SIB-ESS-C - A SPATIAL DATA INFRASTRUCTURE TO FACILITATE EARTH SYSTEM SCIENCE IN SIBERIA

Transcription

1 SIB-ESS-C - A SPATIAL DATA INFRASTRUCTURE TO FACILITATE EARTH SYSTEM SCIENCE IN SIBERIA Roman Gerlach, Christiane Schmullius, Sören Hese Institute for Geography, Dept. Earth Observation, Friedrich-Schiller-University, Grietgasse 6, Jena, Germany roman.gerlach@uni-jena.de ABSTRACT The Siberian Earth System Science Cluster (SIB-ESS- C) currently being established at the University of Jena (Germany) will be a Spatial Data Infrastructure (SDI) for remote sensing product generation, data dissemination and scientific data analysis to support Earth system science in Siberia. The initial set of products has been derived as part of the EU funded project (EVG ). These products cover a 300 Million ha region in central Siberia comprising maps of land cover, fire induced disturbances, phenology, snow depth, snow melt date, onset and duration of freeze and thaw, among others. The study region represents a significant part of the Earth s boreal biome and is believed to play a critical role in global climate change in Northern Eurasia. Hence, a major goal of SIB-ESS-C is to continue product generation in order to build up time series. SIB-ESS-C will be implemented based on standards (OGC ) and the International Organization for Standardization (ISO). 1. INTRODUCTION The concept of spatial data infrastructures (SDI) as a tool for data management, dissemination and visualization has been widely recognized and numerous initiatives (e.g. INSPIRE, GSDI, GDI-DE) emerged to implement such systems. Major drivers in the development of SDI s have been the standards (OGC ), the International Organization for Standardization (ISO) as well as the World Wide Web Consortium (W3C). Initially the prime objective of an SDI was to allow users to search for a geographic dataset utilizing a Catalog Service, access or download the data through a Web Feature/Coverage Service (WFS, WCS) and perhaps visualize it using a Web Map Service (WMS). These basic components may still form a comprehensive set for many applications, but within recent years more and more authors emphasized to incorporate also processing and analysis capabilities into SDI concepts [1][2][3][4]. Especially within the earth science community there has been strong interest to go beyond pure data sharing systems and develop distributed processing and analysis services. An active research topic is also the integration of simulation models into a distributed service architecture where single components can be connected or chained utilizing standard interfaces [4]. Based on these technical advancements the Siberian Earth System Science Cluster is being implemented. SIB-ESS-C is the follow-on activity to the EU funded project (Multi-Sensor Concepts for Greenhouse Gas Accounting of Northern Eurasia, EVG ) [5][6][7]. was a joint Russian-European remote sensing project that improved greenhouse gas accounting over a 300 Million ha area in the central Siberian region [8]. This area represents a significant part of the Earth s boreal biome which plays a critical role in global climate change and has been defined as one of IGBP s boreal transects representing a strong climate change hot spot in northern Eurasia. The overall objective of the project was to demonstrate the viability of full carbon accounting including greenhouse gases (GHG) on a regional basis using state-of-the-art environmental methods, biosphere modelling and advanced remote sensing technologies. The tools and systems which have been employed include a selected yet spectrally and temporally diverse set of 15 Earth observation datasets from 8 satellites, detailed GIS databases and some of the worlds most advanced Dynamic Global Vegetation Models (the Lund- Potsdam-Jena LPJ-DGVM and the Sheffield-DGVM) to account for fluxes between land and atmosphere. 2. THE SIB-ESS-C REGION: FROM THE URAL TO THE PACIFIC The region (Fig. 1) stretched from the Taymyr Peninsula in the north of Siberia to lake Baikal including the administrative entities of the Krasnoyarsk Kray, Irkutsk Oblast, Taymyr and Evenk Okrug. This area in central Siberia will be extended in SIB-ESS-C to the east and west.the extended region now stretches from the Ural to the Pacific covering the Ob-, Lenaand Yenissey river systems and the far eastern federal districts including the autonomous okrug/oblast of Amur, Jewish, Kamchatka, Korya, Khabarovsk, Magadan, Chukotka, Sakha and Primorsky (maritime) (Fig. 1). From an Earth system modeling perspective this larger region is much better representing the northern boreal biome. Most biosphere models already work on this scale and with areas that are much larger than the region. Extending the area to Proc. Envisat Symposium 2007, Montreux, Switzerland April 2007 (ESA SP-636, July 2007)

2 cover the complete boreal region of northern Asia was therefore a needed extension. Figure 1: SIB-ESS-C Study (red line), Study (green line) 3. THE EARTH OBSERVATION LANDSURFACE PRODUCTS Northern Eurasia is home to several processes that are unique, greatly affected by climate change and likely to have big consequences for global climate. The objective of Earth observation was to deliver geo-observational products for monitoring and modeling the key processes. A better understanding of the above processes in turn improved the modeling approaches used in the project to address the key project scientific question: What is the current average greenhouse gas budget of the region and what is its spatial and temporal variability? How will it change under future climatic and anthropogenic impacts? To achieve the goals of the project, a diverse set of multi-sensor Earth observation data was used. The definitions of land surface products to be derived from EO data, their spatial and temporal scales have been driven by the project modeling approaches and also by their use as indicators of global change in the boreal region. Tab. 1 summarizes the main properties of the EO products derived in. Only with a multi-sensor approach could the diverse set of land surface parameters be achieved at spatial and temporal scales required by the modeling approaches for the entire project area. A more comprehensive presentation of the project can be found in [5][6][7] and [8]. For detailed product descriptions refer to [9][10][11][12][13][14]. EO Product Phenology Disturbances Source SPOT-VGT AVHRR MODIS, AVHRR ATSR-2 Table 1. Earth observation data products Temporal Spatial Spatial coverage resolution coverage annual on yearly basis 1km & 10km 1 km Freeze/ Thaw QuikSCAT km Water bodies ASAR WS 2003/ m Snow Depth SSM/I km Snow Melt SSM/I km Land cover Topography MODIS SRTM / GTOPO annual m 3arcsec<60 N I km > 60 N entire region Partner responsible Centre for Ecology and Hydrology Monks Wood, UK TU Wien, Institute of Photogrammetry and Remote Sensing (IPF),Austria TU Wien, Institute of Photogrammetry and Remote Sensing (IPF),Austria University of Wales Swansea, UK Gamma Remote Sensing, Switzerland 4. SIB-ESS-C OBJECTIVES The objectives of the SIB-ESS-C project are to develop a spatial data infrastructure to facilitate Earth system science studies in central Siberia, to set up a web interface to provide access to data products created during the project, to continue remote sensing data acquisition and product generation to build up time series for a larger region in Eurasia and to integrate additional products from other projects as well as from external collaborators. The final stage of SIB-ESS-C will provide online geo-visualization and analysis tools (including a biosphere/earth system modelling interface) for integrated data analysis.

3 5. SIB-ESS-C ARCHITECTURE The overall design philosophy of SIB-ESS-C (fig. 2) follows three major principles: a) adhere to standards to ensure interoperability, b) utilize components that are well established in the Earth Science, Earth Observation and GIS communities and c) implement free and open source software components whenever possible. The core of SIB-ESS-C will be a PC cluster providing enough processing power for high volume remote sensing data processing and complex modeling tasks. Data products created on the cluster will be stored in a product database. A GeoServer will be implemented to provide data access through OGC compliant WFS [15] and WCS [16] interfaces. Metadata will be kept in a PostgreSQL database following ISO19115 [17]. Metadata shall be retrieved through an OGC compliant Catalog Service [18]. At a later stage it is planned to set up a THREDDS Data Server [19] ensuring access and distribution of NetCDF, GRIB or HDF data sets that are widely used in the Earth science community. In order to manage the different data sources and to provide a single interface for data discovery and access the GI-Cat toolbox will be implemented [20]. All services offered by SIB-ESS-C will be embedded into a web based user interface. In the future SIB-ESS-C will be extended with comprehensive spatio-temporal analysis tools as well as model interfaces. Figure 2. The Planned Siberian Earth System Science Cluster (SIB-ESS-C) architecture. 6. SIB-ESS-C IMPLEMENTATION STRATEGY The technical implementation of SIB-ESS-C shall adhere to the following multi-stage concept: Stage 1: development of an online data repository including a metadata database and a web interface to enable users to search, (pre-)view and download existing datasets. Stage 2: set up of a computing cluster for operational processing of large quantities of remote sensing data ensuring continued product generation. The cluster will also include tools for data archiving, storage management and automatic metadata creation. Stage 3: extension of SIB-ESS-C with comprehensive interactive online geo-visualization tools through a web interface: allowing users to analyse the information content of the data sets provided (GIS functionalities, cross-comparison of data products, extraction of results using maps, graphs, text files and real data) including triggering of Earth system model runs (using biosphere models from partner organisations). The last part of stage 3 (biosphere modelling) needs the design of various model-interfaces that allow the use of Earth observation products in biosphere modelling (has been started already in ). Stage 4: following the principle of interoperability SIB- ESS-C is planned to become part of a distributed network of similar systems where not only data is

4 being distributed and shared, but also applications (e.g. analysis functionalities, processing modules) are being offered and used throughout the network. 7. SUMMARY The Siberian Earth System Science Cluster (SIB-ESS- C) currently being established will be a Spatial Data Infrastructure (SDI) for remote sensing product generation, data dissemination and scientific data analysis to support Earth system science in Siberia. SIB-ESS-C will be implemented based on standards (OGC ) and the International Organization for Standardization (ISO). SIB-ESS-C Services contain: Catalogue Service: providing meta data on products and procedures (search and find data), Coverage Service: providing direct access to datasets available from SIB-ESS-C (access and download data), Map Service: visualization of geographic datasets available from SIB-ESS-C, Analysis Service: advanced visualization tools for integrated data analysis (integration of multiple data sets, spatially and temporally), Biosphere Modelling Service based on various datasets (final stage of the SIB-ESS-C implementation) Access to data products is provided free of charge. The preliminary website address for information and news concerning SIB-ESS-C can be reached through 8. ACKNOWLEDGEMENT The Siberian Earth System Science Cluster is being funded by the Friedrich-Schiller University Jena (Germany) for the period commencing January 2006 until December Funding is granted for hard- and software as well as labour cost. 9. REFERENCES 1. Kiehle, C., Greve, K. & C. Heier. (2006). Standardized geoprocessing - taking Spatial Data Infrastructures one step further. Proceedings of the 9th AGILE Conference on Geographic Information Science, Visegrád, Hungary, Descamps, G., Therrien, P. & Therrien, R. (2006). An interactive and open approach for the analysis and diffusion of geoscientific data. Computers & Geosciences, 32(5): Baumann, P. (2006). Towards a Standard for Interoperable Earth System Raster Services, 23rd Intl. FIG Congress, 8-13 October 2006., Munich, Germany. 4. Wytzisk, A. (2003). Interoperable Geoinformations- und Simulationsdienste auf Basis internationaler Standards. PhD Thesis, Westfälischen Wilhelms-Universität Münster, Germany. (german) 5. Schmullius, C., S. Hese, H. Balzter, W. Cramer, F. Gerard, R. Kidd, T. LeToan, W. Lucht, A. Luckman, I. McCallum, S. Nilsson, A. Petrocchi, S. Plummer, S. Quegan, A. Shvidenko, L. Skinner, S. Venevsky, S. Voigt, W. Wagner, U. Wegmüller, A. Wiesmann (2002). Sensor Systems and Data Products in a Multi- Sensor Approach for Full Greenhouse Gas Accounting in Siberia, ForestSAT Conference 2002, Edinburgh, UK. 6. Schmullius, C., S. Hese, D. Knorr (2003). Siberia- II: A Multi-Sensor Approach for greenhouse Gas Accounting in Northern Eurasia, PGM (Petermanns Geographische Mitteilungen), 147, 2003/6. 7. Schmullius, C., C. Beer, R. Gerlach, S. Hese, D. Knorr, M. Santoro, L. Skinner, A. Luckman 2005: - Multi-Sensor Land Cover Products for Greenhouse Gas Accounting of Northern Eurasia, ForestSAT Conference 2005, Conference Proceedings, Umea, Sweden. 8. Beer, C., W. Lucht, C. Schmullius, & A.Shvidenko. (2006) Small net carbon dioxide uptake by Russian forests during , Geophys. Res. Lett., 33, L Delbart N., Kergoat, L., Le Toan, T., L'Hermitte, J. & G. Picard, (2005). Determination of phenological dates in boreal regions using normalized difference water index, Remote Sensing of Environment, 97, 1, pp George, C., Rowland, C., Gerard, F. & H. Balzter (2006). Retrospective mapping of burnt areas in Central Siberia using a modification of the normalised difference water index, Remote Sensing of Environment, Volume 104, Issue 3, 15 October 2006, Pages Bartsch, A., Kidd, R.A., Wagner, W. & Bartalis, Z., (2007). Temporal and spatial variability of the beginning and end of daily spring freeze/thaw cycles derived from scatterometer data. Remote Sensing of Environment, 106(3): Bartsch, A., Kidd, R., Pathe, C., Shvidenko, A. & W. Wagner (2004). Identification of wetlands in central Siberia with ENVISAT ASAR WS data. Proceedings ENVISAT Symposium, Salzburg. 13. Grippa, M., Mognard, N.M., Le Toan, T., & E.G. Josberger (2004). Siberia snow depth climatology derived from SSM/I data using a combined

5 dynamic and static algorithm, Remote Sensing of Environment (93): Skinner, L., & Luckman, A. (2004). Introducing a land cover map of Siberia derived from MERIS and MODIS data. Proceedings of IGARSS'04, Anchorage, September, pp OpenGIS Consortium, Inc., (2004). Web Feature Service (WFS) Implementation Specification, Version 1.1, OGC OpenGIS Consortium, Inc., (2006). Web Coverage Service (WCS), Version 1.0.0, OGC r OpenGIS Consortium, Inc., (2007). Catalogue Service Implementation Specification, Version 2.0.2, OGC r1 18. ISO/IEC 19115:2003, Geographic information Metadata, ISO International Standard, Geneva Domenico, B., Caron, J., Davis, E., Kambic, R. & Nativi, S., (2002). Thematic Real-time Environmental Distributed Data Services (THREDDS): Incorporating Interactive Analysis Tools into NSDL. Journal of Digital Information, 2(4): Article No Bigagli, L., Nativi, S., Mazzetti, P., & Villoresi, G. (2004). GI-Cat: a Web Service for Dataset Cataloguing Based on ISO 19115, Proc. of 1st InternationalWorkshop on Geographic Information Management (GIM 04) 15th International Workshop on Database and Expert Systems Applications (DEXA 04), IEEE Computer Society Press, Saragozza, Spain, 30 August 3 September 2004, pp , 2004.