Water requirement of oil palm in two different edaphoclimatic conditions in Brazil

• Autor(a) principal: Santos, Otávio Neto Almeida
• Data de Publicação: 2019
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: http://www.teses.usp.br/teses/disponiveis/11/11152/tde-16092019-113831/
• topic: Elaeis guineenses Jacq.; Elaeis guineensis Jacq.; Balanço hídrico; Irrigação; Irrigation; Lisímetro; Lysimeter; Remote sensing; Sensoriamento remoto; Water balance

Oil palm is a perennial evergreen crop which stands out as the crop with the highest oil production per planted area among the oilseeds. The main objective of this thesis was to determine the oil palm water requirement in two different edaphoclimatic Brazilian conditions. Specifically, this study aimed to: (i) quantify actual evapotranspiration (ETa), develop crop coefficients (Kc and Kcb), and determine the relationship between the basal crop coefficient (Kcb) and leaf area index (LAI) of young irrigated oil palm growing under the Brazilian Humid Subtropical conditions; (ii) determine the spatiotemporal variation of oil palm evapotranspiration from adjusted satellite images, as well as the adjusted crop coefficient (Kcadj) and its components under Amazon climate conditions. To compute the water requirement of young oil palm trees, we used a large and precise weighing lysimeter (4.0 m diameter x 1.3 m depth) installed in Piracicaba, SP, Brazil. Crop coefficients were obtained by dividing lysimetric measurements over 24 h period by reference evapotranspiration (ETo) calculated by the Penman-Monteith method (FAO-56) and LAI measurements were performed by using the LAI-2200 Plant Canopy Analyzer. Lysimetric measurements showed that ETa and transpiration young oil palm were 2.50 ± 1.39 mm d-1 and 1.43 ± 1.09 mm d-1, respectively, and the corresponding Kc and Kcb were 0.71 and 0.41, respectively. Average Kcadj for plants between 18- and 33-month-old was 0.08 and is recommended only for drip irrigation as it is the result of an adjustment that depends on ground coverage and crop spacing. The Kcb-LAI relationship obtained for oil palm was Kcb = 0.5895 LAI - 0.6674 (R2 = 0.9856) and can be useful to estimate oil palm water use from LAI measurements. To determine the oil palm water use in the Eastern Brazilian Amazon, a remote-sensing-based ET and water balance model was performed over an 8-year consecutive period in a commercial oil palm site near Moju, Pará, Brazil. The results of the water balance model revealed that under Amazon climate influence, the multi-year average of daily and total ETa was about 3.4 ± 0.4 mm d-1 and 1229 ± 127.2 mm yr-1. The oil palm water requirements were lower during the dry season (364.7 ± 88.94 mm) comparing to the rainy period (864.4 ± 80.91 mm) as a result of eventual water stress. In an annual average basis, Kcadj was 0.87 ± 0.42, Kcb for the two first growing years was 0.78 ± 0.29, reaching an average of 1.16 ± 0.04 from the third cropping year. The correlation between in situ measured and modeled oil palm fresh fruit bunches (FFB) yields was described by a linear function (Yieldmod = 0.7626 Yieldmeas + 538.64; R2 = 0.9913). In summary, this thesis presents interesting results on palm oil needs in two Brazilian regions, which may be of fundamental importance in establishing strategies to improve the efficiency of water use in palm plantations.