  It also considered factors such as the probability of people being in an area where a debris flow even occurs and the number of vehicles per second crossing the area.

Dr. Shuai Zhang considered rainfall intensity and duration related to the occurrence and non-occurrence of debris flows at specific locations in order to come up with probability of a debris flow occurring.

UPDATE:  I incorrectly identified the above speaker as Dr. Shuai Zhang.  I have since learned that she is Dr. Hua-li Pan from the Institute of Mountain Hazards and Environment in Chengdu.  She presented a paper titled, The Mechanism of Debris Flow Downcutting Erosion for Movable Bed and Its Critical Conditions.  In her paper, Dr. Hua-li Pan pointed out that debris flows are more erosive than water or even sediment-laden water.