1 / 35

Evidence Based Policy Making: Water resources case studies

Evidence Based Policy Making: Water resources case studies. D. Rios Insua, RAC J. Cano, A. Udías, URJC Kiombo Jean Marie, U. Agostinho Neto Hocine Fellag, U. Tizi Ouzou Paris, December’10. Agenda. Background Case 1: Kwanza river management Case 2: Fair water distribution in Kabylia

nash-burks
Download Presentation

Evidence Based Policy Making: Water resources case studies

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Evidence Based Policy Making:Water resources case studies D. Rios Insua, RAC J. Cano, A. Udías, URJC Kiombo Jean Marie, U. Agostinho Neto Hocine Fellag, U. Tizi Ouzou Paris, December’10

  2. Agenda • Background • Case 1: Kwanza river management • Case 2: Fair water distribution in Kabylia • Common lessons

  3. Background

  4. Background Two case studies in water resources management Africa (Angola, Algeria) Spanish Cooperation Agency (us, local uni, one or more local agents) Complex Policy making (public policy decision making) Evidence based?? Policy analytics

  5. Background: EBDM A. Smith. Towards an Evidence-Based Society Most of us with rationalist pretensions presumably aspire to live in a society in which decisions about matters of substance with significant potential social or personal implications are taken on the best available evidence, rather than on the basis of irrelevant evidence or no evidence at all. Of course, the nature of what constitutes evidence in any particular instance could be a matter of debate. Evidence-based medicine (JAMA) de-emphasizes intuition, unsystematic clinical experience and pathophysiologic rationale as sufficient grounds for clinical decision making and stresses the examination of evidence from clinical research. Evidence- based medicine requires new skills of the physician, including efficient literature searching and the application of formal rules of evidence evaluating the clinical literature

  6. Background: EBDM ODI. Evidence-Based Policy Making. What relevance for developing countries? Replacing ideologically driven politics with rational decision making Better utilization of evidence in policy and practice Helps people make well informed decisions about policies, programmes and projects by putting the best available evidence form research at the heart of policy devlopment and implementation Different types of evidence Hierarchy of evidence What evidence is useful to policymakers

  7. Background: EBDM ODI. Evidence-Based Policy Making. What relevance for developing countries? More troubled political context… Conflict Political volatility Role of civil society Centralised policy processes, Less open Lack of performance indicators Lack of data (and if existing difficult to get)

  8. Case 1: Kwanza river management

  9. Background info. The Kwanza river • Divided in three parts: Upper Kwanza, Middle Kwanza, Lower Kwanza • Length: 960 Km • Basin area: 152,570 Km2 • Flow average: 825 m3/s

  10. Background info. Angola • Estimated population: 17600000 inhabitants • Human Development Index (2007):0.564, place 143 of 182 • Gross Domestic Product per capita (2008): US$ 4961 • Population without access to improved water source (2006): 49% • Population without access to energy (2005): 13,5 million of persons • Electrification rate: 15% between 2000 and 2005 • Proportion of renewable energy supply: approximately 1% in 1990 and 1.5% in 2005 • Number of hydrographic basins: 47, the most important is the Kwanza river.

  11. Background info. Capanda Cambambe

  12. Problem formulation. Decision variables • Monthly water through Capanda turbines • Monthly water through Capanda spillgates • Similarly for Cambambe • Monthly water for irrigation at Southern Luanda agricultural projects • Monthly water for consumption at Luanda

  13. Problem formulation. Uncertainties • Water inflow to Capanda • Evaporated water in Capanda • Similarly in Cambambe • Water demand for irrigation in Luanda • Water demand for human consumption in Luanda

  14. Problem formulation. Constraints 1 • Energy production constraints • Release constraints Water through turbines smaller than the maximum It. Through spillgates Releases should respect navigability minima… Maximum capacity for water treatment • Storage constraints Storage at reservoir smaller than maximum Continuity constraints

  15. Problem formulation. Objectives • Maximize energy production • Minimize water releases through both spillgates • Minimize water deficit for irrigation • Minimize water deficit for human consumption Changed several times

  16. Preference modelling. • Energy production g1(et) = 1.005 – 0.991 * exp(-0.012 * et) • gt(dec, inc) = 0,30g1(et)+0,15g2(U2Pt)+0,15 g3(U2Bt)+0,17 g4(kt)+0,23g5(ht)

  17. Uncertainty modelling Unregulated inflows. DLMs Regulated inflows. Regression models, Water travelling time model. Data available for different years… simulate to generate data for same years Evaporation data not available: A model (Visentini)

  18. Problem formulation The problem consists of maximising the following expected utility: Ψ(U) = ∫…∫∑[(0,30(1.027 – 1.012 * exp(-0.0056 * et)) + 0.15(-0.045 + 1.098 * exp(-0.475 * U2Pt))+0.15(-0.045 + 1.098 * exp(-0.475 *U2Bt))+0.17(0.020 + 0.984*exp(-0.083*kt))+0.23(-0.007 +1.058 * exp(-1.232 * ht))]h(i,u)dîB dîp dûurb dûirr Subject to constraints

  19. Problem solution. MC Approximation of objective function • 1/N ∑[(gt(U1Bt,U2Bt,U1Pt,U2Pt,Ujirrt,Ujurbt, i,d)+ (gt+1(U1Bt+1,U2Bt+1,U1Pt+1,U2Pt+1,Ujirr(t+1),Ujurb(t+1),i,d)+…] Similarly for gradient

  20. Problem solution. Number of stages vs Sample size (MATLAB on a standard Laptop)

  21. Problem solution. Strategy adopted At a given time: Forecast for next twelve months. (Inflows, Demands, ‘Evaporations’) Optimise/plan for next twelve months. (Solve MEU) Implement optimal alternative obtained this month. Collect data for forecasting model. Move to next time. Variants: • A different number of months • Add a penalty term to mitigate miopicity

  22. Problem solution. An example of a run 18 months planning. At each stage, 12 months. MC size 500.

  23. The Kwanza problem Finetuning this DSS Capacity expansion- National energy plan (with A. Simbo)

  24. Case 2: Fair water distribution in Kabylia

  25. Background info

  26. Background info • Total area: 2957 km2; • 67 municipalities (1380 villages); • Total population: 1,100,000; • Population over 900 m: 110,000 (90 villages); • Annual rainfall: 900 mm; • Groundwater: 72%; • In 2009: 6,500,559 m3, up to 80% losses; • 9 wells (Bouaid), pumping capacity: 1,040 m3 h−1; • 5 wells (Takhoukht) pumping capacity: 140 m3 h−1; • Taksebt reservoir, pumping capacity: 140 m3 h−1; • 38 reservoirs (storage capacity: 28,000 m3); • 6 intermediate pumping stations 5,400 m3 h−1;

  27. Description of the problem • Distribute water in a equitable & cost-efficient way. • Difficulties to satisfy water demand adequately. Frequent water shortages create political unrest • (Optimization of the pump operational schedules & strategic planning) • Complex rules in energy fares (daytime & contractual issues). • Pumping water uphill creates important engineering problems & high costs due to electricity consumption

  28. Solutions to the problem • Egalitarian solution (suggested by the water company), inflict same per capita water deficit Lk to all users & minimize Lk; • Smorodinsky-Kalai, min-max Lk; • Utilitarian solution, maximize sum of attained utilities (minimizing sum of Lk minimize average per capita water deficit); • Variance solution (not an arbitration solution), minimize standard deviation of Lk; • Minimize inter and intravariability.

  29. Formulation of the problem • OBJ1: Minimize Cost (total electricity pumping consumption); • OBJ2: Achieve equitable distribution. • CONSTRAINTS: • Pumping capacity; • Storage capacity; • Water flow capacity;

  30. Generation of Pareto frontiers

  31. Case study • Two scenarios of distribution network losses (50% and 80% over-demand). • Four infrastructure scenarios.

  32. Case study • Previously ‘fixed’ demand • Demand data unavailable until this project • DLM model per village. Hierarchical approach. Censored data • Stochastic programming

  33. Common lessons

  34. Common lessons • Multiple objectives • Unclear objectives which evolved • Uncertainty • Lack of data (simulate, extrapolate, censored) • Planning over time • Computational compromises • No participation • First quantitative approaches for our clients • Encouraging results • Unveil new realities • Angola (Capacity expansion, Transmission reliability) • Algeria (Thefts, Modify network, Reliability) • Policy making • Evidence based? • Possibly (best available methods, best data available, careful work with experts) • DSS • Policy analytics

  35. Discussion

More Related