Assessing suitability of the ACRU hydrological model in a rainforest catchment in Ghana, West Africa

Abstract

Hydrological modelling is a challenge in the rainforest dominated Bonsa catchment (1482 km 2) in Ghana, West Africa, because of data scarcity and rapidly changing land uses. The objective of this study was to assess the suitability of the daily time step physical-conceptual ACRU model for hydrological modelling in the Bonsa catchment. Since the catchment is data poor, model calibration was conducted using a careful parameterization and sensitivity analysis, using initial values obtained from literature and field observations, as well as climate data for the period 1987-1999 and 1991 land use. The model performance during calibration and validation was satisfactory, with a monthly NSE of 0.6 and 0.5 and R 2 of 0.8 and 0.7, respectively. The model simulated the rise and the recession of the hydrograph well, but during the validation the accumulated monthly streamflows were underestimated by 10%. The main conclusion from this study is that the ACRU hydrological model is suitable for exploring basic hydrological responses to land use and climate in the Bonsa and similar catchments.

Assessing Impacts of Land Use Changes on the Hydrology of a Lowland Rainforest Catchment in Ghana, West Africa

“…A detailed description of the model is available in Schulze [38]. The ACRU model has previously been applied in the catchment with a Nash-Sutcliffe efficiency (NSE) index of 0.6 and 0.5, and R 2 of 0.8 and 0.7, for calibration and validation of monthly flows, respectively [37]. This study builds on this application, as well as knowledge gained through several applications of the model in a variety of catchments [13,43,44], with a wide range of climates and land uses.…”

“…Climate records were obtained from the Ghana Meteorological Agency (GMA) for the period of 1980 to 2011. Due to data gaps and outliers in the rainfall records, the data was preprocessed [37] before using it in the ACRU model. Reference potential evaporation was calculated using the Hargreaves and Samani method [46].…”

“…An example of the flow sequence of subcatchments is shown in Figure 3. The land use parameters used in this study were the same as those used in the calibration and validation study [37], which were obtained after a sensitivity analysis in the Bonsa catchment. The assumption made was that the land use types remained the same, however, the area and distribution changed depending on land use scenario ( Figure 4; Table 1).…”

“…The substantial increase in streamflows in the current (2011) scenario is largely because of the over 300% increase in shrubs/farms area between 1991 and 2011. The conversion of evergreen forests with higher crop coefficients (CAY) and interception loss by vegetation (VEGINT) [37] to land uses with lower values of the parameters led to the increased streamflows.…”

“…In addition to a baseline (1991) and current (2011) land use scenarios, three future land use scenarios developed in a previous study [36], namely: (i) the business-as-usual (BAU); (ii) the economic-growth (EG), and; (iii) the economic growth and reforestation (EGR) scenarios, were used. This study builds on a companion paper [37], which has demonstrated that the Agricultural Catchments Research Unit (ACRU) hydrological model [38] is a suitable tool for modelling the hydrology of Bonsa catchment. Several studies have also applied the ACRU model successfully to assess the impacts of land use changes on the hydrology of catchments with diverse land uses and climates [21,39], demonstrating the sensitivity of the model to land uses and changes thereof.…”