PhD Thesis Emmanuel Obeng Bekoe - Cranfield University
PhD Thesis Emmanuel Obeng Bekoe - Cranfield University
PhD Thesis Emmanuel Obeng Bekoe - Cranfield University
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
171<br />
intermediate groundwater store and coefficient of baseflow release from the<br />
intermediate store into the stream (Schulze, 1995).<br />
The earlier analysis of the delayed initial wetting up of the catchment following<br />
the 1967 dry season (in comparison to the observed) which the calibration could<br />
not correct and the generally poor simulation during the dry seasons suggests<br />
weaknesses in the recharge and baseflow generation model components.<br />
Everson (2001) and New (2002) also found some limitations of ACRU in relation<br />
to groundwater simulation.<br />
According to Everson (2001) a limitation observed during a relatively wet year<br />
simulation was the inability of ACRU to account for lateral subsurface soil water<br />
flow during a water balance study in South Africa, where analysis of the<br />
simulated unsaturated flow from the B horizon to the groundwater zone and<br />
saturated drainage could not account for the predominance of subsurface flow<br />
found in the catchment. The soils were reported as having organic matter<br />
contents of 6-10%, and high water holding capacities but the subsoils had a<br />
very high clay content and poor infiltration leading to soil water logging and<br />
lateral interflow within the more permeable upper soil layer. These mechanisms<br />
could occur in Densu Basin because of the heterogeneity of soils within this<br />
large catchment with water holding capacities and high clay contents, between<br />
35-55%, in the subsoils (Adu and Asiama, 1992).<br />
New (2002) observed two shortcomings of the ACRU model which affect<br />
baseflow simulation which are 1) the inability of the model to simulate the<br />
wetting-up of the catchments at the beginning of the wet season, which resulted<br />
in over prediction of streamflow and 2) the tendency of the ACRU model to<br />
underestimate dry season baseflow. In this modelling with ACRU for the Densu<br />
Basin, the dry season produced the most overestimation which may be as a<br />
result of reasons deduced in the earlier Sections such as “poor” input data, and<br />
weaknesses in the groundwater routine in ACRU where the delay in the initial<br />
wetting up of the catchment in 1968 (in comparison to the observed) could not<br />
<strong>Emmanuel</strong> <strong>Obeng</strong> <strong>Bekoe</strong> Phd <strong>Thesis</strong> Chapter 6 Discussion