Relating canopy size of pecan (Carya illinoinensis (Wagenh.) K. Koch) trees to transpiration

Abstract

The South African pecan (Carya illinoinensis (Wagenh.) K. Koch) sector is not exempt from water scarcity difficulties, so effective irrigation management techniques are required in pecan orchards to help growers maximize production with a limited water supply. One of the first steps in irrigation management is to have a means of estimating orchard water use (ETc), which usually involves a modelling approach. Canopy size is a key determinant of water use when soil water is not limiting and is used together with prevailing weather conditions in many water use models. The FAO-56 approach, in which ETc is calculated as a product of reference evapotranspiration and a crop coefficient (Kc)., is the most widely used method of estimating ETc. Previous research in several fruit trees demonstrated a linear relationship between Kc or transpiration crop coefficients (Kt) and canopy size, indicating that Kc and/or Kt values, and ultimately ETc or T, can be estimated from a measure of canopy size. It is therefore critical to capture canopy size accurately for future modelling exercises and irrigation scheduling in order to optimise yield, growth, and quality of pecan nuts. This study was therefore initiated to quantify the canopy size and water use of a mature pecan orchard at Innovation Africa@UP in Pretoria. Aerial photography was assessed as a means of providing accurate estimates of canopy cover in pecan orchards. Canopy cover estimates of trees in the orchard were compared using red green blue (RGB) images from above the canopy and the Canopeo app, which selects green pixels, with estimates of fractional interception of photosynthetically active radiation (FI-PAR), leaf area index and canopy cover calculated using the shaded area under the canopy. A sap flow technique was used to monitor transpiration rates in the orchards. Results suggest that canopy size can be accurately estimated with aerial photography as it is digitalized and can capture canopy size for large orchards faster. There was a good relationship between canopy cover determined using the Canopeo app and FI-PAR estimated using the ceptometer, with an R2 value of 0.85. There was a poor relationship between canopy cover determined using the Canopeo App and LAI, with the lowest R2 value of 0.56. The results support the hypothesis that the use of photographs captured from above the canopy and image analysis (Canopeo App which selects green pixels) can provide reliable estimates of canopy size, as compared to measurements of FI-PAR by the canopy and canopy cover calculated using the shaded area. Canopy development is influenced by thermal time, thereby the results from this study demonstrated dependency between growing degree days and leaf senescence. The regression between Kt and the different canopy size measures indicated a positive linear correlation, however, this relationship was not good enough to be used in deriving orchard specific values using canopy cover in pecans. The R2 value for the relationship between canopy cover determined using the Canopeo App and Kt values was 0.66, whilst it was 0.7 for midday FI-PAR and Kt, and 0.54 for canopy cover determined as the area on the ground shaded by the tree and Kt. There was a poor correlation between Kt values and LAI measurements, as indicated by an R2 value of 0.41. Despite the fact that this study revealed a poor correlation between Kt and the canopy size measured with Canopeo App, an attempt was made to use the relationship derived in one season to derive weekly Kt values for the 2020/21 season, but a poor relationship was found between measured and estimated T, yielding an R2 value of 0.58. This underestimation was due to a peak in Kt values near the end of the season, which corresponds to the nut filling stage and a minor vegetative flush. Despite some shortcomings, the findings of this study can potentially benefit the pecan industry as the Canopeo App method provided good canopy cover estimates, when compared to widely accepted methods using very expensive equipment.