An environmental information system linking time series with GIS
Rodrigo Oliveira, rpo@chiron.pt
João Ribeiro da Costa, jrc@chiron.pt
Chiron, Information Systems, www.chiron.pt
Lisbon, PORTUGAL
Introduction
Environmental management decision making and, in particular, water resources planning and management require a large and diverse ensemble of data that includes time series of monitored environmental variables and geographical datasets. The properties of these two types of datasets are quite different which has conditioned the generation of distinct analytical tools and information systems for each data type. Unfortunately, while each tool is very efficient in dealing with a specific data type there is a lack of systems creating a common workspace to deal with all kinds of datasets required in environmental management.
The paper presents an environmental information system that links geographical datasets with time series of monitored environmental variables like precipitation, temperature or dissolved oxygen concentration. The system includes GIS software to manage geographical datasets and a specifically designed library, named TSF, to manage a relational database that stores the time series data.
Background
All around the world costly environmental monitoring programs are being promoted to increase our knowledge over the environment and enhance our analysis and forecast capabilities. However, without appropriate tools we will never be able to extract all the information available in our data repositories and the money spent on monitoring programs will probably not be recovered. Integrated Information systems that include storing capabilities, as well as appropriate tools for data analysis and exploration are thus needed to fully exploit our investment in monitoring programs.
A closer look at most environmental phenomena shows that we are dealing with five different dimensions: the location in space, expressed as a set of 3 coordinates - latitude, longitude and altitude -, the position in time and the particular phenomena or theme being analyzed. The goal of an environmental monitoring and information system is thus collect data in this five-dimensional world, store it and promote its use in the most efficient possible way.
This goal requires the integration of space and time, a complex field, where much research is still needed. Research proceeds in various directions, ranging from the development of the fundamental theory (Raper, Livingstone, 1995), to the visualization of 5-D processes (Hibbard et al., 1995) and the practical implementation of the basic concepts (Langram, 1993). Efforts span also the software vendors, with the implementation of multidimensional databases (Waters, 1995; Oracle, 1995).
The Information System
The core of the Information System is an ORACLE based database server, where data are stored. A complex data model named TSF (Time Series Framework) allows for the storage, management and retrieval of time-series data. 1-D geographic objects are stored inside the database, as well as 2-D objects using Esri SDE.
The system includes several general applications that enable the user to: i) manage the whole system; ii) manage and edit the stored data; iii) insert new data; iv) query and plot data as graphs or maps and v) perform spatial analysis operations.
The system is highly flexible and integrates time-series and geographic data in seamless way. At any time, the user may create a new monitoring network and specify the location of its gages or create a new gage in an existing network. The GIS window may be used to immediately visualize the new or updated network. The user also defines the environmental parameters measured at the different networks with the dimensional consistency of the parameter set being strictly enforced.
The values of any parameter stored in the system can be viewed at any moment as a time series (Figure 1), a map or as time-series of maps (Figure 2). Statistical and spatial analysis tools are available for further in-depth analysis.
Figure 1 – Time series display
Figure 2 –Precipitation surface generated automatically from the time-series data
In addition, other tools are available to deal with more specific problems such as the automatic emergency alert for pre-specified thresholds, the management of groundwater data or the computation of rating curves or the automatic verification of environmental quality standards (Figure 3).
Figure 3 – Automatic verification of environmental
quality standards
A geographic data browser was developed to allow for the storage and retrieval of geographic data. Queries can be performed by browsing a thematic tree (Figure 4) and or by providing a geographic location (Figure 5). Metadata are also stored within the system (Figure 6).
Figure 4 – Thematic tree for browsing geographic data
Figure 5 – Geographic query
Figure 6 – Metadata browsing
User experience
The system is fully operational and has been in use for 5 years. The gained experience has directed the research effort of the developing the team. Today’s main users are the Kennedy Space Center, where the system is known as MAPS (Costa et al., 1998), and the Portuguese Water Institute, where the system is known as SNIRH - National Water Resources Information System (Costa et al., 1995; Costa and Lacerda, 1995, http://www.inag.pt/snirh). SNIRH stores and disseminates all water resources data available in Portugal and supports the daily operation of the Water Institute and of the Portuguese Environmental Regional Offices. At this moment, SNIRH includes 10 monitoring networks of over 3000 gages and 200 parameters.
The system is also being used in the development of several Watershed Master Plans of major Portuguese rivers, providing data and analysis capabilities to a large and diverse team of experts. Several mathematical models where linked to the system, namely a hydrological model, a non-point source model and a water quality model.
References
Costa, J. R., Engel, B., Hinkle, R., Oliveira, R. (1998) GIS, Models and Time-Series Integration in the Kennedy Space Center. Paper presented at GisPlanet98.
Costa, J. R., Jesus, H.B., Lacerda, M. (1995) Integrating GIS and time-series analysis for water resources management in Portugal. Paper presented at HYDROGIS 1995.
Costa, J. R.& Lacerda, M. (1995) The Portuguese Water Resources Information System: using OOP to integrate time series and GIS. Paper presented at the ArcInfo Users Conference 1995.
Hibbard, W., Paul, B., Battaiola, A., Santek, D., Martinez, M.& Dyer, C. Interactive visualization of earth and space science computations. Computer, Nº 7, July 94, pp. 65-72. 1994.
Langram, G. Time in Geographic Information Systems. Taylor and Francis. 1993.
Oracle Multi-dimension. ORACLE Working Paper. 1995.
Raper, J.& Livingstone, D. Development of a geomorphological spatial model using object-oriented design. Int. J. Geographical Information Systems, 1995, Vol.9, Nª4, 359-383. 1995.
Waters, N. GIS Database Technology and Beautiful Formulae. In GIS WORLD SOURCE BOOK 1996, pp. 357-363. GIS World Books.
Authors
Rodrigo Oliveira, Civil Engineer
PhD
rpo@chiron.pt
Tel: +351-212948095
Fax: +351-212948010
Chiron, Information Systems, www.chiron.pt
Madan Park, 2825-114 Monte de Caparica
Portugal
João Ribeiro da Costa, Civil
Engineer PhD
jrc@chiron.pt
Tel: +351-212948095
Fax: +351-212948010
Chiron, Information Systems, www.chiron.pt
Madan Park, 2825-114 Monte de Caparica
Portugal