Geneva, Switzerland, 20 - 24 April 1998

1. OPENING OF THE MEETING

1.1 The Expert Meeting on Emergency Response Activities (CBS, CAS, UN/Humanitarian expert group) was held in Geneva from 20 to 24 April 1998. The meeting was opened by Mr Michel Jarraud, Deputy Secretary-General of WMO. Mr Jarraud welcomed the participants and thanked them for accepting WMO’s invitation to this meeting.

1.2 In his opening remarks, Mr Jarraud noted that the WMO CBS-XI held in Cairo, 1996 agreed on the need to address the issue of meteorological support in response to chemical incidents and the need for this meeting. He recalled the collaboration between WMO and IAEA in development and implementation of regional and global arrangements for the provision of transport model products from WMO designated Regional Specialized Meteorological Centers for environmental emergency response. He gave a brief status of implementation and noted recent consolidation activities covering guidance developed, training activities and exercises simulating environmental nuclear emergencies.

1.3 Mr Jarraud noted that in view of recent environmental disasters such as the South-East Asia forest fires, the relevant WMO designated centers had to and responded to these emergencies. He noted that concrete procedures for response to such other types of emergencies are developing and will continue to evolve. He noted that these issues would be addressed by a specific conference planned for June 1998 in the ASEAN region.

1.4 With regard to chemical incidents, Mr Jarraud noted the preliminary work addressed by the CBS Working Group on Data Processing (1997) and emphasized the need for the meeting to consider the guidelines developed by the Working Group and propose a programme of action for the future which could be eventually approved by the CBS.

1.5 Mr Jarraud noted that opportunity was also being taken to bring to the attention of the meeting proposal updates to the arrangements for the provision of meteorological assistance to UN humanitarian missions through the UN Office for the Coordinator of Humanitarian Affairs.

2. ORGANIZATION OF THE MEETING (agenda item 2)

2.1 Election of the Chairman (agenda item 2.1)

2.1.1 Mr Peter Chen, Canada, was unanimously elected chairman of the meeting.

2.2 Approval of the agenda (agenda item 2.2)

2.2.1 The meeting adopted the agenda which is given in Appendix 1.

2.3 Working arrangements for the meeting (agenda item 2.3)

2.3.1 The meeting agreed on its working hours and tentative work schedule.

2.3.2 There were 10 participants at the meeting as indicated in the list of participants given in Appendix 2.

3. DEFINITION OF REQUIREMENTS CONCERNING CHEMICAL INCIDENTS (agenda item 3)

3.1 During a broad discussion on meteorological requirements aspects related to chemical incidents it was noted that there was a need to have:

• Representative meteorological data reflecting the characteristic atmospheric conditions at and in the vicinity of the incident site,

• Data on evolution of atmospheric conditions within the incident area,

• Chemical information data base,

• Emergency planning and response information including meteorological support aspects,

• Expertise at various organizational levels, (e.g. regional to the local level), to provide model output interpretation and/or run very simple models on site to support rapid response,

• Dispersion model facilities,

• Good communication facilities for rapid receipt of notification and dissemination of results.

3.2 The meeting discussed the above requirements in further detail in three categories:

• Data (meteorological observations and incident related)
• Tools including models and visualisation facilities,
• Expertise.

While the meeting discussed response aspects of chemical incidents, meteorological requirements can also be identified in relation to measures that are related to the prevention of accidents, including long advanced planning that account for safe designs as well as those related to safe operations and practices (in plant and transportation of chemicals). The meeting expressed the importance of including advance planning (preparedness) for emergency response to reduce the consequences of chemical incidents.

3.2.1 Data

The meeting considered that to run dispersion models and understand and evaluate the dispersion processes involved, data that characterize the atmospheric conditions, especially local boundary layer, turbulence regime, of the site normally should include:

• Wind speed at 10 m
• Wind direction and the directional fluctuations (F » )
• Atmospheric stability (e.g. Pasquil category)
• Height of atmospheric boundary layer
• Humidity at surface
• Precipitation – occurrence and type
• Surface temperature
• Boundary layer profile (temperature, humidity, wind)
• Prevailing climatological conditions and dispersion climatology.

3.2.2 There are a number of approaches to obtain the characteristic atmospheric conditions. These may include making arrangements to have deployable mobile weather stations, or to establish standard observing facilities installed and operated at the chemical plant sites and to provide such data routinely from the site. Another approach in countries with dense networks and running mesoscale atmospheric models, their outputs could be used as input to dispersion models. Mesoscale model output could also serve to indicate the evolution of atmospheric conditions within the incident area.

3.2.3 The meeting further considered that to efficiently run such a model and provide outputs for rapid guidance, there is a need to have an up to date integrated system of other related data and information prepared and selectable to include for example, chemical information data base, possible accident scenarios, local surface cover characteristics, and related response planning information. These may include:

• Chemical inventories (e.g. types and quantities, geographically referenced to facilities, and transportation corridors)
• Type of possible release (fixed facility, mobile tanks, explosive, flammable, etc)
• Physical characteristics (mapped) of the incident site and nearby or local areas (e.g. out to 50 km)

• Action levels (e.g. Immediate Danger to Life and Health (IDLH) or Lethal Dose (LD50)) that are established in emergency response plans and linked to activating specific emergency response activities

3.2.4 The meeting was informed on chemical information data base and national emergency planning information of the US EPA/NOAA Computer-Aided Management of Emergency Operations CAMEO: a computer-based planning and response system designed to help emergency planners and responders at regional and local level for, and safely handle, chemical hazards does address these issues. CAMEO operates in two computer environments, IBM compatibles and Apple Macintosh. It contains chemical specific response information and recommendation for 3, 311 chemicals, a planning module for assisting the risk posed by extremely hazardous chemicals, and atmospheric dispersion model to assist in evaluating release scenarios, and a series of related data bases for storing and retrieving information required for planning and response, to serve as place holders for local information. CAMEO information may be displayed on maps to assess the relative risk presented by various chemical release scenarios and determine response actions to chemical emergencies.

3.2.5 Tools

Major tools include atmospheric boundary layer and dispersion models for domains less or equal to 50km. These should assimilate all available local data for diagnostic and prediction purposes. Various factors may come into play in the choice of a suitable model to be run:

• Nature and scale of incident (chemical species, emission conditions, surrounding terrain e.g. open field, urban setting, etc).

• Treatment of different dispersion regimes, e.g. heavy gas, passive tracer, or buoyant plume (explosion, fires).

• Nature of source (chemical release scenario), depending on whether it is a facility or transport incident, or different source term modelling (e.g. jets, pools, flashing liquid) to provide input for specific dispersion model.

• Input data requirements (some models will require extensive data sets); timeliness in receiving input data and turn-around time of running models will constrain the possible support to response, perhaps limiting response options to only major incidents.

3.2.6 With regard to dispersion models, it was noted that several NMSs operate EER procedures and models in support of chemical incidents. These models may be:

• Empirical ones informed by basic meteorological conditions such as wind velocity, vertical stability, existence of precipitation etc.

• Gaussian models of passive dispersion are easy to run on PCs, require little input data (wind and vertical stability) but are of limited applications.

• Integral models of heavy gas dispersion may also be run quickly and efficiently on PCs, and are applicable to a range of chemical releases.

• Complex time dependent three dimensional non-homogeneous models are difficult and expensive to operate in real-time. However, they may be used for scenario simulation of the pollution around a potentially hazardous site, taking into account the small scale features of the topography. The focus of this application is examining worst cast scenarios for prospective of inputs. It is necessary to run the models on a selected sample of worst case characteristic meteorological condition for each site with due consideration for worst case impacts. Such a catalogue of dangerous areas corresponding to the worse case meteorological situation could be useful in order to determine areas of vulnerability depending on worst case meteorological conditions.

• Model intercomparison and validation against observational data (e.g. from field trials) contributes to the understanding of the applicability of the various models.

3.2.7 Assimilation and application of the output information will require:

• Visualisation facilities for estimating hazardous zones from model output (e.g. scalable map size, zoom, pan, sections, animation, overlays, etc.)

• Geographical information system facilities for detailed interpretation of local meteorological effects on the movement and dispersion of chemicals and aiding in communicating output results using geographical references.

• Coupling with established action levels for hazards related decision making.

3.2.8 Expertise

The meteorological expertise required include expert knowledge and information related to the dispersion models and dispersion processes with a view to advising facility operators on meteorological data measurement systems and on implementation and operation of local dispersion models for within 10-15 km of site, and to advising other emergency response organisations out to within 50 km of site, and on interpreting model results:

• Meteorological data measurements (e.g. assessing data quality, representativeness for the model used)

• Selecting and operating suitable models

• Quality and suitability of model results assessed in the context of the specific incident

• Micro and meso meteorological processes especially in the lowest surface layer (Canopy layer) in relation to possible local effects

3.3 Role of National Meteorological Services

3.3.1 The meeting noted that requirement of meteorological support to emergency response for chemical incidents is broad and depends on nature of the incident. Due to wide ranging possibilities for chemical incidents within the interest of individual countries, the role of the NMS can also vary, and may include some or all of the following:

• Collecting and using, and providing all available meteorological data, from NMS network and other agencies facilities

• Implementing and operating tools including meteorological Numerical Weather Prediction (NWP) models, local dispersion models, related visualisation and geographic information facilities and possibly source term models

• Provide meteorological expertise and in particular provide advice to national, regional and local level emergency response teams

• Provide information and advice to facility operators on aspects of meteorological data and atmospheric dispersion models, including occasional review (e.g. biennial workshops) of what models are being run at industries or at emergency response agencies so as to ensure proper use of models and that meteorological data inputs are suitable and available.

• Play an active meteorological support role by participation in emergency contingency planning and simulation exercises.

3.3.2 It is desirable that NMS build its own capability operate at least simple models. The availability at NMSs of 3-dimensional models or access to their output products would assure availability of simple model input data such as atmospheric stability and height of atmospheric boundary layer, humidity and precipitation. Chemical industry plants and emergency response services may have and run their own simplified models on site and establish linkage with NMS to receive local special meteorological forecaster interpretation and advice as support to the response team.

3.3.3 While complex models may be executed in real-time response mode during an incident, they may also be used to pre-compute meteorological data and the dispersion data to evaluate turbulence parameters, in the interest of eliminating turn-around time needed to execute models during an incident. Preparatory calculations may be done based on known poor (or worst case) atmospheric dispersion conditions. This approach would provide for ready reference the estimated hazard, for example affecting nearby populations or sensitive resources, for ready reference. Probabilistic estimates could be generated.

3.4 Possible Gap in the Provision of Services

3.4.1 While it is recognized that the identified roles of the NMSs are broad in relation to requirements in supporting emergency response for chemical incidents, and that individual NMSs may priorize certain roles as more critical to implement than others, it should be noted that due to present day operational realities, NMSs are not in a position today to provide specialized support to response in:

• Very short time scale incidents
• the near field (i.e. near the source)
• Complex source and obstacles to flow (e.g. buildings wake)
• Spills where chemicals are released into a body of water

3.4.2 The operational capabilities and communications for provision of NMS services at national, regional and local offices to meet emergency response operations may in some cases not be assured during emergency situations.

3.4.3 Issues needing to be addressed include minimum standards for dispersion models with a view to harmonization of inputs (in relation to measurement data) and outputs (in relation to parameters used in defining hazards). As well, it was felt that a methodology is needed for calculating uncertainties in the model estimates, or the sensitivities of outputs to varying input parameters.

4. GUIDANCE TO NMS ON MODELING ASPECTS (agenda item 4)

4.1 The meeting agreed that as part of future tasks, guidance to NMS should be developed addressing the following issues:

• Best or optimal observational data set needed for input to modelling to assure useful output: particularly those that indicate the turbulence regime in the surface layer.

• basic micrometeorological information pertaining to atmospheric dispersion and chemical incidents (training document or referenced) as a means for evaluating model outputs

• situations where models can not be used reliably or at all

• modelling and parameterization schemes (e.g. wet and dry deposition schemes for different chemical species), and output parameters including toxicity (exposure) and intervention levels

• situations where certain dispersion models apply

  • In view of a large number of available models, guidance is needed on selecting models for implementation at NMSs

• general aspects of source modelling of different types of incidents/sources

  • For example incidents involving pipes, tankers, chemical storage, and fire and give guidance on which kind of models can be used in each type.

5.1 The meeting taking into account the guidelines of the CBS Working Group on Data-processing addressed the issue of NMS interface with National environmental disaster management and lead emergency response agencies, i.e., interfaces at the national level. This aspect includes (normally) interface with NMSs in neighbouring States. This would recognize as well that some chemical incidents could affect areas outside of national boundaries, although due to the highly localized nature of such incidents, this possibility is considered as a relatively rare situation. In incidents involving fire and or an explosion, the chemicals involved have a greater potential to spread to greater distances and possibly affect more than one State. The following action plan is proposed:

5.2 The NMS could identify existing procedures and make arrangements with authoritative agencies responsible for environmental contingency plans to assure participation of the NMS in national, provincial and local contingency planning. This participation, and extent of involvement developed through careful consultations, and based on evaluation of requirements of specific measures, will in due course define the role of the NMS.

5.3 Specific action will need to be taken for NMS to obtain a registry of chemical sites and holdings. In addition, NMS can assist in developing climatologies of prevailing meteorological conditions at major chemical sites. As well NMS should arrange to acquire in-situ (plant site) meteorological data routinely and in real-time. Similarly, information on high traffic transportation corridors used by chemical transport could be acquired from Transport Ministries and examined jointly with the view to determine high risk zones for transport accidents (e.g. from meteorological perspective, high wind or fog prone zones could represent accentuated risk when combined with other factors).

5.4 NMS should seek information on and review dispersion models used by other national agencies and industries. NMS should take steps to understand and build a knowledge base on what are the response arrangements at various levels of authorities in relation to different incidents including particularly roles and responsibilities and communication arrangements.

5.5 NMS should make arrangements to participate in tests and exercises related to response to chemical incidents.

5.6 NMS will need to set up appropriate emergency response measures or procedures and facilities for their supporting role in emergency response (e.g. as required by up-to-date and approved contingency plans).

5.7 In some cases NMS’ observational systems such as remotely sensed by satellite or radar platforms might be of help in estimating the emission characteristics and plume extent.

5.8 Within the framework of its participation in national and local emergency contingency planning, NMS should provide authoritative information on meteorological aspects of an incident to be factored into public information mechanism.

5.9 Interfaces may be required among NMSs (likely within a limited geographical region), for three distinct reasons; appropriate action by NMS may include:

• Developing bi-national shared stake arrangements with neighbouring states and procedures for response to agreed areas of common concern, in considering trans-national boundary incidents.

• Developing back up arrangements and procedures for response between NMSs and/or NMS (or NMCs) and RSMCs (EER) in accordance with mutually agreed conditions.

• Developing regional economic grouping arrangements and procedures for response by agreed regional grouping centre(s) as may be agreed by the economic grouping concerned.

5.9.1 For neighbouring States, Regional and sub-Regional operational response interfaces could be developed and implemented among NMSs, preferably aligned or harmonized with other existing routine coordination mechanism among the same NMSs. Specific documentation on how NMS and RSMC deal with chemical accidents should be included as part of relevant contingency plans. Coordination of emergency response, for planning and response to incidents should include:

• Identification of contact points (programme and operational)

• Arrangements (e.g. data and information exchange, communications, security of information, exercise)

• Criteria for activation and deactivation of Regional or sub-Regional interface
• Harmonizing of the authoritative information on meteorological aspects of the incident, as well the technical guidance on the dispersion across national boundaries.

In general, for Back-up mode arrangements, not necessarily developed for trans-national boundary incidents, the required meteorological support role could be provided by another NMS, likely within the same Region or sub-Region (a neighbouring NMS might be convenient) or an RSMC (EER), as may be agreed to through a bilateral interface between the parties concerned. However, in view of this support role requiring highly specialized knowledge of local meteorological conditions and geographical information, and as well a potential for over demand for support services, only limited support from RSMC (EER), defined on a case-by-case basis, should be expected for chemical incidents within the current framework. This concept of back-up is applicable in similar fashion in the case of regional economic groupings.

It is encouraged that all NMSs improve their capabilities in relation to possible roles for NMSs for providing meteorological support in preparedness and response to chemical incidents, in their own country’s context. While the noted possible actions are gradually undertaken, NMSs with existing specialized capabilities in this regard are encouraged to provide some support (limited by expertise and resources), as well as leadership and encouragement to the other NMSs with evolving capabilities.

5.10 The meeting learned of possible coordination and cooperation among International Organizations that share interests in improving the safe management of chemicals and chemical accidents. The International Programme on Chemical Safety, involving UNEP, ILO, and the WHO was presented. It was noted that a meeting will be held in Geneva (June 18, 1998) of the U.N. Interagency Coordination Group for Chemical Accidents, led by OECD. This Group was prescribed following UNCED in Agenda 21 (Chapter 19). In addition, the OECD’s Chemical Accidents Programme (ref: Highlights and Plans for Future Work. February 1998) could provide opportunities for WMO and Members to participate in information sharing or specific project undertakings (for example where certain National participation is already indicated). It was noted that the Principle of Outreach is a major objective of an established Expert Group within this Programme. Some Projects carried in the coming period appear to be directly associated with preparedness and response aspects of chemical accidents. The UNEP’s Awareness and Preparedness for Emergencies at Local Level (APPEL) Programme promotes a process for responding to technological accidents and has adopted the U.S. EPA’s CAMEO programme (see item 3.2.4 above) for international APPEL applications.

5.11 The meeting noted the proposals of the Working Group on Data-processing for revision of Appendix I.5 of the Manual on the GDPS – Arrangements for the provision of meteorological assistance to UN Humanitarian Missions. No specific issues were raised on the proposed revision. The meeting agreed that in the event of large scale chemical incidents which trigger requests for international assistance for counter measures, these arrangements would be applicable in the provision of relevant chemical response products together with other meteorological information to UNEP/OCHA and the WMO Secretariat. Noting that chemical response products are not intended for public nor wide distribution, the posting of such products on the WEB should require restricted access.

6.0 CLOSURE OF THE MEETING (agenda item 6)

The meeting was closed by the Chairman on Friday, 24 April 1998.

______________________

APPENDIX 1

PROVISIONAL AGENDA

1. OPENING OF THE MEETING

2. ORGANIZATION OF THE MEETING

2.1 Election of a chairman

2.2 Approval of the agenda

2.3 Working arrangements for the meeting

3. DEFINITION OF REQUIREMENTS CONCERNING CHEMICAL INCIDENTS

4. DEVELOPMENT OF GUIDANCE TO NMSs ON MODELLING ASPECTS

5. CONSIDERATION OF INTERFACE PROCEDURES

6. CLOSURE OF THE MEETING

APPENDIX 2