Flood Forecasting Warning Karro by de443B3


									Subgroup on flood
 forecasting and
     Activity report
      Ilmar Karro

     RAVI WGH XI session, Toulouse 25
              March 2009                1
   General
   Review on the subgroup activities
   Identified related activities
   Progress of the subgroup
   Remarks
   Proposals/recommendations

                   RAVI WGH XI session, Toulouse 25
                            March 2009                2
             ToR SG FFW from Heidelberg
                 (RA VI session XIV)
   The subgroup is considered to be composed of
    meteorologists and hydrologists with the aim of
    improving the capability of NMHSs in flood
    forecasting and warning for different types of floods.
   To study and report on present applications of:
       The development and operational use of common
        probabilistic methods.
       The development and operational use of ensemble
       The development and operational use of methods for
        estimating, calculating the uncertainty in the forecasts.

                          RAVI WGH XI session, Toulouse 25
                                   March 2009                       3
         ToR for SG FFW from Heidelberg
               (RA VI session XIV)
   To promote co-operation between meteorologists,
    hydrologists and end-users both in operational activities
    as well as in research and development.

   To cooperate with relevant actors in the area of flood
    forecasting and warning, mainly with respective
    members of CHy AWG on flood forecasting and with
    the EU Expert Circle on Flood Forecasting (EXCIFF)
    and the European Flood Alert System (EFAS)

                      RAVI WGH XI session, Toulouse 25
                               March 2009                    4
                    RAVI SP and AP
   During this period the RAVI Action Plan was updated
    for the years 2008-2011 and the subgroup was assigned
    two tasks.
       Review of operational forecasting methods and development
        targets in the region including existing cooperation between
        NMHSs with trans-boundary river basins. (2009).
       To facilitate activities of NHSs in the region for
        improvement of hydrological forecasting including regional
        initiatives (EFAS, EXCIFF). To cooperate with EC on
        mechanism for the future implementation of EFAS including
        the system of relevant data exchange. (2009)
                         RAVI WGH XI session, Toulouse 25
                                  March 2009                       5
    Review on the subgroup activities
   Participation in planning of the WMO
    workshop on “Enhanced flood forecasting in
    Europe” in Bratislava 12-14 December 2005.
     During the workshop the chairman made a
      presentation of the subgroup.
     The final report is available from WMO.

                   RAVI WGH XI session, Toulouse 25
                            March 2009                6
    Review on the subgroup activities
   The chairman was member of the organizing
    committee (together with other members of the
    subgroup representing CHR) for the joint WMO/CHR
    workshop in Berne in March 2006.
   The themes of the workshop were
       Uncertainty in numerical weather predictions
       Hydrological Ensemble forecasts
       Communication of uncertainties
   Reports from the workshop is found on the CHR
    Webb side www.chr-khr.org

                         RAVI WGH XI session, Toulouse 25
                                  March 2009                7
    Review on the subgroup activities
   The chairman acted as a test pilot in the EXCIFF action 5c,
    which aimed at defining the good practices for informing the
    general public about flood forecasting and to write a guide
    relating to these good practices.
   The chairman participated in one writing session in Brussels and
    other members provided valuable input to the work with the
   The guide “Good Practice for delivering Flood-related
    Information to the general public” is now finalised and available
    from JRC.
   Planned workshop on flash flood in Norrköping in November
    2006 together with EXCIFF, was unfortunately cancelled.

                         RAVI WGH XI session, Toulouse 25
                                  March 2009                            8
    Review on the subgroup activities
   Subgroup meeting during the 3rd Hepex-
    conference in Stresa Italy 27-29 June 2007. (First
    discussion about the questionnaire)
   Subgroup meeting in Prague 4 October 2007.
    (real start on the work with the questionnaire)
   Subgroup meeting in Koblenz March 2009.
    (Finalization of the evaluation of the

                    RAVI WGH XI session, Toulouse 25
                             March 2009                9
       Identified related activities
   EFAS (used pre-operational)
   Meteoalarm (operational-vigilance)
   Preview (finished project)-operational
   COST Action 731 (ongoing project)
   MAP and D-PHASE (ongoing project)
   HEPEX (research - testbeds)

                   RAVI WGH XI session, Toulouse 25
                            March 2009                10
        Progress of the subgroup (1)

   The subgroup consisted of both meteorologists and
    hydrologists and we have found that in the identified
    related activities also cooperation between the
    disciplines is being considered as a prerequisite for
    success. Also end users are being included in the
   Co-operation with EXCIFF was established and the
    subgroup took part in the preparation work to the
    EXCIFF guide “Good Practice for Delivering Flood-
    Related Information to the general Public”.
                     RAVI WGH XI session, Toulouse 25
                              March 2009                    11
      Progress of the subgroup (2)
   The development of EFAS and the work with
    making EFAS operational was monitored by the
    chairman and the issue is recommended to be
    monitored also in the future. The financing of
    the running of an operational EFAS will be an
    issue to follow and report on for the coming
   A questionnaire was issued and evaluated to deal
    with the first objective in the RAVI Action plan.

                   RAVI WGH XI session, Toulouse 25
                            March 2009                12
         Progress of the subgroup
          (The questionnaire 1)
   Preparation phase. (web-based, took time, answering
    rate 56%, 26 countries)
   Contents
    1. general
    2. forecasting methods and models
    3. forecasting products and dissemination systems
    4. flash floods
    5. verification and evaluation
    6. strengths and weaknesses in the forecasting process
    7. economic and legal aspects

                    RAVI WGH XI session, Toulouse 25
                             March 2009                  13
                Progress of the subgroup
                 (The questionnaire 2)
   1. General (questions 1-8) A majority of the respondents are from joint hydrometerological
    services, but the work is mostly done by separate units. If you are in a separate hydrological
    institute the national meteorological service is the main provider of forecasts but there is also
    a multiple use of available forecasts, including forecasts from private sector providers.
   Almost all of the respondents say that their service is responsible for flood forecasting on a
    national scale, but many have also responsibilities on river basin and regional scale. Almost half
    of the respondents say that they are responsible on the trans national scale. Most of the
    answering services say that no other organisation is responsible for warnings for the same area,
    but there are also a number of respondent’s claming that there exist other providers for
    forecasts or warnings.
   The most common types of floods that occur are flooding due to precipitation, due to snow
    melting, urban floods and flash floods. The precipitation due to precipitation and snow
    melting are also forecasted by a majority of the respondents but urban floods and flash floods
    are forecasted by round a third of the respondents.

   2. Forecasting methods and models (questions 9-12) A variety of different methods and models
    are used (see question 9 in annex 2). 65% of the respondents report on use of meteorological
    ensembles in their forecasting process. The most commonly used real time data are
    precipitation, air temperature, water stage, snow cover and reservoir data. The most used
    processes to model according to the respondents are snow cover accumulation and melting,
    rainfall-runoff, evaporation and potential soil moisture.

                                   RAVI WGH XI session, Toulouse 25
                                            March 2009                                             14
              Progress of the subgroup
                (The questionnaire)
   3. Forecasting products and dissemination systems (questions 13-20). Mostly
    the produced products are deterministic but over half of the respondents
    report production of probabilistic products. The users are different kind of
    authorities (governmental, local, water and civil protection), general
    public and media. More than 60% also say that neighbouring countries
    are users. Under question 15 (annex 2), links to some examples on
    probabilistic information is given.
   Lead time for forecasts and warnings show a wide variety amongst the
    respondents. A majority of the answering institutes say that they in some way
    deal with uncertainty in the forecasting process but only 50% report on
    indication of uncertainty given to the users. Only one third say that
    training is being organised for end users who are using ensembles and
    uncertainty. In the comments a need for probabilistic information and
    uncertainty measures are mentioned.
   The most used dissemination media are web, telephone/fax, e-mail and
    mass media.

                             RAVI WGH XI session, Toulouse 25
                                      March 2009                                15
              Progress of the subgroup
               (The questionnaire 3)
   4. Flash Floods (questions 21-23) A large majority of the respondents say that
    they don’t have special products for flash floods. The ways of
    dissemination are mostly the same as for other kinds of flooding. Around two
    thirds of the respondents report that they miss a local warning system.
    Some systems building on alerts from gauges are in use.

   5. Verification and evaluation (questions 24-28) Verification of the
    hydrological forecast/warning after a flood event is made by a large majority
    of the respondents and regularly by more than half of the answering
    institutes. The most common verification method is still subjective/manual.
   The evaluation report is distributed internally and also to the
    government/ministry and in some cases to other authorities/users. Mostly
    the verification is done as a report but many also only verify by criterion
    values. Both the meteorological part and the hydrological part are evaluated,
    separately as well as integrated.

                             RAVI WGH XI session, Toulouse 25
                                      March 2009                                16
               Progress of the subgroup
                (The questionnaire 4)
   6. Strengths and weaknesses in the forecasting system (questions 29-31, 37) The
    strengths reported are, expertise, models, co-operation. Expertise, especially when
    dealing with extreme weather events is very important and also requires training.

   In the part of weaknesses there is noted a general demand for increase of lead time,
    more frequent updated forecasts, more detailed forecasts and more accurate
    forecasts. Also a need for better meteorological input and probabilistic
    products as well as better use/understanding of uncertainty is highlighted. In question
    31 a lot of ongoing research activities are reported, mainly national..

   7. Economic and legal aspects (questions 32-36) Just a few of he respondents say that
    there has been an economic valuation of the flood forecasting service carried out.
    Most of the answering institutes report that they have a legal mandate when it comes
    to issuing forecasts and warnings. The present legal framework for the work with
    flood forecasting and warning is considered adequate by nearly 70% of the
    respondents and around half of the respondents report that liability aspects has been
    an issue in the past. The private sector competition is seen to be an issue in the
    flood forecasting in the aspects, data collection, modelling and dissemination.

                                RAVI WGH XI session, Toulouse 25
                                         March 2009                                      17
   In the questionnaire it can be identified that many institutes use probabilistic
    forecasts, some operationally and some in research mode. This is encouraging
    and should be followed up. The general conclusion is that research is
    going on in many institutes to be able to fully adopt probabilistic
    forecasting into practise.
   The quantification and communication of uncertainties is still a big issue.
    That should also be combined with training of the staff as well as training of
    the end users.
   There are a number of reports on national co-operation between the
    meteorological and hydrological institutes as well as co-operation with other
    authorities. This is also a good sign and further improvement should be
   Generally it is important that training is included in the organizations
    strategic planning.
   It seems there might be a possibility to further enhance tinternational co

                             RAVI WGH XI session, Toulouse 25
                                      March 2009                                  18
   It is important to identify reports and results from ongoing
    activities/projects in order to give an input to those institutes who wish to
    implement the results (i.e. developed methodologies). The EXCIFF guide is
    one example of documented guidelines.
   There is still a need for improvement regarding flash floods, both
    regarding modelling and dissemination. In dissemination there may be an
    organisational difficulty looking at the responsibility between the
    meteorological and hydrological side.
   Post processing of meteorological parameters is a technical issue where there
    also may occur questions about the responsibility between the meteorological
    and the hydrological side.
   There may be a need to study and report on real time updating techniques of
    the hydrological forecasting models.
   There are two relevant WMO publications in this area. The technical
    regulation nr 49, volume 3 Hydrology and Guide to hydrological practices
    (WMO 168)

                             RAVI WGH XI session, Toulouse 25
                                      March 2009                                19
   The fact that private sector companies are being more and more involved in both
    forecasting and warnings in meteorology as well as hydrology may be cause problems
    if there should be different messages sent to the users. For the benefit of the user it
    is preferable to speak with a single voice.
   Not only flood forecasting is interesting. Also forecasting for low and normal water
    conditions becomes increasingly important.
   Seasonal forecasting is used to forecast extreme hydrological events, as low water
    (droughts). Seasonal forecasts are also a important tool in predicting heat waves
    affecting the water temperature, which then has an impact both economically
    (transport) and ecologically. Forecasts with longer lead times can also be used for
    generally forecasting the inflow to reservoirs. Winter time you could use seasonal
    forecasting for predicting ice.
   The issue of glacier lakes is identified. This can be an increasing problem in the future
    in the climate change perspective.
   Taking into the account the reports on climate change there may be an increasing need
    to be able to better forecast flash floods, urban and drainage floods and ice jams back

                                RAVI WGH XI session, Toulouse 25
                                         March 2009                                        20
              (to RA VI WGH)
   Continue the work to study and report on progress regarding the use of
    probabilistic forecasting and the inclusion of uncertainty in products to users.
   Follow the operationalisation of EFAS.
   Study and report on progress regarding flash flood forecasting and also
    generally dissemination of forecasts with short lead times.
   Identify crucial questions from the questionnaire distributed in 2008 and
    make a follow up in some year’s time to be able to identify trends.
   Make an inventory and report of Open Source software for the forecasting
   Look in to further co-operation with organisations within the RAVI region
    and also global activities as i.e. HEPEX.
   Inform about related projects/activities so that the results can be made used
    more widely within the RAVI region.

                              RAVI WGH XI session, Toulouse 25
                                       March 2009                                  21
                (to NMHSs)
   In co operation with the users try to adapt to probabilistic forecasting for
    flood forecasting in daily practice.
   To look into the results from activities done internationally in order to see if
    the results are nationally applicable and to avoid duplication of work.
   An increase in objective verification methods/systems could be fruitful.
   Make sure that education of personnel for forecasting and warning
    continuously is included in the strategic planning.
   To further increase the cooperation between the meteorological and
    hydrological communities as well as the user community. This includes
    defining clear responsibilities between the different partners involved in flood
   To further increase international co operation in research and operational
    activities in flood forecasting and the efforts to make research results

                             RAVI WGH XI session, Toulouse 25
                                      March 2009                                  22
Thanks for your
   attention !

    RAVI WGH XI session, Toulouse 25
             March 2009                23
               RAVI SP and AP
   Resolution in Heidelberg to develop a Regional
    Strategic Plan (SP) for RAVI 2008-2011 in line
    with the WMO long term plan.
   For the previous period 2006-2007 an Action
    Plan (AP) was adopted in Heidelberg. (2
    hydrological tasks)
   As an annex to SP a new AP for 2008-2011 will
    be prepared.

                   RAVI WGH XI session, Toulouse 25
                            March 2009                24

To top