OPTIMA INCOMPC Management Board Meeting April 12 2005

OPTIMA INCO-MPC Management Board Meeting, April 1/2 2005 Izmir DDr. Kurt Fedra kurt@ess. co. at ESS Gmb. H, Austria http: //www. ess. co. at Environmental Software & Services A-2352 Gumpoldskirchen 1

WP 03: Modelling MODELS provide a • Formal • Structured • Quantitative description of the problems and possible solutions. 2

WP 03: Modelling WP 1: identifies problem issues, develops a structure for the description of the cases, identifies data needs and availability, constraints; WP 2 analyzes perceptions and preferences, institutional or regulatory frameworks, plausible socio-economic developments; WP 4 compiles the set of ALTERNATIVE WATER TECHNOLOGIES that can be used; WP 5 looks into LAND USE change as one of the major driving forces, consistent with WP 2. 3

WP 03: Modelling WP 1, 2, 4 and 5 develop the boundary conditions and specifications for • Complete • Consistent • Plausible Set of SCENARIOS for simulation modelling and optimization. 4

WP 03: Modeling Water. Ware dynamic water resources model (daily, annual) optimization Embedded models: • RRM rainfall-runoff model for subcatchments (incl. erosion estimates) with automatic calibration • STREAM water quality model Related model (optional): • LUC dynamic land use change model 5

WP 3: Modelling Models provide estimates for 1. Economic efficiency 2. Environmental compatibility 3. Equity (intra- and intergenerational) 6

WP 03: Modelling LUC: land use change model • • Discrete state (LUC) transition model Markov chain with stochastic transition probabilities Rule-based constraints and TP adjustments Temporal resolution: year, scope: decades (2050 years) Spatial resolution: ha to km 2 Resource use and pollution as land-use specific output; Possibility for external, global driving forces 7

WP 03: LUC Modelling Global/local adjustments of the transition probabilities expressed as First-order logic RULES in relative terms (INCREASE, DECREASE in %). http: //www. ess. co. at/SMART/luc. html 8

9

10

11

WP 03: LUC Modelling Interactive editors for 1. Land use classes 2. Transition probabilities 3. Modifying rules 4. Class specific resource needs/outputs are available on-line together with the viewer (player for animated results) Links from http: //www. ess. co. at/SMART will be moved to http: //ww. ess. co. at/OPTIMA 12

13

WP 03: LUC Modelling Derived values per unit area, class specific: 1. Water consumption 2. Waste water generated 3. Energy use 4. Solid waste production OTHERS ? ? 14

15

WP 03: Modelling LUC EXTENSIONS: Include transportation network in rules (connectivity) Other external variables (specified as time series) More LUC specific coefficients and processes (employment, value added, etc) 16

WP 03: Modelling LUC OBJECTIVES: 1. Hypothesis testing 2. Developing CONSISTENT scenarios with high explanatory value that can also be used directly in the rainfall-runoff basin water budget model 3. Independent estimate on water budgets 17

WP 03: Modelling RRM: rainfall-runoff model • • Dynamic, daily time step Uses daily rainfall and temperature Major basin characteristic: LAND USE (summarized from LUC scenarios ? ? ) Estimates runoff and dynamic water budget for ungaged basins, provides input for WRM start nodes (catchment) 18

19

WP 03: RRM Modelling • • • Includes automatic calibration with runoff observation data Method: Monte Carlo, evolutionary programming; Extract reliable features (Gestalt) from observations, define as constraints on model behavior, • FROM TO (period) CMIN < FEATURE < CMAX FEATURES: min, max, avg, total, values 20

21

22

WP 03: WR Modelling WRM: water resources model • • • Dynamic, daily time step Topology of NODES and REACHES Demand nodes (cities, irrigation, industry, tourism) Estimates dynamic water budget, supply/demand, reliability of supply Complete on-line implementation with editors 23

WP 03: Modelling User/scenario management: • User authentication by name and password (monitored … ) • User can see and copy ALL scenarios, edit/delete only their own ! • TEST scenarios installed as EXAMPLES to demonstrate features implemented • On-line manual pages 24

25

WP 03: Modelling Model structure: Topology (network) of NODES, connected by REACHES; NODES represent functional OBJECTS in the basin: • Sub-catchments, well(s) fields, springs • Reservoirs, structures • Water demand: cities, irrigation districts, industries, environmental uses (wetlands, minimum flow) 26

WP 03: Modelling Model structure: Topology (network) of NODES, connected by REACHES: Represent natural and man-made channels, canals, pipelines that transfer (route) water between NODES. Networks include: • Diversions (splitting the flow) • Confluences (merging flow) 27

28

29

Water demand NODES Consumptive use Intake quality constraint, conveyance loss return flow (pollution) Water demand use: 1. domestic, 2. agricultural, 3. industrial losses Costs of supply Benefits of use recycling 30

WP 03: Modelling DEMAND NODE is defined by • Its type (domestic, industrial, agricultural) • Its connectivity (upstream, downstream, aquifer) • Its water demand (time series) • Conveiance losses (evaporation, seepage) • Consumptive use fraction, resulting in • return flow, and its losses • Quality changes (pollution) • Costs of supply – Benefits of use 31

32

33

WP 03: Modelling WRM EXTENSIONS: 1. Full groundwater coupling, single or multi -cell aquifers with Darcy-flow coupling, in/exfiltration for reaches 2. Quality integration (return flow) 3. Economic analysis: 1. Water efficiency; added value/unit water 2. Cost-benefit analysis, requires, per node: Investment, lifetime, OMR, discount rate 34

WP 03: Modelling Full groundwater coupling, single or multicell aquifers with Darcy-flow coupling, in/exfiltration for reaches Every node is optionally connected to an AQUIFER OBJECT: 1. Extracting water from it (wells, infiltration (lateral inflow, baseflow contribution) into reaches, depending on relative levels 2. Returning water to it: seepage losses, explicit recharge 35

WP 03: Modeling Water Quality Modeling : STREAM • • • Uses WRM networks and results (flow scenarios) and dedicated editor; Dynamic (daily) BOD/DO, plus an arbitrary pollutant (conservative or first order decay) Input at start nodes and demand nodes: – Concentration TS – Pollutant load TS – Concentration as a piecewise linear function of flow • • Overall mass budget and compliance Dynamic Output at control nodes 36

WP 03: Modeling Water Quality Modeling : STREAM • • • Uses WRM networks and results (flow scenarios) and dedicated editor; Dynamic (daily) BOD/DO, plus an arbitrary pollutant (conservative or first order decay) Input at start nodes and demand nodes: – Concentration TS – Pollutant load TS – Concentration as a piecewise linear function of flow • • Overall mass budget and compliance Dynamic Output at control nodes 37

WP 5 -9: Modelling REMEMBER: • Model applications are THE central part of the case studies !!! • All data compilation in view of model input data requirements 38

WP 03: Model steps 1. Define the domain or system boundaries (river basin including any transfers !) 2. Describe all important OBJECTS: • Inputs = sub-catchments, wells, springs, transfers, desalination, Aquifers • Demands: cities, tourist resorts, industries, agriculture (irrigated) • Structures: reservoirs 2. Define NETWORK: link nodes through reaches (connectivity) 39

WP 03: Model steps 1. Compile and edit the DATA for the NODES and REACHES: – Time series of flow, pumping, water demand, diversion, reservoir release as rules or explicit time series, – Loss coefficients – Consumptive use fractions, – Costs (investment, OMR, and benefits per units water supplied/used; 2. Edit one or more scenarios, document 3. RUN the model, evaluate runs. 40

WP 03: OPTMIZATION steps 1. Define • CRITERIA, sort into 1. OBJECTIVES (min/max) and 2. CONSTRAINTS (inequalities), set numerical values, symbolic targets; 2. RUN the optimization model on-line (that may take a while …) 3. ANALYZE results as input to WP 14, 15 41

WP 03: OPTMIZATION steps OPTIMIZATION generates sets of feasible alternatives, each optimal in some (well defined) sense; Discrete multi-criteria methodology SELECTS a single preferred solution from that set by defining preferences and trade -offs (multi-criteria) interactively: Users explore the decision space to learn what can be obtained, and for what price (the trade-offs) and how to approach their UTOPIA solutions. 42

Work Plan (simple version) OPTIMIZATION (multi-criteria) Maximize • Supply/demand ratio (by sector) • Reliability (% time, volume), • Efficiency (GRP/unit water), • Benefit/cost ratio meeting constraints (minimum or maximum allowable levels of selected criteria), minimizing non-linear penalty functions

Work Plan (simple version) OPTIMIZATION Maximize …. by • Choice of water technologies of different costs (investment, OMR) vs performance including structures • Different allocation strategies • Selecting criteria, setting constraints

Work Plan (simple version) 3. Case Studies (WP 7 -13) • Run parallel • SHARE models • Use SAME structure for end user involvement, reporting

Work Plan (simple version) 4. Evaluation (WP 14) Post-optimal analysis: • Analyze model decision and behavior spaces (feasible set, pareto set, preference structures, trade-offs) • Cross-correlation, sensitivity analysis • Test for criteria independence

MC Decision Support Reference point approach: criterion 2 utopia A 4 A 5 A 2 A 6 dominated nadir efficient point A 1 A 3 criterion 1 better

Work Plan (simple version) 4. Evaluation (WP 15) Comparative evaluation across case studies • Ranking by criteria, MC clustering • Discrete Multi-Criteria Optimization with end user involvement • Common patters and trends