Assessing the influence of rainfall-intensity-dependent time of concentration on spillway design under future climate scenarios in the knuckles basin, Sri Lanka

Abstract

Although conventional hydrological design treats time of concentration (Tc) as a static constant, Tc is inversely related to rainfall intensity, making its dynamic representation important under climate change. This study assesses how rainfall-intensity-dependent Tc affects spillway design under future climate scenarios in the Knuckles Basin, Sri Lanka. Future climate projections were obtained from three CMIP6 GCMs (CNRM-CM6-1, HadGEM3-GC31-LL, MRI-ESM2-0) under SSP2-4.5 and SSP5-8.5, statistically downscaled to watershed level for 2041–2060 and 2081–2100 using LARS-WG. The resulting daily rainfall series were then used to develop future Intensity-Duration-Frequency (IDF) curves, with daily rainfall values disaggregated using a regionally developed by Engr. Ponrajah. An existing rainfall intensity–Tc relationship, developed using Detrended Moving-Average Cross-Correlation Analysis (DMCA), was adopted to estimate rainfall-intensity-dependent Tc, and hydrological simulations were performed in HEC-HMS using the SCS Unit Hydrograph method. The dynamic Tc relationship was incorporated into design-storm simulations to evaluate hydrograph shape, peak discharge, and spillway response. Results demonstrate an inverse relationship between rainfall intensity and Tc, with Tc decreasing from approximately 2.5 to 1.0 hours as intensity increases from 25 to 45 mm/h. Projected peak discharge varied substantially across climate projections, ranging from 81.6 to 205.4 m³/s. MRI-ESM2-0 under SSP5-8.5 for 2081–2100 produced the highest projected peak, reaching 205.4 m³/s, approximately 2.5 times the baseline peak discharge of 82.2 m³/s. To convey this maximum inflow, the spillway design required an increase in crest length from 58 m to 80 m and crest elevation from 72 m to 73 m, which attenuated the reservoir peak discharge from a baseline of 525 m³/s to 179.6 m³/s. These findings highlight the importance of dynamic Tc and multimodel climate projections for climate-resilient spillway design.