Modeling daily runoff and probabilistic estimation of design maximum daily runoff from selected watersheds of Udaipur, Rajasthan

Probabilistic estimation of one-day maximum runoff is essential for safe and cost-effective planning and design of surface runoff storage systems, especially in arid and semi-arid regions of developing countries. In present study, one-day maximum and consecutive 2 and 3-day maximum runoff were computed for eight watersheds of Udaipur, Rajasthan. Of the eight watersheds, six are interconnected and two are independent. Daily runoff was estimated for 30-year period by using Soil Conservation Service Curve Number (SCS-CN) model. The SCS-CN model was validated for one independent Bari watershed. Furthermore, 1, 2 and 3-day maximum runoff was fitted with three probability distributions (i.e., log normal, log Pearson type-III, and Gumbel). The chi-square goodness-of-fit test revealed that the best-fit distribution for the one-day maximum runoff from Bari, Chhota Madar, Lakhawali, Fatehsagarand Udaisagar watersheds is log Pearson type-III. Whereas, Gumbel is the best-tit distribution for rest of the watersheds. For 2 and 3-day maximum runoff, either Gumbel or log Pearson type-lit distributions were the best-fit except log normal for 3-day maximum runoff from Lakhawali. Finally, 1,2 and 3-day maximum runoff was estimated at return periods of 1.05, 1.11, 1.25, 2, 5, 10, 15, and 30 years,