Climate change-sensitive hydrologic design under uncertain future precipitation extremes


  • Ramesh S. V. Teegavarapu

    Corresponding author
    1. Department of Civil Environmental and Geomatics Engineering, Florida Atlantic University, Boca Raton, Florida, USA.
    • Corresponding author: R. S. V. Teegavarapu, Department of Civil, Environmental, and Geomatics Engineering, Florida Atlantic University, Rm. 217, Bldg. 36, 777 Glades Rd., Boca Raton, FL 33431, USA. (

    Search for more papers by this author


[1] Precipitation as an important component of hydrologic cycle bears a significant influence on hydrologic design and water resources management. The uncertainties associated with future climate change coupled with limitations of climate change models and uncertainties in projections, our inability to quantify these introduce additional complexities in hydrologic design using future precipitation extremes. A new optimal compromise hydrologic design of a stormsewer system using a fuzzy mixed integer nonlinear mathematical programming (MINLP) model with discrete, binary variables and logical constraints is developed and evaluated in this study. Preferences of hydrologists toward expected uncertain future changes in precipitation extremes are modeled using linear, nonlinear, triangular and Gaussian fuzzy membership functions. Methods for deriving membership functions based on multimodel multiple scenario GCM-based simulations are also suggested. Incorporation of triangular functions in optimization formulation required the use of binary variables. A hypothetical hydrologic design example with realistic parameter values is used to obtain compromise elements of stormsewer system. Results from the study suggest a compromise design is possible based on the preferences and a balance between over and under design of stormsewer infrastructure is achievable. The nature of membership functions reflecting the preferences attached to future extremes influence the optimal solutions obtained resulting in changes in cost-based performance measures.