WaM-DaM: A Data Model to Synthesize and Organize Water Management Data Adel M. Abdallah & David E. Rosenberg 2014 AWRA Spring Specialty Conference GIS & Water Resources VIII - Data to Decisions May 12-14 Snowbird, UT 1 Water Management Data Model (WaM-DaM) 1. Why Do We Need a Standard? 2. Design Methods 3. WaM-DaM A Proposed Method to Organize Network-Based Water Schema Management Data 4. Results 5. Conclusions WaM-DaM Model quicker. Publish faster. 2 Why Do We Need a Standard? • We use different data sources and models to manage water resources • Scientists and managers spend up to 75% of their time to build models Supply Demand 3 Different Sources and Descriptions Lake Mead e.g., outflow Hoover Dam e.g., release Reservoir? Water body? HydroDesktop 4 Different Model Needs Reservoir Simulation (ResSim) Simulates reservoir operation and management Requires operational data (e.g., gates and release rules), river network, diversions, etc. Water Evaluation and Planning system (WEAP) Allocates water to meet basin demands Requires supplies, delivery requirements, priorities, allocation priorities, etc.. 5 Objective Design a data model/standard to organize, share, and publish water management data Integrate data from different sources Support consistent metadata descriptions Open-source, generic, programming language and technology independent 6 Needed features for a standard to organize network-based water management data # Feature Example data models that support it fully or partially 1 Generic and open source information model Observations Data Model (ODM) 2 Create dynamic networks from objects and instances ArcHydro, WEAP, HydroPlatform, and, HEC-DSS (Data System Storage) 3 Enforce controlled vocabulary Observations Data Model (ODM), ArcHydro, 4 Impose descriptive and explicit metadata Observations Data Model (ODM) 5 Support multiple data formats (time series, tabular, text, parameters) HEC-DSS and HydroPlatform 7 Design Methods 1. Identify the essential data for water management. 84 fields 2. Represent these fields in tables according to Relational Model Theory and Jim Gray’s rule. 58 tables Categories of questions That WaM-DaM will answer: i) Build a network ii) Compare datasets iii) Query data (1) Project metadata Objects and attributes (4) Data Formats Text: Release rules (2) Network metadata Instances and scenarios (3) Attribute metadata Variables, methods, sources, time, and units, [Observations Data Model (ODM)] 8 WaM-DaM 0.2 Schema People PK PersonID Samples PK SampleID PoliticalJurisdictionalNodeInstace PK PoliticalJurisdictionNodeInstaceID Project Metadata FK1 NodeInstanceID FK2 PoliticalJurisdictionID Controlled Vocabulary Description FK1 PersonID Definition Definition NodeObjectCategory PK NodeObjectCategoryID NodeObjectNameCV PK NodeObjectName Definition FK2 NodeAttributeCategotyCV FK1 ParentNodeAttributeCategoryID FK3 NodeObjectID FK1 AttributeID FK2 CategotyID Attributes PK AttributeID FK2 LinkObjectNameCV Description Color Shape LinkObjectCode ProjectID FK1 LinkObjectCategoryID FK1 AttributeNameCV Description AttributeCode NoDataValue LinkObjectAttributes PK LinkObjectAttributeID LinkObjectNameCV PK LinkObjectName Definition LinkObjectCategory PK LinkObjectCategoryID FK2 LinkObjectCategoryCV FK1 ParentLinkObjectCategoryID Definition NodeInstanceName Description NodeInstanceCode ExistingOrProposed Longitude Latitude LocalX LocalY FK1 NetworkID FK3 SpatialReferenceID FK2 NodeObjectID FK1 AttributesID FK3 LinkObjectID FK2 LinkCategotyID AttributeNameCV PK AttributeName Definition ObjectName Networks PK NetworkID NetworkName Description DateCreated FK1 ProjectID LinkInstances PK LinkInstanceID LinkInstanceName LinkInstanceCode Length_m Description ExistingOrProposed FK3 StartNodeInstanceID FK4 EndNodeInstanceID FK2 NetworkID FK1 LinkObjectID SourceName SourceWebpage Citation FK2 OrganizationID FK1 ModelID FK3 SampleID ScenarioNodeInstanceAttributeData PK ScenarioNodeInstanceAttributeDataID Scenarios PK ScenarioID ScenarioName Abbreviation TimeHorizon ReferenceStudyName FK1 NetworkID Description DataTypeCV PK DataType FK1 AffiliationID FK2 SourceID Methods PK MethodID Models PK ModelID MethodName MethodLink MethodDescription FK2 OrganizationID FK3 PersonID FK1 MethodTypeCV UnitType UnitName UnitAbbreviation TextValueControlledCV PK TextValueControlled Affiliations PK AffiliationID AffiliationStartDate AffiliationEndDate Phone Email FK1 OrganizationID FK2 PersonID Webpage Address MethodTypeCV PK MethodType Definition SiteLatitude ODMSiteName SiteLongitude IsRegular TimeSupport UTCOffset FK1 TimeUnitID Definition TextDescriptorControlledData PK TextControlledID AttributeMetaData PK AttributeMetadataID FK2 AttributesID FK1 AttributeFormatCV FK5 UnitID FK3 DataTypeCV FK4 MethodID Definition FK2 TextValueControlled FK1 AttributeMetadataID CensorCodeCV PK CensorCode Definition Data Values Storage TimeSeriesData PK TimeSeriesID LocalDateTime DateTimeUTC Value FK2 CensorCode FK3 TimeSereisMetadataID FK1 AttributeMetadataID ColumnsData PK ColumnID AttributeFormatCV PK AttributeFormat FK1 AttributeMetadataID TextDescriptorFreeData PK TextFreeID Definition ParameterData PK ParameterID ScenarioLinkInstanceAttributeData PK ScenarioLinkInstanceAttributeDataID FK1 ScenarioID FK2 LinkInstanceID FK3 AttributeMetadataID PersonFirstName PersonMiddleName PersonLastName TimeSeriesMetadata PK TimeSereisMetadataID Units PK UnitID SourceOriginOrder FK1 AttributeMetadataID FK2 SourceID Definition FK1 LinkAttributeCategotyCV ParentLinkAttributeCategoryID Definition Attribute Metadata AffiliationSources PK AffiliationSource ModelName Citation ModelMainPurpose SoftwareUsed Description ModelDateCreated FK3 NodeInstanceID FK1 ScenarioID FK2 AttributeMetadataID Network Metadata LinkAttributeCategory PK LinkCategotyID LinkAttributeCategoryCV PK LinkAttributeCategoty SampleMediumCV OrganizationTypeCV PK SampleMedium FK1 OrganizationTypeCV PK OrganizationTypeCV OrganizationCode Definition OrganizationName Definition OrganizationLink OrganizationDescription ParentOrganizationID SourceData PK SouceDataID NodeObjectAttributes PK NodeObjectAttributeID LinkObjects PK LinkObjectID LinkObjectCategoryCV PK LinkObjectCategory NodeInstances PK NodeInstanceID JurisdictionName State County Country AgreementName City Sources PK SourceID NodeObjects PK NodeObjectID FK2 NodeObjectNameCV Description Color Shape NodeObjectCode FK3 ProjectID FK1 NodeObjectCategoryID Projects PK ProjectID ProjectName Description Purpose Domain DateCreated Status FK1 UserID SRSID IsGeographic Notes NodeAttributeCategory PK NodeAttributeCategotyID FK2 NodeObjectCategoryCV FK1 ParentNodeObjectCategoryID Definition FK2 SampleTypeCV LabSampleCode FK1 SampleMediumCV PoliticalJurisdictions PK PoliticalJurisdictionID NodeAttributeCategoryCV SpatialReference NodeObjectCategoryCV PK NodeAttributeCategoty PK SpatialReferenceID PK NodeObjectCategory Users PK UserID SampleType PK SampleType Organization PK OrganizationID DateCollected Assumptions ParameterValue Description FK1 AttributeMetadataID BinaryData PK BinaryID BinaryValue ValueMeaning FK1 AttributeMetadataID TextValue Description FK1 AttributeMetadataID MultiColumnsData PK MultiColumnID FK2 ColumnID FK1 AttributeMetadataID FileBasedData PK FileBasedID SeasonalParameterData PK SeasonalParameterID SeasonName Value Description FK1 AttributeMetadataID FileName FileType Description FileLocationOnDesk FileWebpage FK1 AttributeMetadataID MultiColumnValues PK MultiColumnValueID Value ValueOrder FK1 MultiColumnID 9 Load Different Data to WaM-DaM US Dams dataset 23 attributes 8,121 instances Time Series data 32 attributes Water Management Data Model (WaM-DaM) US Water Bodies and Wetlands Dataset 15 attributes 26,872 instances WEAP Model Lower Bear River, UT 10 53 instances Represent the Little Bear River Network, Utah in WaM-DaM Site node object River link object Reservoir node object Site node object 11 Results What are the attributes for Hyrum Reservoir and their units? Dam, Reservoir, water body, lake Object name Node instance name Attribute format Attribute name Unit name Organize multiple data formats and maintain consistent metadata 12 Results What are the reservoir attributes, values, units, and sources? MAX_STOR and Total Capacity Node instance name Attribute name Value Unit name Source Integrate data sources and support explicit descriptive metadata Incorporate uncertainty in models Foster integrated understanding of systems data 13 Future Work • Finalize the data model design and test it with larger networks and national datasets • Integrate WaM-DaM with Hydra software to visualize and edit networks • Automate data loading, retrieval, and use WaM-DaM to populate and run models Online Published Data (e.g., Bureau of Reclamation) Prior Models Outputs (e.g., WEAP, HEC-ResSim) Discover Existing Data Files from Stakeholders (e.g., cities and operators) GAMS Transform and Organize Data and Introduce Controlled Vocabulary Water Management Data Model (WaM-DaM) Retrieve and Transform and Publish Data in Desired Format and Units WEAP CUAHSI HIS Other Models 14 Conclusions • Propose WaM-DaM as a method to organize networkbased water management data • Organize multiple data formats like time series, text, multi-column, and parameters from different sources • Foster integrated analysis and understandings of water systems • Future work will automate the process to discover, transform, and publish of data plus populate models 15 Acknowledgement http://ci-water.org/ This project is funded through EPS – 1135482 and EPS – 1135483. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation 16 Thank you! Questions? For more info: Adel Abdallah [email protected] http://www.engr.usu.edu/cee/faculty/derosenberg/students.htm WaM-DaM Model quicker. Publish faster. 17
© Copyright 2024 ExpyDoc
