Loading...
Quantifying Uncertainty and Error in River Basin Models: Lessons from Pangani and Rio Grande
Citations
Altmetric:
Abstract
Legal and administrative procedures require policymakers to justify financial investments in water infrastructure projects based on some computational forecast of its benefits. Such forecasts are typically generated using computer models of river basins. However, ex-post evaluations of actual water projects show that many of the benefits predicted by the model fail to materialize after the project comes online. This discrepancy between forecast and actual benefits is a symptom of the models’ inevitable errors and uncertainties.
The model errors and uncertainties arise due to a spectrum of difficulties ranging from practical intractability to theoretical difficulty. At the heart of these difficulties is the issue of predicting how societies will decide to operate water infrastructure and use water for their needs. Human water management decisions are essential to river basin models and their effects across the river basin may even dominate those of purely geophysical processes. The scientific field of water resources systems analysis creates models of human decisions yet pays little attention to these inherent errors.
The objective of this dissertation is to address this common oversight and use two case studies to demonstrate methods to quantify the uncertainties due to representing human decisions in river basin models.
Type
Dissertation
Date
2024-05
Publisher
Degree
License
License
Research Projects
Organizational Units
Journal Issue
Embargo Lift Date
2025-05-17