Team Leader- Hydrology The Regional Integrated Multi-Hazard Early Warning System for Africa and Asia (RIMES) Ensembles Probabilistic Long Lead Flood Forecasts For Community Level Applications S.H.M. Fakhruddin Team Leader- Hydrology fakhruddin@rimes.int www.shmfakhruddin.com Measuring ‘Real’ Impact Monday 25th June 2012 UK CDS, Wellcome Trust, 215 Euston Road, London NW1 2BE Free Powerpoint Templates

Discussion Topics About RIMES & Key Activities End to End EWS Case Study on Flood forecasting and Agriculture Risk Management

RIMES Member States

Purpose and objectives Purpose: Provide early warning services for enhanced preparedness, response, and mitigation of natural hazards, according to differing needs and demands of its Member States Objectives: Facilitate establishment and maintenance of core regional observation and monitoring networks and ensure data availability for early warning Provide regional tsunami watch within the framework of UNESCO Intergovernmental Oceanographic Commission (IOC) Support National Meteorological and Hydrological services for providing localized hydro-meteorological risk information within the framework of the World Meteorological Organization (WMO) Enhance warning response capacities at all levels (national to community) within each national early warning framework

Governance Council Heads of NMHSs/ national scientific and technical agencies generating multi- hazard early warning information, empowered to make policy decisions, on behalf of governments, concerning regional early warning arrangements for enhanced preparedness for, response to, and mitigation of natural hazards Chair: Government of India Secretariat Carries out the decisions and tasks assigned by the Council, and provides support to the Program Unit in managing the regional early warning center Government of Maldives (Presidential Task Force led by the H.E. Vice President of Maldives) serves as Administrative Secretariat; Government of Mongolia (Ministry of Foreign Affairs) as Program Secretariat Program Unit Responsible for the day-to-day operation and management of the regional early warning center and the implementation of programs and activities Co-located with the RIMES regional early warning Center

Organizational Chart Chief Scientist Tsunami Early Warning Warning Coordination Scientist Seismologist Oceanographer Telecommunications Specialist System Analysts (2) Decision-support Tool Development Specialist Seconded Scientists (6) Team Leader Early Warning Tsunami Hazard and Risk Assessment Expert Climate Risk Management Earthquake Hazard and Risk Assessment Expert Project Teams Climate Impact Assessment Team GIS and Survey Specialist Secretariat Council Director, Program Unit Program Management Tsunami Watch Provision Support to Hydro-Met Services Societal Applications Capacity Building Specialist Chief Finance Officer Human Resource and Administration Officer ICKM Specialist Climate Forecast Application Team Climate Change Seasonal Forecasting Severe Weather Synoptician Hydrologist System Analyst Seconded Scientists (2)

Key services Tsunami Watch Provision to National Tsunami Warning Centers Seismic and sea level monitoring and data exchange Provision of earthquake alerts and regional tsunami bulletins Tsunami hazard and risk assessment tools for local coastal inundation forecasting Support to National Meteorological and Hydrological Services Customization of climate and weather forecasting models for generation of more reliable, location-specific severe weather and short- and medium-term weather forecasts, and seasonal climate outlook, having longer lead times Downscaling of global climate models for generating high-resolution climate change information for national and local level planning Development of decision-support tools Translation of products of research into operational forecast products and testing these for local level application

Key services continued Capacity Building on End-to-End Early Warning Early warning system audits Assistance in establishing and maintaining observation and monitoring stations of regional benefit Training of scientists (in-country and RIMES secondment program) Development of decision-support tools Strengthening national early warning provider and user interface Application of tailored risk information at different time scales in decision- making Enhancing community responses to early warning

EW System

EW System Structure Detection Subsystem Management Subsystem Monitoring, detection, data Assessment, data analysis, prediction Management Subsystem Risk Assessment, interpretation, communication Response Subsystem Interpretation, confirmation and response

Reasons for Warning Failure ?

Gaps Regulatory framework for warning Stakeholders involvement and roles Observation/ monitoring Aging and insufficient observation and data communication facilities Data analysis Data sharing among agencies Numerical prediction capability Skilled human resource Capacity to make use of new generation forecasts Prediction Risk assessment Potential impact assessment Local level potential impact assessment not done Language Localized, relevant Warning formulation Preparation of response options Institutional mechanism, linkages SOPs Redundant communication systems Reach to special groups Dissemination to at-risk communities Public awareness Communication of forecast limitations Lack of trainers/ facilitators Resources to respond to warning Emergency response plans Public education/ awareness Mitigation programs Community response

A Case Study- Bangladesh

Probabilistic Flood Forecasting and Applications in Agriculture Research Project initiated since 2000 and completed in 2007 GoB requested RIMES to continue to support RIMES provides 10 days lead time flood forecast to GoB and build capacity

Institutional Collaboration For Sustainable End-to-end Flood Forecasts System Climate (rainfall and di scharge) forecasting technology RIMES- CFAN BMD RIMES Flood forecast RIMES FFWC Discharge translation Agro met translation Interpretation DMB, DAE RIMES, Local Partners Communication RIMES, Local Partners End users

Forecast lead time requirement Flood risk management at community level decisions and forecast lead time requirement Target groups Decisions Forecast lead time requirement Farmers Early harvesting of B.Aman, delayed planting of T.Aman 10 days Crop systems selection, area of T. Aman and subsequent crops Seasonal Selling cattle, goats and poultry (extreme) Household Storage of dry food, safe drinking water, food grains, fire wood Collecting vegetables, banana 1 week With draw money from micro-financing institutions Fisherman Protecting fishing nets Harvesting fresh water fish from small ponds DMCs Planning evacuation routs and boats 20 – 25 days Arrangements for women and children Distribution of water purification tablets Char households Storage of dry food, drinking water, deciding on temporary accommodation

Discharge Forecast Schemes (I). Initial Data Input (II). Statistical Rendering (III). Hydrological Modeling (IV). Generation of Probabilistic Q (V). Forecast Product Hydrological Model Lumped Distributed Multi-Model Discharge Forecasting Discharge data Accounting for uncertainties Final error correction Generation of discharge forecast PDF Critical level probability forecast Hydrologic model parameters NOAA and NASA (i.e.CMORPH and GPCP) satellite precipitation & GTS rain gauge data ECMWF Operational ensemble forecast Downscaling of forecasts Statistical correction

2007 Flood- Brahmaputra Ensemble Forecasts and Danger Level Probabilities 7-10 day Ensemble Forecasts 7-10 day Danger Levels 7 day 8 day 7 day 8 day 9 day 10 day 2005 (not shown) and 2006 (shown here) discharges for both Brahmaputra (left) and Ganges were well below severe flooding levels, as forecasted. 9 day 10 day 18

2012- Brahmaputra Ensemble Forecasts 2005 (not shown) and 2006 (shown here) discharges for both Brahmaputra (left) and Ganges were well below severe flooding levels, as forecasted. 19

Plumes and probability pies for the first Brahmaputra flood July 28-August 6, 2007 Model able to meet three fundamental information needs of communities at risk

Distribution of H combined with DEM --> probabilities of flood classes Distribution of H values

Vulnerability & Flood Risk Assessment Development of flood risk map which will include: low probability medium probability high probability

Flood Risk Map

Decision Support System High flood J F M A S O N D T.Aman 1 3 T.Aus 2 Jute S.Vegetables 4 Cattle 5 … Community Outcomes Delayed seedling raising, gapfilling, skipping early fertilizer application Advance harvest of paddy ( 70-80 % mature) Early harvest of jute for rotten in water Pot culture (homestead), Use resistant variety Food storage, flood shelter, vaccination de-warming

Decision Support System (DSS) High flood J F M A S O N D T.Aman 1 3 T.Aus 2 Jute S.Vegetables 4 Cattle 5 … Recommendations Delayed seedling raising, gapfilling, skipping early fertilizer application Advance harvest Early harvest Pot culture (homestead), Use resistant variety Food storage, flood shelter, vaccination de-warming

USER MATRIX on Disasters, Impacts and Management Plan for Crop, Livestock and Fisheries

Decision Tree

Risk Communication of flood forecasts

Risk Communication for Flood Forecasts Mobile phone Sending SMS to Mobile Flag hoisting 29

Community responses to flood forecasts

Economic- Benefits In 2008 Flood, Economic Benefits on average per household at pilot areas Livestock's = TK. 33,000 ($485) per household HH assets = TK. 18,500 ( $270) per household Agriculture = TK 12,500 ($180) per household Fisheries = TK. 8,800 ( $120) per households Experiment showed that every USD 1 invested, a return of USD 40.85 in benefits over a ten-year period may be realized (WB).

Expansion of Areas

Team Leader- Hydrology Thank you S.H.M. Fakhruddin Team Leader- Hydrology fakhruddin@rimes.int www.shmfakhruddin.com