Estimating Evaporation Rates From Lake Nainital (Uttaranchal), India: Comparison Of Various Models

摘要

Estimates of lake evaporation are needed for various purposes in water resources planning and management. However, selection of a model for evaporation estimation has many practical constraints. The present study has been undertaken on Lake Nainital, (Uttaranchal) India, with an objective to evaluate the suitability of the various existing models of lake evaporation estimation for the lake. Pan coefficients have been used for the study area to obtain actual lake evaporation from pan data. Evaporation estimates obtained from 12 other models have been compared with these actual lake evaporation rates. The 12 models tested in the study include Jensen-Haise Model, Stephens-Stewart Model, Priestley-Taylor Model, Penman Combination Model, Brutsaert-Stricker Model, Brutsaert Model, DeBruin Model, Makkink Model, Horton Model, Rohwer Model, Fitzerald Model and Thornthwaite Model. Statistical analysis of the data has been carried out. The criteria of, comparison for the study are mean absolute error (MAE), mean absolute relative error (MARE) and root mean square error (RMSE). The results obtained from the study indicate that none of the models are able to estimate the evaporation rate precisely and results show a high degree of error for different models. The models either underestimate or overestimate the evaporation rates in different months. All the models underestimate the peak rates by a high degree. The study concludes that based on the criteria adopted for the study, the Priestley-Taylor Model performs relatively better over rest of the models. The study suggests development of three sub-models for the lake. The study also recommends development of true pan coefficients for the Nainital region based on actual energy balance estimates so that the results could be used for other lakes in the region as well as in similar other regions of India.