Evaporation reduction and prediction model for reservoirs

• Main Author: Mostafa, Ali Mohamed Benzaghta
• Format: Thesis
• Language: English
• Published: 2011
• Online Access: http://psasir.upm.edu.my/id/eprint/42870/1/FK%202011%20117R.pdf; http://psasir.upm.edu.my/id/eprint/42870/

In this study, experiments were carried out at the Universiti Putra Malaysia, Fac-ulty of Engineering in order to investigate the performance of di®erent types of covers for evaporation reduction. The covers used were Mengkuang mat, pieces of plywood and galvanized iron corrugated sheet. The performance of the covers was tested using four PVC tanks. Measurement from the tanks revealed that evaporation reduction using Mengkuang mat was approximately 64% while it were about 50% and 36% for tanks covered with plywood and that shaded with a corrugated sheet respectively. The results obtained from experiments show that signi¯cant evaporation saving are possible if covers are used as barriers between water and atmosphere. Water samples were collected weekly from the four tanks to assess the impact of selected covers on water quality. The collected samples were taken to the laboratory and physical and chemical water quality analysis was conducted. The analysis included testing pH, EC, Tw, DO, BOD and COD. Test results were compared with the standards. The comparison revealed that there was a minor change in water quality of the covered tanks compared to the uncovered. These results confirmed the efectiveness of the covers in evaporation reduction with no harmful efects on water quality. In this study, models for predicting evaporation from reservoirs were proposed. The models were based on the concept of Neural Network. The proposed models were applied to predict the evaporation from two selected reservoirs. One of these reservoirs is Algardabyia Reservoir (located in Libya), while the second one is Batu Dam Reservoir (located in Malaysia). The meteorological data used to run the proposed models were air temperature (T), wind speed (W), relative humidity (RH), and solar radiation (Rs) for the period from 1997 to 2008. Data from 1997 to 2005 was used for models training and validation. Data from 2006 to 2008 was used for model testing. The proposed models simulating the evaporation from Algardabyia Reservoir was named (based on the number of input data) as EM4¡G, EM3¡G and EM2¡G. The same basis is used to name the models proposed for Batu Dam Reservoir. These models are EM4 ¡ B, EM3 ¡ B and EM2 ¡ B. Outputs from the proposed models were validated and it was found in agreement with the historical records. The validation process was based on computing selected statistical indices such as RMSE, MBA, MAE, R2, CE and d. The computed indices show that EM4¡G and EM4¡B models gave the most accurate predictions.Also, selected climate based models such as Penman (Pen) and Priestley-Taylor (P ¡T) were used to predict the evaporation from the Algardabyia Reservoir and Batu Dam Reservoir. The performance of P ¡ T model was found to be better than the Pen model.