Seite - 596 - in Book of Full Papers - Symposium Hydro Engineering

𝑆 𝑒 𝑑 = 11.8× (𝑄 𝑠 𝑢 𝑟 𝑓 ×𝑞 𝑝 𝑒 𝑎 𝑘 × 𝑎 𝑟 𝑒 𝑎 ℎ𝑟 𝑢 ) 0.56 × 𝐾 𝑈 𝑆 𝐿 𝐸 × 𝐶 𝑈 𝑆 𝐿 𝐸 × 𝑃 𝑈 𝑆 𝐿 𝐸 × 𝐿 𝑆 𝑈 𝑆 𝐿 𝐸 × 𝐶 𝐹 𝑅 𝐺 [1] Where; 𝑄 𝑠 𝑢 𝑟 𝑓 - Surface runoff volume (mm/ha) 𝑞 𝑝 𝑒 𝑎 𝑘 - Peak runoff rate (m3/s) 𝑎 𝑟 𝑒 𝑎 ℎ𝑟 𝑢 - Area of hydrologic response unit (ha); defined under methodology 𝐾 𝑈 𝑆 𝐿 𝐸 - USLE soil erodibility factor 𝐶 𝑈 𝑆 𝐿 𝐸 - USLE cover and management factor 𝑃 𝑈 𝑆 𝐿 𝐸 - USLE support practice factor 𝐿 𝑆 𝑈 𝑆 𝐿 𝐸 - USLE topographic factor 𝐶 𝐹 𝑅 𝐺 - Coarse fragment factor As per the MUSLE all other factors except 𝑄 𝑠 𝑢 𝑓 and 𝑞 𝑝 𝑒 𝑎 𝑘 are unique characteristics of a basin area. As the same land use and soil map were used throughout the modelling period (1960-2016) of this study, as described under methodology, only 𝑄 𝑠 𝑢 𝑟 𝑓 and 𝑞 𝑝 𝑒 𝑎 𝑘 were the varying parameters. Hence, any increase or decrease in erosion rates during the period considered for the analysis can be attributed to the variability of rainfall. 2.2 TREND ANALYSIS FOR A TIME SERIES One objective of this study was to check whether climate change has any impact on MRUB soil erosion rate. For finding that time series output of total annual sediment has to be checked for presence of any trend. The Spearman’s rank correlation method is a useful tool for trend analysis of a time series [3]. The Spearman’s rank correlation coefficient 𝑅 𝑆 𝑃 is defined as follows [Eq. 2]. 𝑅 𝑆 𝑃 = 1- 6×∑ 𝐷 𝑖 2𝑛 𝑖 =1 𝑛 (𝑛 2−1) [2] Where 𝑛 is the total muber of data and 𝐷 is difference, and 𝑖 is chronological order number. 𝑡 𝑡 = 𝑅 𝑠 𝑝 [ 𝑛 −2 1−𝑅 𝑠 𝑝 2 ] [3] Where 𝑡 𝑡 , has Student’s t-distribution with 𝑣 = 𝑛 − 2 degrees of freedom. At a significance level of 5 per cent (two-tailed) and the time series has no trend if; 𝑡 {𝑣 ,2.5%} < 𝑡 𝑡 < 𝑡 {𝑣 ,97.5%} [4] 596