Abstract:
In Ethiopia, available rainfall records are mainly limited to daily time steps. Though
rainfall time series data at shorter time steps are important for various purposes like
modeling of erosion processes and flood hydrographs, they are hardly available in
Ethiopia. The development of a disaggregation approach allows for the production of
hourly data that add up to certain daily totals. The objectives of this study were (i) to
examine the various methods of disaggregation of rainfall using the higher-level and
lower-level, (ii) to evaluate the performance of stochastic and HYETOS
disaggregation models, and (iii) to develop a rainfall distribution curve and evaluate a
design storm using the US Soil Conservation System (SCS) approach as a base
through the Tana basin. The analysis was based on rainfall data of Dangla, Debre
Tabor, Bahir Dar and Gondar Automatic Gauge Stations. The disaggregation model
used was the Stochastic and Modified Bartlett–Lewis Rectangular Pulse Model
(MBLRPM) called HYETOS. The stochastic disaggregation model result was stated
in monthly Cumulative Density Function (CDF) and monthly convergence parameter
(ε) from 1.05 to 8.95 mm. HYETOS rainfall disaggregation model results obtained in
the optimum magnitudes by model parameters for each month. The wettest months of
July and August, which had extremely low chances of being dry, corresponded to
higher Poisson arrival rates and mean cell rain depths. Their model validity has been
examined using statistical comparison of variance, skewness, probability of dry
period, and Lag-1 Autocorrelation function (ACF). The findings result show that the
methods can preserve statistical characteristics and daily total rainfall depth. The US
SCS method's underlying assumption was used to construct 1-day rainfall distribution
curves and as a consequence have a better effect on the study area for each station
with 24-hour rainfall duration and for design storm computation the designers are
better to use 24-hour rainfall distribution curves by increasing computations.
Key words; Disaggregation, Distribution curve, Downscaling, Rainfall Profile,
HYETOS, Stochastic