Estimation of transpiration in oil palm (Elaeis guineensis Jacq.) with the heat ratio method

Estimación de la transpiración en palma de aceite (Elaeis guineensis Jacq.) por el método del radio de calor

This work is licensed under Creative Commons Attribution 4.0 International.

Oil palm cultivation in Colombia has grown significantly in recent years (Fedepalma, 2014), with a planted area of 450 thousand hectares distributed in different regions of the country. Oil palm requires large amounts of water for good production (Corley and Tinker, 2003; Paramananthan, 2003; Carr, 2011) assuming an average water requirement of 6 mm per day. However, the measurements used to determine the overall transpiration rate of oil palm have been based on allometric parameters (Kallarackal et al, 2004; Legros et al, 2009) or on gas exchange measurements (Suresh et al., 2010; Rivera et al., 2012; Suresh et al., 2012). All of these studies have allowed us to partly understand the water demand of oil palms under different controlled conditions; however, the water requirements of oil palm in the field are not well known. Therefore, it is necessary to take direct measurements to determine water use under field conditions over time.

Different measurement methods have been used to quantify the transpiration of plants, including the estimation of water use based on water balance and measurements of precipitation, drainage and changes in the water reserve available to plants (Ruiz-Peñalver et al, 2014), along with the lysimeter method, which, for perennials, requires a large infrastructure, such as that used in the study conducted on date palm (Tripler et al., 2012). However, even though this method has been improved, the size of the plant is still a major limitation. For smaller species, inexpensive lysimeters can be used, as reported by Ruiz-Peñalver et al. (2014). Another method used is the Granier method (Granier, 1987) and its modifications (Burgess et al, 2001), a technique that measures sap flow rates and the volumetric flow of water in xylem tissue with a short heat pulse as an indicator. This is performed by measuring the ratio of the amount of heat transported to two symmetrically arranged temperature sensors to calculate the quantity and direction of the water flow (Smith and Allen, 1996; Bleby et al, 2004). Research with sap flow meters has been corroborated by other techniques, such as the lysimeter and water balance, with a high correlation between the different techniques. These results show the usefulness of this technique (Trcala and Cermák, 2012; Langensiepen et al, 2014).

The aim of most of the studies that used the heat pulse method was to determine the water balance and flow of assimilates, such as the study by Haijun et al. (2015) under mesh-house conditions, in which the total transpiration of banana plants was determined. Based on information provided by the device, they were able to determine the water demand of the plants and, therefore, the optimum irrigation schedule to prevent water stress. The same has been done with cacao (López-López et al. 2013) and eucalyptus (Bleby et al., 2012); various studies were able to determine and quantify the sap flow rate with a high degree of accuracy using the heat pulse method.

The aim of this study was to measure the total transpiration of a commercial oil palm plantation under field conditions. We examined climate variables in order to understand which of them play significant roles in transpiration fluctuations. Furthermore, we studied the influences of leaf age on whole plant transpiration. The results have a potential to be used as a decision making tool in water management and supply under a precision agriculture scheme.

During the sampling days, the soil water potential was constant in a range of -0.1 to -0.2 MPa, and the sap flow showed no significant difference in any of the evaluated palms. However, differences were found between leaf 9 and leaf 25, where leaf 17 had an intermediate value between them. A similar pattern was found in each of the sampled leaves (Fig. 2), where the sap flow rate ranged from 10 to 20 cm. h^-1 overnight - from 18:00 HR to approximately 6:00 HR - increasing 20-fold 2 h after the arrival of the first rays of light (Fig. 2A). The highest flow rates occurred between 9:00 and 16:00 HR (220 cm. h^-1), with fluctuations between 20 and 40 cm. h^-1. At 16:00 HR, the flow rate started to decrease, reaching the lowest values after 18:00 HR; this cycle repeated every day during the sampling period. Additionally, a reduction in the sap flow was observed after day 7 of the evaluation, associated with a decrease in vapor pressure deficit (VPD).

Figure 3
Accumulated volume of water transported by 3 leaves of oil palm during a 12-day sampling period by the heat ratio method (n = 4).

The volume of water transported through the leaf petiole bases during the sampling period was calculated based on the recorded information (Fig. 3). Leaf 9 transported the largest volume of water (21.65 L), followed by leaf 17 (17.20 L) and leaf 25 (14.6 L) during the 12 d of sampling. This variable showed statistically significant differences over time from day 4. No statistically significant differences were found between palms, which indicated a good repeatability of sampling. Additionally, the volume of water transported increased during the day, as illustrated by valleys related to the lower volume of water transported during the night.

Figure 4
Relationship of sap flow rates of 3 leaves of oil palm by the heat ratio method with climate variables. PAR, photosynthetically active radiation; RH, relative humidity; AT, ambient temperature; VPD, vapor pressure deficit (n = 4). Leaf 9, leaf 17 and leaf 25.

The results are similar to those found in pindo palm (Madurapperuma et al., 2009), pine (Chang et al., 2014) and wheat (Langensiepen et al., 2014), among other species. Daily fluctuations are directly associated with soil water conditions and weather variations that influence gas exchange in plants. For example, in jujube (Chen et al, 2014), a reduction in soil water potential and change of meteorological factors caused a direct response in photosynthesis and transpiration rates, which was reflected in reduced sap flow. We found that sap flow rates are closely related to weather variables (Fig. 4), where the vapor pressure deficit (VPD) was the most correlated factor. These results are similar to those reported for oil palm in Indonesia (Niu et al., 2015). A multiple regression was also performed, taking into account the main components (AT, PPFD and VPD) (Fig. 5), in which a high linearity was found. However, it is necessary to include other variables to complete a predictive model for irrigation, as was done for jujube and banana(Lu et al., 2002; Haijun et al., 2015; Chen et al, 2014).

Figure 5
Multiple regression for leaves 9, 17 and 25 of oil palm, taking VPD, PPFD and RH into account to develop predictive models (n=4).

Based on the water flow data, including 33 leaves, our estimates of whole-plant water use were calculated as the sum of the rates from individual leaves, which is necessary and useful to assess canopy variability. We found that water use per palm was substantially higher in young leaves than in older leaves. The estimated water consumption per palm was 80.5 L d^-1, for a transpiration of 1.15 mm d^-1, very similar to the value reported by Niu et al. (2015). In the literature (Mutert et al., 1999; Corley and Tinker, 2003), the estimated values based on evapotranspiration and allometric calculations are higher, possibly because those estimations do not directly measure plant water consumption.

The analysis of sap flow measurements showed that each leaf had a different water consumption rate, and young leaves had higher water consumption values. This monitoring allowed us to determine that, for the measurement conditions, the transpiration of one hectare of oil palm was 1.15 mm d^-1. The sap flow rates in the oil palm were closely related to weather variations, particularly vapor pressure deficit, VPD, which is the variable that best explains the fluctuations in water movement in oil palm. Additionally, measurements showed high repeatability using the heat ratio method (HRM) of ICT-International (Australia). However, it is necessary to assess the crop in different seasons to determine water requirements during the dry season and to integrate crop transpiration data with the evapotranspiration data of the ecosystem.

This research at Cenipalma was funded by the Colombian Oil Palm Promotion Fund (FFP), administrated by Fede-palma (Federación Nacional de Cultivadores de Palma de Aceite) and Ministry of Agriculture and Rural Development of Colombia.