FLOOD EARLY WARNING SYSTEM FOR THE NZOIA RIVER BASIN Johnson M. Maina KMD
MT. ELGON THE NZOIA RIVER SYSTEM CHERANGANY HILLS FLOOD PLAIN LAKE VICTORIA
FLOODING SCENARIOS
FLOOD HAZARD FREQUENT FLOODS DYKE PROTECTION FAILURE Economic and social impacts of floods on the Nzoia Basin Structural damage to buildings, roads, communication lines, and land degradation Deaths of people and animals from drowning and injuries Displacement from homes. Possible outbreaks of diseases like malaria, cholera, dysentery etc due to presence of mosquitoes and contamination of water sources by the floodwaters. Contamination of wells and ground water which is a major source of drinking water by most rural communities Loss of harvests and crops in farms, loss of food stocks, supplies and produce from farms. Mental and physical stress (e.g. anxiety, depression, loss of security, domestic problems) Increased conflicts over water resources Nutrition problems/food insecurity- lack of food as the floods destroy food reserves
Integrated Flood Management Approach The Western Kenya Flood Management Project was informed by the report on Strategy for Flood Management for Lake Victoria Basin, jointly prepared by WMO/APFM and Kenya Government Min of Water & Irrigation, KMD. A component of the Western Kenya Community Driven Development and Flood Mitigation Project (WKCDD&FMP Report recommended the Integrated Flood Management Approach Structural: dykes, river training, check dams-control of siltation, multi-purpose dams (design stage)-hydropower, irrigation, fishing Non-structural: Catchment management (reforestation, agricultural practices, conservation of riparian areas), community awareness and education, early warning system, clearing of channel (vegetation, sediment deposits), dyke management, poverty reduction Participatory: community, inter-sectoral, inter-ministerial, CBOs, religious Organizations.
FLOOD EARLY WARNING SYSTEM Components of a flood forecasting & Warning System 1. Hydro-Met Networks: Weather and River Gauging Stations 2. Data acquisition, processing and analysis system: 1&2: Real-time Flood Monitoring System-data collection: observation and rapid communication system, processing, analysis, data base management system (dbms) 3. Forecasting and Warning Centre: models, software, hardware, expertise 4. Forecast dissemination: email, press, TV, Radio, telephone, Internet website 5. Response plan 6. Forecast review and development: feedback, improvement of forecasts/warning (reach, timeliness, accuracy)
NZOIA RIVER BASIN FLOOD EWS FEWS NZOIA RIVER BASIN Indigenous knowledge Weather Forecasts Experts Models FDFC (KMD) Data, Forecasts Forecasting/warning Feedback Feedback FWDC (OP) Warning response organizations/ groups Dissemination: Media, Bulletin, RANET FM Radio COMMUNITY Hydromet data
THE FLOOD DIAGNOSTICS AND FORECASTING CENTRE (FDFC) The Flood Early Warning System : (i) Set up a telemetric flood monitoring network to collect, in real-time, meteorological (e.g. flood producing storms) and river flow/flood information (ii) Established a National Flood Early Warning Centre (NFEWC now FDFC) at which, (i) data will be acquired in real-time, (ii)data will be integrated using models to generate flood forecasts and warnings to inform decisions in flood management. (iii) The center should operate 24 hours a day for 7 days a week all the year round have advanced telecommunications facilities to ensure access to data sources, including remote sensing technologies (e.g. satellite and radar); have personnel skilled in hydrology/ hydrometeorology. iv) FDFC is located at Kenya Meteorological Department HQs: To ensure rapid access to real-time meteorological data/ Information- forecasts (QPFs), Satellite imagery, others
OPERATIONS AT THE FLOOD DIAGNOSTICS AND FORECASTING CENTRE 1. Data acquisition Meteorological Network i) 35 Synoptic stations (distributed nationally, 3 are within the Nzoia Basin) ii) 16 Automatic Hydromet Systems & 3 Rainfall Stations within or close to the basin River Gauging Network: 3 Radar Water Level Stations: Rwambwa Bridge, Semogere and Webuye; Method: GPRS Telemetry and Use telephone calls Quantitative Precipitation Forecasts (QPF) from National Meteorological (Forecasting ) Centre (NMC) Satellite images - NMC
Automatic Telemetric Hydromet Stations WEBUYE (Automatic Telemetric) SEMOGERE Kaimosi FTC RWAMBWA
Solar Panel REMOTE UNIT/ HYDROMET SYSTEM GSM NETWORK HYDROMET CENTRAL BASE SYSTEM SUTRON 9210 Data Logger X-Connect will poll the 9210 and update the Synoptic User Interface with the newest data. Connection use radio. Fig 13: Observer Window GSM Modem GSM Network (900/1800MHz) GSM Modem AT/RH RG/TB WS/WD PYRO BARO BASE STATION COMPUTER
Fig 13: Observer Window HYDROMET AND RGS BASE STATIONS Fig 9: HydromET Central Base Station
OPERATIONS OF FDFC (contd) 2. Data processing and analysis i. Data entry (into database) and validation -Rainfall and water level ii) Spatial analysis of rainfall data iii) Estimation of areal rainfall (i.e. spatially averaged rainfall over Nzoia river basin) iv) Computation of river flow from rating curve v) Compilation of model data input files: areal rainfall, flow, evaporation, QPF vi) Data storage/archival for subsequent use
OPERATIONS OF FDFC (contd) 3. Streamflow/River level Forecasting i. Model in Use o The SMAR Model with Updating. SMAR means Soil Moisture Accounting Routine SMAR is a Component of the Galway Flow Forecasting System (GFFS). Model was calibrated using rainfall, flow and evaporation data for 2000 to 2007 period o KMD provided the MET data while the Water Resources Management Authority-WRMA & Ministry of Water and Irrigation provided flow data, rating equations oother Models: GeoSFM, MIKE Flood Watch ii. Update input data files- Rain, Discharge, Evaporation, QPF (Quantitative precipitation Forecasts) iii. Run model to obtain forecast flows iv. Convert forecast flows to water levels v. Prepare forecast chart vi. Determine whether to issue flood warning or not
7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 3.09 0.47 4.18 3.72 1.49 3.45 0 2.31 2.01 4.76 7.22 10.48 16.32 16.6 15.32 4.63 22.54 3.15 24.91 3.81 3.33 2.73 2.93 2.83 2.6 2.83 2.58 2.45 1.13 4.65 4.57 4.54 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Water Level (m) 26-Oct-08 27-Oct-08 28-Oct-08 29-Oct-08 30-Oct-08 31-Oct-08 01-Nov-08 02-Nov-08 03-Nov-08 04-Nov-08 05-Nov-08 06-Nov-08 07-Nov-08 08-Nov-08 09-Nov-08 10-Nov-08 11-Nov-08 Areal Rainfall (mm) Forecast chart Rainfall Forecast Level Water Level Flood Alert Flood Warning
4. Products presentation Main and regular Products Daily Flood Watch Update OPERATIONS OF FDFC (contd) Summary: weather forecast, observed and forecast water levels, flood risk category (colour coded) Flood risk indicators in detail: Rainfall distribution, areal rainfall amount and forecast Water level forecast chart Monthly Flood Diagnostics Bulletin Summary Hydrometeorological Analysis and Modeling Impacts during the month Rainfall Forecast, expected impacts and advisory 5. Dissemination of products/ Warnings to the Flood Warning and Dissemination Centre Methods Email Road Other methods? 6. Forecast review and development River level (m) 6 5 4 3 2 1 0 Observed and Predicted River levels at Rwambwa RGS Observed Predicted 1 Sep 08 6 Sep 08 11 Sep 08 16 Sep 08 21 Sep 08 26 Sep 08 1 Oct 08 6 Oct 08 11 Oct 08 16 Oct 08 21 Oct 08 26 Oct 08 31 Oct 08 5 Nov 08 10 Nov 08 15 Nov 08 20 Nov 08
FLOOD WARNING AND DISSEMINATION CENTRE(FWDC) USERS Budalang i community Public Emergency management e.g NOC, RED CROSS: logistics, response, evacuation, relief, shelter, medics Private sector, media Government Agencies Civil Society NGOs, CBOs Others- UNOCHA, FEWS NET Hotlines including mobile sms FORMS OF WARNING MESSAGES Daily Flood Watch Updates: Colour Code: Green-No Flood Risk Amber: Medium Risk RED: High risk Monthly Bulletins METHODS Mailing lists: email, fax Internet website Weather Radio RANET
Role of the Community in EWS Observers: Observe rainfall and water level Communication Network: information to FDFC and FWDC Relay data and Provide Indigenous Knowledge: forecasting and early warning. integration into Provide flood information/data Provide impacts information during floods Provide security for flood monitoring equipment Disseminate warnings in local languages e.g through the Bulala RANET FM Community Radio Station Community disaster and response plans RESPONSE TO WARNINGS
ISSUES THAT NEED CONSIDERATION QPFs are made available in a suitable format Build Capacity at the FDFC in analysis (especially ArcGIS) Use of various flood forecasting models Staffing levels necessary for 24/7 operation Sustainability of FDFC/FWDC Integration of Remote Sensing data sources - radar, satellite
THANK YOU FOR YOUR TIME
Modeling using GFFS Data Collection Analysis Correction Process Model Set Up Calibration Validation Model Implementation Flood forecasting Flood mitigation Flood mapping SMAR (EFFICIENCY = 90.35) 800 0 10 20 600 30 Discharge in cumecs 400 Observed Estimated rainfall 40 50 60 Rainfall in mm 70 200 80 90 0 100 2000 2001 2002 2003 2004 2005 Time in days