Modeling water resources management at the basin level: Methodology and application to the Maipo River Basin

With increasing competition for water across sectors and regions, the river basin has been recognized as the appropriate unit of analysis for addressing the challenges of water resources management. Modeling at this scale can provide essential information for policymakers in their resource allocation decisions. A river basin system is made up of water source components, instream and off-stream demand components, and intermediate (treatment and recycling) components. The river basin is thus characterized by natural and physical processes but also by human-made projects and management policies. The essential relations within each component and the interrelations among these components in the basin can be represented in an integrated modeling framework. Integrated hydrologic and economic models are well equipped to assess water management and policy issues in a river basin setting. McKinney et al. (1999) reviewed state-of-the-art modeling approaches to integrated water resources management at the river basin scale. Based on that review, this report describes the methodology and application of an integrated hydrologic-economic river basin model. Today we are faced largely with a "mature" water economy (Randall 1981), and most research is conducted with an explicit recognition of and focus on the need to cope with resource limits. Research has focused on a multitude of situations that might be presented in a mature water economy. Much work has been motivated by expanding municipal and industrial demands within a context of static or more slowly growing agricultural demand. More recently, expanding instream environmental demands have been added. On the supply side, conjunctive use of groundwater has been considered in addition to simple limits on surface water availability. In addition, some researchers have worked on waterlogging and water quality effects (primarily salinity). The work presented in this report examines a "complex" water economy: one in which demands grow differentially not only within but also among sectors, and one in which limited opportunities for increasing consumptive use exist. In particular, the growth of high-value irrigated crop production (such as table grapes) within the case study basin (the Maipo River Basin in Chile), together with the rapidly growing urban area, provides a rich context in which to examine the general problem of basin-level water resources management. The methodology presented is optimization with embedded simulation. Basinwide simulation of flow and salinity balances and crop growth are embedded with the optimization of water allocation, reservoir operation, and irrigation scheduling. The modeling framework is developed based on a river basin network, including multiple source nodes (reservoirs, aquifers, river reaches, and so on) and multiple demand sites along the river, including consumptive use locations for agricultural, municipal and industrial, and instream water uses. Economic benefits associated with water use are evaluated for different demand-management instruments - including markets for tradable water rights - based on the production and benefit functions of water use in the agricultural and urban-industrial sectors. The integrated modeling framework makes use of multiple techniques, such as hydrologic modeling, spatial econometrics, geographic information system (GIS), and large-scale systems optimization. While these techniques have been adapted in other studies, this study represents a new effort to integrate them for the purpose of analyzing water use at the river-basin level. The model's main innovative feature and advantage lie in its ability to reflect the inter-relationships among essential hydrologic, agronomic, and economic components and to explore both economic and environmental consequences of alternative policy choices. The model can be used as a decision-support tool to assist water management authorities and policymakers in the selection of appropriate water policies and in the establishment of priorities for reform of institutions and incentives that affect water resource allocation. The Maipo River Basin in Chile was chosen as the case study site in response to (1) increasing demands and competition among the major water-using sectors of agriculture, urban areas, and industries; (2) growing concerns over how these demands can be efficiently, equitably, and sustainably met; (3) increasing concerns over water pollution from agricultural, urban, and industrial users; and (4) innovative water management and allocation policies in the basin, including markets for tradable water rights. The model is applied to the Maipo River Basin to address site-specific research questions regarding (1) the role of water rights and water rights trading in enhancing allocation efficiency; (2) the role of water use efficiency in saving water in the irrigation sector; (3) the impact of changes in physical irrigation efficiency on basin-level economic and physical efficiency; (4) the effects of hydrologic uncertainty on irrigation technology choice; and (5) the potential for substitution among water and other crop production inputs. Model applications presented thus focus on the relative benefits of alternative water allocation institutions. The major findings of the report can be summarized as follows: 1. Modeling tool. The holistic integrated economic-hydrologic modeling framework developed for this study reflects the major hydrologic and economic processes related to water supply and demand in a river-basin context. Such a tool can be used to assist policymakers in their strategic water allocation decisions. 2. Intersectoral water transfers between irrigated agriculture and domestic and industrial uses. Under full basin optimization - assuming an omniscient decisionmaker with perfect foresight, optimizing the entire water-related basin economy, and current irrigation system and water supply charges - crop area in the basin can increase by 15 percent. Further, total crop revenue comprises an increase in high-value crops of up to 50 percent and a reduction in low-value crops, such as annual prairie and maize, of 40 percent. Under this scenario, total agricultural water withdrawal increases slightly, and domestic and industrial withdrawals almost double. This is a key and (for most contexts) surprising conclusion of the study. 3. Benefits from water rights trading with water moving into higher-valued domestic and industrial uses. When water rights can be traded, as opposed to being fixed, net farm incomes can increase substantially. Moreover, agricultural production declines only minimally as a result of water trading because economic efficiency increases. Net benefits can be even larger than under basin optimization for some irrigation districts because farmers can reap substantial benefits from selling their unused water rights during months with little or no crop production. Reducing transaction costs has significant benefits because both the volume of water traded and the benefits from trade are increased. 4. A shift from fixed to tradable water rights can lead to substantial gains in economic efficiency without prior changes in physical efficiency levels, such as water distribution/conveyance efficiency and field application efficiency. 5. Both an ideal optimal solution of basinwide water allocation, and water right trading and pricing incentives indicate that existing water and land use shift from lower-value to higher-value crops. Note, however, that additional factors like soil type and socioeconomic constraints to transfers are not taken into account in this analysis. 6. From a basin perspective, the potential for water savings from increases in water use efficiency in irrigation systems is lower than individual system efficiencies might indicate. If flow returning from an irrigation system to a water supply system can be reused in the system, then the actual or effective efficiency at this irrigation system will be higher than the traditional estimated system efficiency. This study shows that, for the Maipo River Basin, improvements in irrigation system efficiencies do increase basinwide economic efficiency over a wide range of efficiency increases. At low levels of local efficiency, improvements in system-level efficiency generate significant basinwide profits. However, when local efficiencies reach higher levels, the relative contribution to basinwide water use profits declines. At high levels of irrigation system efficiency, the contribution to basin profits from further improvements in local efficiency is minimal. 7. Tradable water rights also induce improvements in physical efficiency because it becomes more profitable for farmers to invest in improved irrigation technologies and to sell their surplus water. 8. Higher water charges result in higher basin efficiency in terms of physical indicators, such as water distribution/ conveyance and field application efficiencies, and economic indicators, such as total net profit of water use and net profit per unit of water use because farmers reduce water use, shift from lower-value to higher-value crops, and shift to higher levels of irrigation technology for some crops. Moreover, incentive prices have little impact on the efficiency of irrigation systems at low levels of infrastructure development, and the improvement of physical structures can significantly strengthen the effectiveness of water prices or other economic incentives. 9. Physical and economic efficiency levels in the basin depend on the level of water being withdrawn under the current water supply and infrastructure conditions. Both physical and economic efficiency indicators improve at relatively low (but increasing) withdrawal levels compared with water needs, and increased irrigation withdrawals lead to significant increases in farmer incomes. At high withdrawal levels, both efficiency indicators decrease.