Decision Support System for Integrated Lake Basin Management

Decision Support System for Integrated Lake Basin Management Mohammad Fikry Abdullah 1, Bashirah Mohd Fazli 2, Ir. Lee Hin Lee 3 & Khairul Anam Musa @ Mahmud 4 1 ational Hydraulic Research Institute of Malaysia ( AHRIM), fikry.nahrim@1govuc.gov.my 2 ational Hydraulic Research Institute of Malaysia ( AHRIM), bashirah.nahrim@1govuc.gov.my 3 ational Hydraulic Research Institute of Malaysia ( AHRIM), hllee.nahrim@1govuc.gov.my 4 ational Hydraulic Research Institute of Malaysia ( AHRIM), khairul.nahrim@1govuc.gov.my ABSTRACT Emergence of big data has created various formats of data and information that is important and substantial to be used as part of decision making process and knowledge generation. From the standpoint of Malaysia s current situation, there is an urgency to improve extensively the process of defining, assembly, analysing and disseminating knowledge and information from various sources and format as a competitive edge to implement effective control measures in the lake management. Therefore, establishing a decision support system (DSS) is a medium to binding information and knowledge together as a core or a supporting basis in managing lake to produce quality outcomes. The proposal of DSS development is integrating two parts of data which are spatial data and non-spatial data. Classification technique is used during data fusion process in this development. This classification process will associate and integrate data according area of interest. The expected output from this DSS will equipped stakeholders and lake managers with knowledge and information to understand the overall scenario of lake management issues. Thus, it will assist in making a quality, timely and responsive decisions through an innovative knowledge. Apart from that, the proposed DSS will encourage more data generation, data sharing, knowledge creation and knowledge sharing among stakeholders and lake managers. This paper intended to give a conceptual insight on the development of proposed DDS for Lake Basin governance in Malaysia to implement sustainable development and management of lakes including their basins. Keywords: Decision Support System, Knowledge Management, Spatial Data, Non-Spatial Data, Unstructured data and Structured Data. I I TRODUCTIO The challenge of having a centralised lake management system to manage water resources sustainably is one of the most pressing issues facing the world today. The scientific community, the lake manager and the public perceptions about the lakes (including reservoirs) management may differ without the data generation, sharing and disseminating of knowledge and information among the stakeholders. Consequently, the human and financial resources mobilised for the management of lakes will become futile and unproductive if the differences are not bridged in one interactive platform. Current scenario of data and lake management in Malaysia has created issues and problems among stakeholders and lake managers as follows:- a. Information and data overload that lead to data integrity; b. Information and data not reliable and not updated; c. Data accuracy and ambiguity because of poor information exchange process; d. Incomplete data and information due to difficulties in having access to sources; e. Redundant and inconsistent of data leading to information misleading; Updated, accurate, reliable, fast, understood and accepted information and data is central to any decision making in a lake basin management. Without it, institutions can be inefficient, policies can be ineffective and technology can be misapplied leading to problems such as the desiccation of the Aral Sea (Kazakhstan) or the complexity of trans boundary issues of the Great Lakes (North America) or sedimentation of Tasik Chini or severe degradation of water quality of Ringlet Dam (Cameron Highland) (ILEC 2005). Hence, the aim of this research is to propose a DSS based from the water quality modelling tools which would be able to facilitate the stakeholders of federal and state government in implementing integrated lake basin management (ILBM) at national level. The proposed DSS act as a platform to coordinate efforts in lake management and formulate the National policy on lake management for the http://www.kmice.cms.net.my/ 108

country s benefits. With the holistic and proper set up of the database institutionally, it will essentially guide in the implementation of medium and long term strategies plan for the lake sustainability. Understanding the importance of data and information from various sources, DSS for lake management will cater 2 aspects of data which are spatial and non-spatial data. Both data represent a different set of information and data that can be used as a source of decision making and knowledge generation. For non-spatial data, there are 2 subsets of data that need to consider which are structured data and unstructured data. The concept of data fusion was infused in this DSS as an approach to get a better quality of information and data as an output which is a more comprehensive and collective to support decision making process and knowledge generation. Figure 1 shows the process flow of DSS for lake data base. Figure 1: Process Flow of DSS for Lake Management. Input data for this DSS is obtained from various sources in various formats which are spatial and non-spatial data, thus prior data fusion process, classification of these data are importance. Understanding types and format of spatial and non- an overall spatial data are crucial in providing information and data. The objectives of DSS for lake management are aims to: a. To provide a centralised knowledge-based platform that collect, store, compile, analyses and disseminate data, information and knowledge to support understanding, learning and decision making; b. To acknowledge and increase the value and perception towards data, information and knowledge as an intellectual capital that exist in various formats and sources with regards to lake basin management; and c. To establish a comprehensive, collective, centralised and visualised knowledge-based platform to facilitate the management to address the issues arising from degradation with a view of developing and instituting appropriate remedial measures. The proposed DSS is intensely established on utilising and manipulating the benefits and characteristics of map to help in presenting more meaningful information in a form that is easily understood by user. Thus, by understanding characteristic of maps, web mapping which is the focal basis of the proposed DSS will lead to an interactive application. Interactivity is the key concept, as users will have a greater opportunity to explore and study any particular lakes of their interest. II CO CEPTUAL DESIG OF DD FOR LAKE MA AGEME T Spatial data consist of data that represent maps and satellite images, while non-spatial data consist of i) structured data such as sampling and monitoring data and ii) unstructured data such as pictures, reports and images. Associations between spatial and non-spatial data provide a wide-ranging of data that helps stakeholders and managers in getting more comprehensive and overview about specific lake. Table 1 shows list of spatial data potentially used in development of DSS for lake management. Each of the layers will provide a different data thus will generate information for stakeholders and lake managers. Table 1. List of Spatial Data. o Layer 1 State Boundary 2 District Boundary 3 Sub-district / City/Town Boundary 4 River Network 5 River Basin 6 Rainfall distribution 7 Ex-Mining area 8 Topography 9 Road / Transportationn 10 Land Use / Land Cover 11 Mineral Sources 12 Mining 13 Lake Statistic distribution (population, employment, education, health, prices, external trade, national 14 accounts, environment as well as data for the various sectors of the economy) 15 Utility; (Hospital, School, Clinic etc) The main features of DSS for spatial data include online information updates, overseeing data through spatial layers and ability to define and obtain information based on theme according to users preferences. The access is through GIS web publishing software to support the DSS where the system is accessible to stakeholders and interested parties through the Internet. http://www.kmice.cms.net.my/ 109

The features offered allow users to explore and comprehend lakes visually, enabling them to fathom the actual conditions and current scenarios of the lakes, their ecosystem and basins. Information provided will generously assist stakeholders to make more effective and thorough decision that is relevant and precise in handling water resources issues in Malaysia. There are two types of non-spatial data involve in this system which are structured data and unstructured data. Structured data usually contain important information and often retrieved from underlying databases and displayed using fixed templates (Zhai & Bing, 2006). Sampling data and data collection are among the activities that always store data in this format. Manipulation, usage and optimisation of structured data can easily be done via various tools and software. Table 2 shows list of non-spatial data for structured data format. Data Lake Profile Lake Water Quality Table 2. List of on-spatial Data (Structured Data). Trophic State Index Flora Fauna Socio Economy Lake Management Heavy Metals Hydrological Data Point Source Inventory Content Basic information of Lake (depth, perimeter, location, function) Sampling data, Malaysia Water Quality Index, sampling site, frequency of sampling and monitoring (date, year, time) Secchi Disk, Total Phosphorus, Chlorophyll a Inventory of distribution, types and number of species in the lake ecosystem and its catchment area Inventory of distribution, types and number of species in the lake ecosystem and its catchment area Inventory of the indigenous people, resident in the lake catchment, infrastructure, utilities The owner, manager and authority Heavy metals information (contents in lake water ) Rainfall distribution and intensity, wind data Types and number important because revenue, profitability and opportunity can go up, while risks and costs may go down (Abidin, Idris, & Husain, 2010). Table 3 shows list of non-spatial data for unstructured data format (Fikry, 2014). According to (Geetha & Mala, 2012), the importances of unstructured data are as follows:- a. Business Value; b. Better information; c. Apropos information; d. Pertinent Information; and e. More information is available to store, manage and modeled; Table 3. List of on-spatial Data (Unstructured Data). o Data o Data 1 Email 15 Proceeding 2 Report 16 Proposa Paper 3 Slaid Show 17 Journal 4 Memo 18 Technical Document 5 Newspaper 19 Video 6 Website 20 Attachment 7 Graph 21 Data set 8 Table 22 Diagram 9 Chart 23 Engineering Draw 10 Map 24 Google Earth Image 11 Photograph 25 Image 12 Legal Document 26 Satellite Image 13 Guideline 27 Report Chapter 14 Mannual Document 28 Audio Reference: Fikry (2014) By knowing and understanding the importance of unstructured data, there is a need to include this format of data as part of data input for this system. Figure 2 shows the framework of DSS for lake management. Unstructured data is understood as e-mail files, word-processing text documents, PowerPoint presentations, JPEG and GIF image files, and MPEG video files (Blumberg & Atre, 2003). Unstructured data has a non-uniform structure, and is stored as raw data that cannot be understood and processed directly by computers (Xianglong, Bo, Wei, Junwu, & Lei, 2011). Unstructured data is http://www.kmice.cms.net.my/ 110

Figure 2. Framework of DSS for Lake Management. There are various meaning of data fusion from one scientist to another. Data fusion is a combination of multiple sources to obtain improved information in a context less expensive, higher quality and more relevant information (Castanedo, 2013). Data fusion techniques combine data from multiple sensors, to achieved improved accuracy and more specific inference that could be achieved by the use of a single sensor alone (Llinas & Hall, 1998). Wald, Ranchin & Mangolini (1997) refer data fusion more to quality where it defines data fusion as a set of methods, tools and means using data coming from various sources of different nature, in order to increase the quality of the requested information. Data fusion is a formal framework in which are expressed means and tools for alliance of data originating from different sources with aims to obtain greater quality of informationn (Buchroithner, 1998) and (Wald, 1998). Regardless the various definition of data fusion, in this context, data fusion is understood as a process of combination, classification, integration and association of data from various sources and format to produce a better quality of information in area of lake management. In this study, Dasarathy s Classification system is a technique used in data fusion process. This technique is based on data-in and data-out. This is the most basic and elementary data fusion technique. In this technique, input and output data is process and the result typically more reliable or accurate (Castanedo, 2013). The data from various sources are gathered, associated and classified according to area of interest in lake management thus it will produce an output data that is more quality, comprehensive and collective. In this study, areas of interest are based on non-spatial data (Refer Table 2). Figure 3 shows classification process from data input during dataa fusion process to nine knowledge components. Figure 3. Classification Technique during Data Fusion Process to 9 Knowledge Components. The result from classification in data fusion process, input data in DSS will be classified according to knowledge component of interest in lake management perspective. Figure 4 shows data layer of DSS for lake management where each knowledge component consists of variation format of data that can support in decision making process. Figure 4. Data layer of DSS for Lake Management. III IMPLICATIO Optimisation usage of data from various format and sources give a tremendous impact in learning, knowledge sharing and decision making process. Lake managers and stakeholders able to equipped themselves with exact and rigorous knowledge pertaining to lake issues. Two types of impacts have been identified in context of harnessing data and information from this system are: - http://www.kmice.cms.net.my/ 111

a. Generation of innovative knowledge based on current data and information; b. Competitive edge for lake managers and stakeholders to tackle issues and problems Through the proposed DDS, there are an opportunity to innovate new knowledge based on current data and information that reside in the system. Knowledge created from this platform will be more comprehensive and collective where combination, merging and association of various knowledge components during data fusion process has expanding and broadening the data and information to a bigger perspective. The proposed DDS will give a competitive edge to lake managers and stakeholders to tackle issue and problems arise in lake management much earlier. Using better quality, timely and meaningful data and information from various sources will equipped users with broadening knowledge to make a good decision that might increase quality and timely decision at a lower risk and cost. DSS for lake management will benefit stakeholders and lake managers through: a. Facilitating in a sound decision making process in water resources management; b. Information exchange process to assist in efficient, sustainable and holistic approach of lake management; c. Accurate and reliable source of reference; d. Optimising research and implementation of integrated lake basin management (ILBM); e. Appropriate planning of land-use development within the lakes basin, without jeopadising the water quality and the ecosystem of the lakes; f. Avoid poor data management, lack of thorough understanding of Lake Basin ecosystem, poor information exchange locally and international. g. Identify gaps in lake management and stakeholder conflicts. h. Streamlining the research work and avoid redundancy. IV CO CLUSIO Developing a centralise DSS for managing lakes is pertinent to allow continuous knowledge generation by lake managers and stakeholders. Organising current data and information from various sources and managing it according to exact knowledge components will ensure users of the proposed DSS gained more competitive edge in dealing with lake issues. Furthermore the proposed DSS is significantly vital to facilitate stakeholders decision making process to avoid repetitive actions and plans to ensure a better financial and strategic planning in water resources management. By referring to a system which is supported by various formats of data and information, stakeholders and decision makers will have sufficient knowledge to assist them in making better and effective decision with regards to the issues of lakes including their catchment management. ACK OWLEDGME T The authors would like to thank Director General of NAHRIM, Datuk Ir. Hj. Ahmad Jamalluddin Shaaban, Director of Research Center for Hydrogeology, Dr. Saim Suratman and Director of Information Management Division, Ong Suan Ee for the advice, support and guidance rendered during the course of this research. REFERE CES Abidin, S. Z. Z., Idris, N. M., & Husain, A. H. (2010, 17-18 March 2010). Extraction and classification of unstructured data in WebPages for structured multimedia database via XML. Paper presented at the International Conference on Information Retrieval & Knowledge Management, (CAMP), 2010. Blumberg, R., & Atre, S. (2003). The Problem with Unstructured Data. DM Review, 42-46. Buchroithner, M. F. (1998). Geodata Interrelations: Inventory and Structuring Attempt of Taxonomic Diversity. Paper presented at the 2nd International Conference Fusion of Earth Data Sophia Antipolis, France. Castanedo, F. (2013). A Review of Data Fusion Techniques. The Scientific World Journal, 2013, 19. Fikry, A. M. (2014). Business Intelligent Model for Unstructured Data Management: Case Study NAHRIM. UKM, Bangi. Geetha, S., & Mala, G. S. A. (2012, 27-29 Dec. 2012). Effectual extraction of Data Relations from unstructured data. Paper presented at the Chennai 3rd International osustainable Energy and Intelligent Systems (SEISCON 2012), IET. Llinas, J., & Hall, D. L. (1998, 31 May-3 Jun 1998). An introduction to multi-sensor data fusion. Paper presented at the EEE International Symposium oncircuits and Systems, 1998. ISCAS '98. Proceedings of the 1998 I. International Lake Environment Committee Foundation (ILEC), 2005. Managing Lakes and their Basins for Sustainable Use: A Report for Lake Basin Managers and Stakeholders. Technical Report, 67-74. Wald, L. (1998). Data fusion: A conceptual approach for an efficient exploitation of remote sensing images. Paper presented at the Proceedings. Wald, L., Ranchin, T., & Mangolini, M. (1997). Fusion of satellite images of different spatial resolutions: assessing the quality of resulting images. Photogrammetric Engineering & Remote Sensing, 63(6), 691-699. Xianglong, L., Bo, L., Wei, Y., Junwu, L., & Lei, H. (2011, 26-28 Oct. 2011). AUDR: An Advanced Unstructured Data Repository. Paper presented at the 6th International Conference on Pervasive Computing and Applications (ICPCA), 2011. Zhai, Y., & Bing, L. (2006). Structured Data Extraction from the Web Based on Partial Tree Alignment. Knowledge and Data Engineering, IEEE Transactions on, 18(12), 1614-1628. http://www.kmice.cms.net.my/ 112