Experience and Lessons Learned in utilizing geo-referenced - - PDF document

Experience and Lessons Learned in utilizing geo-referenced information in DRM

Country Presentation Sri Lanka

Kumari M . Weerasinghe Senior Scientist National Building Research Organisation Sri Lanka

§

Floods

§

Landslides and mass movement s

§

Drought

§

Cyclones

§

Coastal erosion

§

Subsidence and erosion

§

Ground settlements

§

Tsunami

§

Earthquake

NATURAL HAZARDS IN SRILANKA

Geo-referenced information are utilized in various ways for disaster Risk reduction and sustainable development in the country

• Studying the distribution of landslide potential in the central

highlands of Sri Lanka and landslide hazard zonation mapping

• Identify the areas subjected to flooding and preparation of

flood inundation maps

• Study and map the areas subjected to

– costal erosion – Sea level rise – Tsunami inundation

• Predict and map the areas subjected to agricultural drought
• M apping cyclone paths and areas subjected to wind hazard
• and many more

Among the International Cooperation received over the years…

• Sri Lanka was selected as a Pilot Country for implementation
• f the 5 year Action Plan (2012-2017) formulated by the

UNESCAP for developing the capacity of the countries in the region to use space technology applications for Sustainable Development and Disaster Risk Reduction.

• Sri Lanka was also selected as one of the first two pilot

countries to implement the UNESCAP project on Regional Cooperative M echanism for Drought M onitoring and Early Warning .

As a result, S ri Lanka was able to participate in various capacity building programmes organized under the blessings of UNESCAP

• Use of M ulti GNSS information in Disaster Risk Reduction, held in

Bangkok, Thailand 2012

• Sub-regional training on “ Development of geo referenced information

system for disaster risk management held in Dehradun, India from 26th – 29th August, 2013

• Capacity building programme on space technology and GIS application for

disaster risk reduction, held in Seoul, Korea from 28th October to 17th November, 2013.

Availability of geo-referenced information

• National hazard profile (www.hazard.lk)
• National Risk profile (www.riskinfo.gov.lk)
• Landslide hazard zonation maps (www.nbro.gov.lk)

Cabinet decision has been taken for Development of National Spatial Data sharing Infrastructure (NSDI) The thematic areas of Disaster M anagement and Environmental M anagement are selected for the pilot project Drafting the spatial data infrastructure road map for the thematic area DM was commenced in mid M ay, 2014 and is expected to be completed by the end of July 2014. Testing the road map as a model, fine-tuning it, and using as a basis for building the NSDI will be implemented subsequently.

Visualizing NSDI

SLS DI is viewed as the technology, policies, standards, human resources and related activities necessary to collaborate, acquire, process, distribute, use, maintain and preserve spatial data throughout all levels of government, private and and citizens for efficient service delivery

Steps towards a National SDI

Utilizing geo-referenced information on landslides disaster risk reduction and Experience and Lessons Learned

In the past majority of landslides occurred due to natural causes In 1986, a comprehensive study on landslides was implemented by the government In 1990, 5 year long Landslide Hazard Zonation M apping project was implemented with the technical and financial support of UNDP for Identification of landslide potential and landslide disaster risk reduction At the inception, M aps were intended to be used as a tool for development planning

Field verification and additional data collection Transferring data to base maps

Integration of weighted maps and preparation of hazard zonation map Preliminary hazard zonation map Calibration with actual landslide map and field checking

(Aerial Photo / ABMP maps and geology maps)

• Landslide distribution map
• Engineering Geological map
• Hydrology map
• Material distribution map

(overburden deposits)

• Geomorphological map

(landform map)

• Landuse and management map

Digitizing of attribute maps

Statistical analysis to evaluate and assign weights

Research on further improvements

• zonation techniques
• integration of run-out zone of

landslides

• Map presentation methods
• Data base developments

Preparation of Landslide Hazard zonation maps

Cartographic preparation

Final Landslide Hazard Zonation M ap Final Landslide Hazard Zonation M ap www.nbro.gov.lk

Recommended Land Use Practices for Different Landslide Hazard Zones

Landslides not likely to occur (expect for flood plains) Landslides can be expected Landslides most likely to occur

Degree of Reliability of Predicted Landslide Hazard

1:50000 scale maps are published in www.nbro.gov.lk 1:10000 scale maps are issued on request M aps are being utilized by both government and private sector organizations for disaster impact assessments and development planning Accuracy levels were considered as sufficient for originally intended use of the hazard maps

RESEARCH ON APPLICATION OF DETERMINISTIC SLOPE STABILITY EVALUATION TOOL - SINMAP

Output of the participation in Mini Project 2006/ 2007 at AIT

Total Statistics

Stable Moderately Stable Quasi Stable Lower Threshold Upper Threshold Defended Total Area (km2) 89.30 10.20 13.50 28.30 9.60 2.00 153.70 % Area 58.10% 6.64% 8.78% 18.41% 6.25% 1.30% 100.00% No of Landslides 6 2 8 7 6 29 % Landslides 20.69% 0.00% 6.90% 27.59% 24.14% 20.69% 100.00% Landslide Density 0.07 0.00 0.15 0.28 0.73 3.00 0.19

Indicate instability Indicate marginal instability

72.41%

Predicted reliability

Issues while utilizing landslide hazard zonation maps for development planning

• High hazardous zones are already occupied
• Growth of population
• Availability of suitable land is limited for relocation and also

for future development

• Sociological and political issues related to relocation

Based on the lessons learned, national approach on landslides disaster risk reduction has been reformulated

Awareness among communities on living with landslides

Structural and/or non structural Mitigation of landslide risk

landslide early warning

Landslide hazard zonation mapping NBRO DOM EW Model

Landslide early warning message is issued by NBRO EW is disseminated through DMC and Media

Step 1 - Issuance of Landslide early warning at national level

75 mm/ day 100mm/ day 75 mm/ hour or 120mm/ day

National level landslide warning is issued at three levels

Landslide watch Landslide Alert Be Prepared for Evacuation

Approximate rainfall threshold values

Increasing trend of manmade landslides

Awareness among communities on living with landslides

Structural and/or non structural Mitigation of landslide risk landslide early warning

Step 2 - Government’s intervention by enactment of new regulation in the year 2011

Landslide hazard zonation mapping

Obtaining landslide clearance prior to approval of all construction in landslide prone areas Government’s regulation Identification of landslide hazard potential Performing site specific investigations Implement const ruction by adhering t o the sit e specific recommendations and guidelines Builder NBRO

With the new regulation, demand for landslide hazard maps was increased General public also started to utilize the maps Problems were faced in interpretation of hazard potential by the non technical map users

1. Resolution of the map (presently available at 10 m and 50 m accuracy ) was not sufficient in borderline cases 2. Difficulty in demarcating hazard zonal boundaries

• n the actual ground, if printed maps are used

3. Lack of access to and/ or knowledge in using GIS software for using the maps in digital form 4. Inabilit y to identify t he possible adverse impacts due to hazard potentials in upslope areas 5. Lack of t echnical knowledge to ident ify the threats due to active and/or dormant landslides locat ed in the upper slopes 6. Overlooking the increase in t he severity of hazard potential due to haphazard ground modifications

Even w hen the maps are used in digital form, misinterpretations occur when boundaries of several hazard zones cross a plot of land Example: Land parcel A – Belongs to a single hazard zone represented by the color yellow. Land parcel B – Belongs to two hazard zones interpreted by colors orange and yellow Land parcel C – Belongs to four hazard zones interpreted by colors green, yellow , orange and red In the case of land parcels B and C hazard potential may possibly be interpreted to the advantage of the user.

Problems were faced in interpretation of hazard potential by the non technical map users contd… A Web based tool was developed and proposed as a solution to minimize misinterpretation of hazard maps.

Information from the landslide hazard zonation maps, map of slope angle ranges, map of sub catchments and past landslides were considered .

User Select ion of District Select ion of Admin Division Select ion of Grama Niladhari Division Select ion of Land Parcel Number

Is there a past record of landslides/ rock falls/ subsidence within the land parcel Yes No Is there a past records of landslides/ rock falls/ subsidence within the enclosing catchment No Yes Is the land located within a forested area No Yes Construction/ development is not allowed Obtaining clearance from forest conservation Dept may be required Yes Construction/ development is allowed. Obtaining landslide clearance is not required Is the building density of the land parcel is <= 0.1 No Is the inferred landslide hazard potential is very high Is the land parcel located in an extremely steep slope Is the inferred landslide hazard potential is very low and the land parcel is located in a flat terrain Yes No No Yes Obtaining landslide clearance is required prior to approving construction/ development No Yes

Landslides are most likely to occur in this plot of land Building construction or any other development is not allowed within this land parcel

Expected future developments § Issuance of landslide early warning

§ In the present system, increase of landslide potential with the rainfall

values are manually evaluated and the early warning message is prepared

§ NBRO is presently developing her capacity to receive real time rainfall data § NBRO expects to develop a web based model for generation of real time

landslide warning map, to eliminate the human errors and delays associated with the presently used system

§ Development of web based decision making tool

§ Implement the system first as a pilot project in the study area and then

nation wide

§ Present 10 m accuracy level needs to be further improved to accurately

interpret the hazard potential levels

§ Present methodology heavily depends on field data collection and

therefore, improvements to the methodology to use high resolution spatial data and advanced GIS modelling is required to reduce the time involved

• Following two proposals are made to solicit UNESCAP’s

support – Building NBRO’s capacity for development of a geo portal for

real time landslide early warning

– Building NBRO’s capacity for improving the landslide hazard

zonation mapping methodology by using advanced GIS modelling to reduce the time involved in preparation of maps with increased accuracy (A project has been approved for LIDAR survey for a part of the country and NBRO is to be benefitted by the LIDAR data)