Docstoc

REPORT TAG MEETING

Document Sample
REPORT TAG MEETING Powered By Docstoc
					   Centre for Research on the Epidemiology of Disasters (CRED)




                          Report of the


          EM-DAT Interim Technical Meeting




                     Brussels, July 8-9, 2002




CRED                                                  UCL
                                                       -2-




With the participation of :
Asian Disaster Reduction Centre (ADRC)
Kobe, Japan
Masaru Arakida

Centre for Research on the Epidemiology of Disasters (CRED)
Brussels, Belgium
Debarati Guha-Sapir, Regina Below, Philippe Hoyois, Jean-Bernard Germaux, Darren Shaw*

National Oceanic and Atmospheric Administration, Office of Global Programs
(NOAA-OGP)
Washington, USA
Aurelia Micko, Geoffrey Rowland

Office for the Coordination of Humanitarian Affairs United Nations (OCHA),
Reliefweb
Geneva, Switzerland
Craig Duncan

Office of U.S. Foreign Disaster Assistance, U.S. Agency for International
Development (OFDA/USAID)
Washington, USA
Nate Smith




* CRED Advisor, Centre for Tropical Veterinary Medicine (CTVM), Royal (Dick) School of Veterinary Studies,
University of Edinburgh, Edinburgh, UK




                                                             EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                  -3-


1. Background
Following the third EMDAT Technical Advisory Group meeting hosted by UN-OCHA in New
York in March 2002, the Climate Information Project (CIP/NOAA) offered to help CRED develop
web-based services to allow for dynamic display (tables, graphs, etc.) of EM-DAT and GLIDE
information, as well as online management of its products. Through informal discussions with
CRED, the CIP offered to help develop web services and maintain these for at least 2 years, thereby
giving CRED time to decide if such dynamic access and management of GLIDE and EMDAT was
useful and sustainable.

2. Meeting objectives
The EMDAT Technical Advisory Group (TAG) meeting in New York concluded that EMDAT
should:

    1. Review its current configuration and transform it to a relational database such that
       differences between Reliefweb and CRED will be minimised and data management will be
       rationalised;
    2. Develop a search engine to facilitate the data extraction;
    3. Rework the data entry procedures as a function of the relational model and to reduce interval
       between addition of records and availability on the net.

The workshop in Brussels was convened to discuss the above three issues focused around prototypes
prepared by CIP/NOAA. The meeting hammered out the detailed technical questions between the
programmers and those involved with data management. The goal was to finish the finalised
prototype and the detailed specifications for related modifications.

3. CIP-NOAA prototype

3.1. Technical brief

Currently, the demonstration site (located at http://www.em-dat.net/start.html) serves to showcase
what features might be developed.

Although the CIP is starting the development of the dynamic web services for CRED, the intention
is to allow multiple institutions to help construct the site, as well as ensure that CRED retains full
ownership of the site and eventually the means and ability to further develop and maintain it after 2
years. For this reason the CIP purchased commercial web hosting from Hostway.com in order to
bypass institutional technical restraints (FTP policy, web address/URL naming conventions,
firewalls, etc.). Currently the CIP has been the only one actively developing the site, however anyone
given permission by CRED to work on the site could be given FTP access. There would only need
to be minor coordination to ensure work done by various individuals was not disruptive to existing
portions of the site or its layout.

In terms of database and server-side scripting support, the CIP chose MySQL and PHP. As freeware
anything constructed on the commercial hosting site could relatively easily be transferred to another
or local server. It is relatively easy to learn for persons with little or no programming experience, but
is extraordinarily powerful despite its ease of use. Additionally, the decision to use MySQL and PHP
reflects the desire to allow relatively easy integration with other remote databases – either pushing or
pulling data.



                                                        EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                  -4-


As a final note the CIP is also using various PHP libraries to develop dynamic content. One of these
is JPGraph; a free graphing library used to display various pie charts.

3.2. Site structure

With the two major purposes of the site being visualization of data (EM-DAT and GLIDE), as well
as online management of the databases, the site is currently divided into two major sections:

The “mainpage” is designed to allow visitors to access charts, tables, download data, as well as sign-
up/register to receive e-mail notices and updates of GLIDE and other content. Conceivably the
“mainpage” section might be further subdivided as more features are added.

The “managers” section is where CRED can develop their databases, edit material, and send email
updates, etc. The section can be subdivided if necessary, and all material will eventually be password
protected under the “managers” section.

3.3. Basic database structure

There are actually numerous databases ensuring the various functions of the site work, and more
databases will likely be added, however regarding the EM-DAT, there are two copies. One which is
public (viewable by the visitor) and is used to produce the tables and graphing features of the site,
and the other EM-DAT database, which is a working/developing version based upon the content
and information entered by CRED.

It is intended that CRED, after having developed the content, could on a quarterly basis (or whatever
timescale) simply transfer new records to the public EM-DAT database used for visualization
purposes. All associated graphs, tables, etc would automatically reflect the update. Batch/automated
process to transfert all new records could be created in order to avoid CRED reprogramming or go
through a complicated process of mirroring databases using script commands - literally the public
version of the EM-DAT could be updated with a few clicks of the mouse.

The following graphic illustrates the basic databases available and who is capable of access:




                                                        EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                -5-




Most of the other databases should be relatively straightforward, as how to manage and update
content. For the e-mail list, CRED can simply add users the same way visitors to the site could add
their name to the list. Additionally CRED can view the entire list of subscribers and delete them as
necessary.

Following CRED’s request to keep multiple records associated with the EM-DAT (ie: multiple loss
estimates for various sources), the CIP has created a “note card” system (parent/child) in which
CRED can simply enter in basic information as it is available. Although there is a category for
GLIDE and DisNo associations, these, as well as other information, can be added later. The idea is
that multiple note cards would be assigned to one GLIDE or DisNo, however only one would be
used/selected to represent the losses viewable in the EM-DAT. The following diagram illustrates the
proposed process.




                                                      EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                  -6-




4. Recommendations
4.1. Software, Hosting, Web based

It has be decided to use MySQL as the software.

CRED will stay with the commercial host at least for the development stage.

The data entry will be be made online (PHP, HTML, web access and security). It has been agreed
upon - automation software to download the data base to local CRED computer in case the internet
connection is cut and in order to let CRED contine to work with the database.

4.2. Database management

The actual data entry form structure (flat database) has to be modified and a new data entry flow will
be designed in order to make the database relational. Different technical solutions have been
pointed out to:
- relation “child” forms and “composite” form (primary and secondary sources)
- allow more than 1 country/region/event/location/year entry
- allow internal validation of data (an extra programme will be developed by CRED)
- add more extra fields (developed by CRED)




                                                        EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                                       -7-



                                       D a ta e n try flo w
                                                   P R IM A R Y
         E vent
                                          S o u rc e
                                            S o u rc e
                                               S o u rc e                                    u n t tr
                                                                                      CCoo u n r yy
         G L ID E (sa n s IS O )                                                          C o u n try
                                                             + C o u n try              ( o nn it iuee t y I O
                                                                                      ( cc oCt onnnntn ,t ,I SS O, ,r e g i o n ,
                                                                                                           r
         D is a s te r ty p e                                   + C o u n try             (c o n tin e n t, IS O , re g io n ,
                                          D i s a s t e r s u b s e+ C o u n t r y
                                                                   t                                         euu r eScc ,   y)
                                                                                        r e g( icoonn, t i nccnrtr, eInnO y )r e g i o n ,
         (d is a s te r g ro u p )           D is a s te r s u b s e t                                          c u rre n c y )
                                          n a m ei s a s t e r s u b s e t
                                                D                                       l c a t i n / / d m nn r r
                                                                                      l oo c a itoo n aa d mici u 11e n c y )
         (d a ta e n try d a te )            nam e                                        lo c a tio n /a d m in 1
                                                nam e                                                                 Q
                                                                                                             nI/ a d Q nt t1.
                                                                                              l o c a t i o I FF EEm iee cc .
         E U /O F D A                                                                                           IF E Q e tc .
                                                                                                                   I l l e
                                                                                                                     a t/t Qn g t
                                                                                                                    lF a El/oo n g c .
                         + so u rc e                                                                                   l a t/ l o n g
                                                                                      k ik li l l
                                                                                           l                               l a t/ l o n g
                                                                                                    l
                                                                                              kr i ld
                                                                                             u e
                                                                                      i n ijn j u r e d
                                                                                           ff n e r d
                                                                                                    u
                                                                                      a fa eifcetjc tdee d
  E vent                                                                                         f est
                                                                                              a efle c se d
                                                                                      h o m m e le s s
                                                                                          ho
  S o u r c e e n tr y                                                                        hom
                                                                                      d a m mgae eel e$s sl o lc acla l
                                                                                          da a g$ € € o
                                                                                              d             e d€ o
                                                                                      y e aeraarmoangtn t/$ /a yal y c a l
  S o u rc e                                                                              y /m m o h h d
                                                                                                    /
                                                                                                      /m o n t d e
                                                                                              yle a rG M T t hi/m a y
                                                                                      l o lc ac a l/ / G M T t i m e
                                                                                            o
  D is a s te r ty p e                                                                        lo a l /
                                                                                      c ocm mcmnetn tG M T t i m e
                                                                                             ome s s
                                                                                              c o m m e n ts
  D is a s te r s u b s e t
  C o u n tr y e n tr y
  C o u n tr y                                         D IS A S T E R


4.3 Data visualization

       Incorporate user friendly functions to graphs and charts – image, print, raw data, pdf version;
       Add new functions to current dynamic graphs – graph by region, by multiple year, type,
        country;
       Work on new ways to visualize the data – new charts and tables;
       Mapping – Basic mapping that exists currently, automation;
       Mapping – Come up with new ways to map data;
       View by Glide – 1 event

4.4 GLIDE

4.4.1. Policy issue

GLIDE and TAG have overlap but are not identical. The next meetings should be held in the same
time and venue (initiated by CRED and partner) but has to be distinguished one from the other.
CRED will notify of next TAG meeting and distribute the information. GLIDE meeting will be
held during a special day before/after the TAG meeting but will not be part of the TAG.

The GLIDE partner group is constituted by : ADRC, CRED, NOAA, OFDA, World Bank, UN-
OCHA/ReliefWeb, FAO, Columbia University + invited experts. The core group must provide
value added data. Everybody could be a partners but only those who bring an added value and
quality could become a new core partner




                                                                                EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                 -8-


4.4.2. GLIDE database, generation and mailing list (CRED – CIP)

The GLIDE database base will be linked to EM-DAT updated by CRED. It will be owned by the
core group, but managed « gate keeped » by CRED.

The fields of the GLIDE database will be :

          Event NAME
          Abstract (slightly more than present email weekly list description)
          Data entry info – who and when ?

The E-mail list will be self maintained – but CRED will be in charge of it.

Different options will be added :
- possibility of receiving the list weekly or all in real time
- possiblity of browsing and searching for GLIDE number
- place the 20 last GLIDE number on the entry page
- possibility of viewing the GLIDE number by year or by country

There will be a trial period for multi partner GLIDE generation (first CRED and ReliefWeb).

4.4.3. GlideNumber.net (ADRC)

The actual web-site owned, developed and maintained by ADRC will be « gate keep » by them. The
hosting question (keep or not the GlideNumber.net on ADRC server or put it on a commercial site)
is not an issue now.

The GlideNumber.net web-site will be divided in 3 parts with different parts :

       (1) Public access

              general information about GLIDE
              how to use/how to join GLIDE partner group
              browse GLIDE numbers
              GLIDE events and demos
              contacts

       (2) Semi – public

            steering committee
            upcoming meetings

       (3) Private

            GLIDE generation
            draft documents
            « Sandbox » geeky things




                                                       EM-DAT Technical Mtg- Report - July 2002 - Draft
                                               -9-


4.4.4. Promotion

ADRC and ReliefWeb will jointly collaborate on making a brochure on GLIDE, with the approval
of the core group on the text to be included and also on produce slides/powerpoint summary.

ADRC has to source funding for brochure.

ADRC is sent instructions on how to access GLIDE data on other core member sites.

NOAA will incorporate changes into the generator and be road tested by ReliefWeb/CRED.

ADRC will link that into glidenumber.net (liaise with NOAA how it looks)




                                                     EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                       - 10 -



5. Time and tasks
                                                                A   S O N D J       F M A M J        J
CRED
Comment on prototype screens
Decide on additional fields/categories from comments
Prepare dummy tables, graphs, maps, trend lines
Finalise screen/forms
Unpacking comments
Multi-country/year record re-entry
Send data without multi-country/year records
Meeting NOAA/CRED
Validation in mySQL NOAA 1975-2001 data PHP
Link to em-dat.net graph / profiles
4th EM-DAT TAG meeting
GLIDE Generator road tested
Send slides/powerpoint to ADRC
Send to ADRC how to access GLIDE data
Work flow

NOAA
Emdat – dummy prototype screens/forms
Finalise screen/forms/composite
Design code by NOAA
Retrofitting (without multi-country/year records)
Meeting CRED /NOAA
ERS for CRED /populate JB (pdf)
Mapping initiation
Create account em-dat.net for CRED
Validation methodology/coding
Visualisation additions/corrections
Co-ordinate changes to the generator (ADRC, RW, CRED)
Approve brochure text & produce slides/powerpoint
Send to ADRC how to access GLIDE data
Send instructions for ERS/PDF creation to JB

ADRC
Link GLIDE Generator into glidenumber.net
Source funding for brochure
Collect slides/powerpoint and produce summary
Brochure final package preparation
GLIDE LOGO

ReliefWeb
GLIDE Generator road tested
Send slides/powerpoint to ADRC
GLIDE LOGO
Send to ADRC how to access GLIDE data


                        OFDA
Send slides/powerpoint to ADRC
Lead promotion of rebirth of EMDAT
GLIDE LOGO
Send to ADRC how to access GLIDE data




                                                                EM-DAT Technical Mtg- Report - July 2002 - Draft
                                           - 11 -


Annex 1 - Agenda

                EM-DAT Interim Technical Meeting

               Centre for Research on the Epidemiology of Disasters - CRED
                                Brussels, 8-9th of July 2002
            _______________________________________________________

  MONDAY July 8, 2002

    9.00 am to 1.00 pm : NOAA prototype (1) web based relational database

     Presentation and detailed discussion on the development of a dynamic website based on EM-
     DAT:

     1. Data management (relational, online database structure and access)
     2. Dynamic data visualization (searching, mapping, graphing etc.)
     3. Integration (dynamic linking of EM-DAT with Glide # and other databases – ReliefWeb,
        CIP and others with EM-DAT and Glide#)

     Discussion of compatibility with Reliefweb, appropriateness of environment and languages,
     capacity of CRED to continue development, mechanisms of debugging, updating (cf: data
     entry screens)

           Output: modifications to the prototype for finalisation


     2.00 pm to 5.30 pm: Modification forseen in EM-DAT

     1.     New variables
     2.     Generation of Glide
     3.     Conflict/Complex emergencies
     4.     Internal validity check

           Output : modifications required for data entry screen

  TUESDAY July 9, 2002

    9.00 am to 1.00pm: NOAA prototype (2) search engine (additional issues)

    Where do we go from here? Discussion about:

          1. MySQL vs ACCESS
          2. Commercial hosting and server restrictions
          3. Online data entry (PHP and HTML), web access and security




                                                    EM-DAT Technical Mtg- Report - July 2002 - Draft
                                      - 12 -


Detailed discussion of the mechanics of the search engine, programming simplifications,
outputs reports based on Reliefweb experience, dynamic aspects of EMDAT website, how
CRED would carry it on, debugging, updating.

         Output: modifications for search engine (next steps for CRED and CIP)

2.00pm to 5.00 pm: Summary of modifications and tasks


    1. Overall discussion of programming issues
    2. Time table and attribution of responsibilities
    3. What can be achieved for the next TAG-Meeting

         Output: Summary who does what, when




                                               EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                    - 13 -


                                  Annex 2 - List of Participants

                 EM-DAT Interim Technical Meeting
                      Centre for Research on the Epidemiology of Disasters
                                   Brussels, 8-9th of July 2002
                _______________________________________________________



Masaru Arakida
Asian Disaster Reduction Centre (ADRC)
IHD Bldg 1-5-1
Wakihamakaigan-dori
Chuo-ku
Kobe 651-0073, Japan
Telephone: (81-78) 230 03 46 - Telefax: (81-78) 230 03 47
E-mail: arakida@adrc.or.jp

Regina Below
Documentalist
Centre for Research on the Epidemiology of Disasters (CRED)
Ecole de Santé Publique
Université Catholique de Louvain
30.94 Clos Chapelle-aux-Champs
1200 Brussels, Belgium
Telephone: (32-2) 764 33 27 - Telefax: (32-2) 764 34 41
E-mail: below@epid.ucl.ac.be

Craig Duncan
Reliefweb
Office for the Coordination of Humanitarian Affairs United Nations (OCHA)
Palais des Nations
1211 Geneva 10, Switzerland
Telephone: (41-22) 917 1234 - Telefax: (41-22) 917 0023
E-mail: duncanc@un.org

Jean-Bernard GERMAUX
Centre for Research on the Epidemiology of Disasters (CRED)
Ecole de Santé Publique
Université Catholique de Louvain
30.94 Clos Chapelle-aux-Champs
1200 Brussels, Belgium
Telephone: (32-2) 764 33 69
E-mail: germaux@epid.ucl.ac.be

Philippe Hoyois
Centre for Research on the Epidemiology of Disasters (CRED)
Ecole de Santé Publique
Université Catholique de Louvain
30.94 Clos Chapelle-aux-Champs
1200 Brussels, Belgium
Telephone: (32-2) 764 33 66 - Telefax: (32-2) 764 34 41


                                                             EM-DAT Technical Mtg- Report - July 2002 - Draft
                                                   - 14 -

E-mail: Philippe.Hoyois@epid.ucl.ac.be

Debarati Guha-Sapir
Director
Centre for Research on the Epidemiology of Disasters (CRED)
Ecole de Santé Publique
Université Catholique de Louvain
30.94 Clos Chapelle-aux-Champs
1200 Brussels, Belgium
Telephone (32-2) 764 3327 - Telefax: (32-2) 764 3441
E-mail: sapir@epid.ucl.ac.be

Aurelia Micko
Climate Information Project
Office of Global Programs
National Oceanic and Atmospheric Administration
1100 Wayne Ave. Suite 1210
Silver Spring, MD 20910, USA
Telephone: (1-301) 427 2089 - Telefax : (1-301) 4272082
E-mail: aurelia.micko@noaa.gov

Geoffrey Rowland
Technical Officer, Climate Information Project
Office of Global Programs
National Oceanic and Atmospheric Administration
1100 Wayne Ave. Suite 1210
Silver Spring, MD 20910 - USA
Telephone: (1-301) 427 20 89 x-186 - Telefax: (1-301) 427 20 82
E-mail: geoffrey.rowland@noaa.gov

Darren Shaw
CRED Advisor
Centre for Tropical Veterinary Medicine (CTVM)
Royal (Dick) School of Veterinary Studies
The University of Edinburgh
Easter Bush, Roslin, Midlothian
Scotland - EH25 9RG - United Kingdom
Telephone: (44-131) 650 7573 - Telefax: (44-131) 445 5099
E-mail: Darren.Shaw@ed.ac.uk

Nate Smith
OFDA/USAID
U.S. Agency for International Development
Office of U.S. Foreign Disaster Assistance
Ronald Reagan Building
Washington D.C. 20523-8602, USA
Telephone: (1-202) 661 9321 - Telefax: (1-202) 216 37 06
E-mail: nsmith@ofda.net – nasmith@usaid.gov




                                                            EM-DAT Technical Mtg- Report - July 2002 - Draft

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:6
posted:6/10/2012
language:English
pages:14