1
DESIGN OF AN UML CONCEPTUAL MODEL AND IMPLEMENTATION OF A GIS WITH METADATA INFORMATION FOR A SEISMIC HAZARD ASSESSMENT COOPERATIVE PROJECT. Abstract Y. Torres 1 , M.P. Escalante 1 and the RESIS II Working Group * 1 ETSI Topografía, Geodesia y Cartografía, Universidad Politécnica de Madrid, Spain Conclusions Central America is characterized by a high seismic activity due to its tectonic setting, controlled by its location in the vicinity of a convergent plate margin. In order to understand more deeply this aspect, a seismic hazard assessment is carried out (Benito et al) in the frame of the cooperative project RESIS II . This work illustrates the advantages of using a Geographic Information System (GIS) in such a big project, with researchers of different countries, when input information presents diverse formats. To achieve data homogenization and to integrate them in the GIS UML Conceptual Model Once all information is selected and organized, a Conceptual Model is designed, which must define the project goal/s (estimate ground motion in Central America) and determine the GIS characteristics (scale, data reference system, software…). It allows the GIS designer defining model architecture and interoperability, since classes and their relations are represented. It is carried out in two phases: analysis of requisites and conceptualization. The Unified Modeling Language (UML) is used to compose the conceptual model of the GIS (Figure 3), which complies with ISO 19100 norms. GISCA Structure GISCA is implemented using ArcGIS 9.2 software. Data are classified in folders. Every layer is created indicating its reference system (WGS-84), its geometry for geographical representation (point, polyline, polygon) and its bounding box. For tables, links are defined only when necessary. The structure is divided in two distinct parts: Data Base and Atlas (Figure 4). The GIS provides a frame for the combination of large geographic-based data volumes, with an uniform geographic reference and avoiding duplications. GISCA Structure ATLAS folders Atlas folders contain calculations, analyses and query results. They are used to generate maps that show the results. DATA BASE folders Data Base folders store input the information needed to calculate seismic hazard: seismic catalogue, tectonic, source models, recurrence models, attenuation laws. At this point of the analysis, hazard results are a mesh of values within GISCA. Metadata Information References All data managed within GISCA contain their own metadata following ISO 19115 normative (Figure 5). Metadata give “information about data”, such attributes, data about records, location, ownership… Metadata may include descriptive information about the context, quality, lineage or characteristics of the data. As RESIS II is a cooperative project and all information is shared between contributing users and researchers, metadata is a great support for specialists of different countries to know deeply the data they manage, what make easier their teamwork. In a cooperative project context it is important to establish a contact point where every researcher may demand data, supply information to other members of the working group and make partial and final results accessible. The GISCA manager, who centralizes information transmission during the development of the study, is such a contact point for the RESIS II project . GISCA fulfils its objectives, since it helps the GIS specialist to follow and manage the different project phases: to select and catalogue heterogeneous input data, to combine them in a single frame, to analyze them and make queries and calculations, and finally, to estimate the seismic ground motion. GISCA is really helpful to integrate in the same environment active faults and subduction slabs geometries, which combined with the epicentres location, facilitates the definition of seismogenetic regions. The GIS capacity for making queries (by location and by attributes) and geostatistical analyses is used to The GISCA operator manages input data following conceptual model indications about relations between elements from distinct defined classes to generate additional layers with the aim of getting RESIS II final objective. All these processes results are detailed in Figure 6 and a sample of them is given in Figure 7. Central America Tectonics Data Collection As collected input data present different formats (raster images, vectorial data, tables, charts…), cover distinct geographical areas and are subjected to different interpretations (Figure 2), data inconsistencies may appear and their management get complicated. To avoid this, the first step of the study was to examine, classify and homogenize the original information. 1 2 3 4 Metadata Information Metadata language: Spanish Last update: 20090401 Metadata contact: Individual's name: Yolanda Torres Fdez Organization's name: Grupo de Ingeniería Sísmica de la Universidad Politécnica de Madrid (UPM) Contact's position: Técnico de SIG Phone: 913366441 Delivery point: ETSI Topografía, Geodesia y Cartografía. Av. Valencia, km 7,5. Despacho 326 City: Madrid Postal code: 28031 Country: Spain e-mail address: ytorres@topografía.upm.es Name of the metadata standard used: ISO 19115 Geographic Information - Metadata Version of the metadata standard: DIS_ESRI1.0 Resource Identification Information: Reference date: 20090315 Edition date: 20090328 Presentation format: digital map Place keywords: Centroamérica, Costa Rica Discipline keywords: Geofísica, Sismología, Tectónica, Física, Geología, Ciencias de la Tierra, Estadística, Geodinámica Thesaurus name: AGROVOC, UNESCO Resource's bounding rectangle: *West longitude: -85.911392 *East longitude: -82.561392 *North latitude: 11.212669 *South latitude: 8.025669 Spatial Representation - Grid: Number of dimensions: 2 Dimension size (x-axis): 3350 Dimension size (y-axis): 3187 Resolution: 0,001 Units of measure: Degree Lineage statement: De un cálculo previo externo al SIG se creó una malla de puntos equiespaciados 0,1º de longitud y latitud. Estos puntos tienen asociado un valor de PGA (Aceleración Pico del Suelo), es decir, un atributo en su tabla que recoge dicho valor. Mediante una operación de análisis geoestadístico (KRIGING) se pasó esta información discreta a continua, generando así la capa CostaRica500 y se guardó como ráster en formato GRID de ESRI. Abstract: Esta capa ráster contiene información continua que representa el valor de PGA (aceleración pico del suelo) para un periodo de retorno de 500 años para todo el territorio nacional de Costa Rica. El tamaño de las celdas (píxel) es 0,0010º x 0,0010º. Su valor mínimo es 204,567 gales y su valor máximo es 601,062 gales. CostaRica500 Data format: Raster Dataset Coordinate system: GCS_WGS_1984 Theme keywords: Ciencias de la tierra, Terremotos, Desastres, Desastres naturales, Movimiento del suelo, Temblores de tierra, Sismo, Zona sísmica, Prevención antisísmica, Sismicidad, Movimiento del suelo, Aceleración Pico, Periodo de Retorno whit Metadata information, it is convenient to develop a Conceptual Model previously with the aim of helping the GIS designer taking the project general idea. Figure 1 illustrates the seismic hazard assessment phases followed: data collection and homogenization, conceptual model design, implementation of a GIS and metadata creation, seismic hazard estimation and results presentation. Data and processes in violet forms are managed or developed in the GISCA (GIS of Central America). Some of them are presented in this poster. Figure 1. Seismic Hazard Assessment flow diagram Figure 2. Sample of study input data. Previous zonations for Cental America National Seismic Catalogues Figure 3. Seismic Hazard Assessment Conceptual Model. Figure 4. GISCA folders organization. ArcGIS view. Figure 5. Sample of a GISCA layer metadata. Figure 7. Sample of RESIS II seismic hazard results. •Benito, B., E. Camacho, A. Climent, G. Marroquín, E. Molina, W. Rojas, E. Talavera, J.J. Escobar and G. Alvarado (2008). RESIS II Project. Seismic Hazard Assessment for Central America. Final report. 227pp. •ISO Normative. http://www.iso.org/iso/home.htm Acknowledgements The RESIS II Project is financed by the Govenment of Norway, managed by CEPREDENAC. 5 Panama Seismic Hazard in terms of PGA (gal) for the 2500-year return period. Crustal national seismic sources model 0 75 150 225 300 4.5 5 5.5 6 6.5 7 7.5 8 0.0E+00 1.0E-05 2.0E-05 3.0E-05 4.0E-05 5.0E-05 6.0E-05 7.0E-05 8.0E-05 9.0E-05 %Prob R (km) Mw San Salvador. PGA, PR 500 años 4.5 4.75 5 5.25 5.5 5.75 6 6.25 6.5 6.75 7 7.25 7.5 7.75 8 Mw a) San Salvador Hazard Deaggregation interpolate discrete data resulting from seismic hazard calculations and to create continuous maps as well as to check and validate partial results of the study. GIS-based products (complete, homogenized databases and thematic cartography of the entire region and the six Central American countries) are distributed to the participating researchers, facilitating cross-national communication, the project execution and results dissemination. Results Figure 6. Seismic hazard assessment results diagram ETSI Topografía, Geodesia y Cartografía, Universidad Politécnica de Madrid, SPAIN [email protected] + 34 913366441 CONTACT: * RESIS II Working Group for Seismic Hazard Assessment: B. Benito (Spain), C. Lindholm (Norway), E. Camacho (Panamá), A. Climent and W. Rojas (Costa Rica), G. Marroquín (El Salvador), E. Molina (Guatemala), E. Talavera (Nicaragua), J.J. Escobar (Honduras)
