Deriving crop coefficients for evergreen and deciduous fruit orchards in South Africa using the fraction of vegetation cover and tree height data

Inaccurate crop coefficients are major contributing sources of uncertainty that lead to inefficient use of limited available water resources. Understanding the need to improve water use efficiency in South Africa's fruit in-dustry, this study evaluated the method of deriving crop coefficients developed by Allen and Pereira (2009) over a variety of irrigated fruit tree crops. Detailed data of transpiration, evapotranspiration and weather variables measured using the heat ratio method, eddy covariance method and automatic weather stations, were collected from a water research funding body established by the South African government. This study adjusted the sto-matal sensitivity function (Fr) in the model by replacing the ratio of the leaf resistance (rl) to the standard leaf resistance of a reference crop (100 sm- 1) with rl/alpha where alpha is a resistance parameter for the specific crop. The resistance parameter was solved accordingly for each fruit type. Respective unique alpha values were obtained as: macadamia nuts (200 sm-1), citrus (50 s m-1), peaches (20 s m- 1) and pecans (20 s m-1). These unique values were used to simulate basal and single crop coefficients that produced satisfactory results when compared to the actual measured values. Overly, no unique standard alpha value exists for most tree crops although a value close to 20 sm-1 may give reasonable estimates for pome and stone fruit. Crop coefficients derived using locally measured data were standardised and tabulated in a format that facilitates their transferability between sites. However, there is still a need to acquire crop specific information to parameterize alpha and improve accuracies.

Automated summary

Inaccurate crop coefficients lead to inefficient use of water resources. This study evaluated the method of deriving crop coefficients developed by Allen and Pereira (2009) for various irrigated fruit tree crops in South Africa. The study adjusted the stoma-tal sensitivity function in the model by replacing the ratio of the leaf resistance to the standard leaf resistance with a resistance parameter specific to each crop. The derived crop coefficients were standardized and tabulated for transferability between sites, but more crop-specific information is needed to improve accuracy.