Introduction
The African oil palm (Elaeis guineensis Jacq.) has become one of the most important industrial crops in Colombia and worldwide. Additionally, the interspecific hybrid [Elaeis oleifera (H.B.K.) Cortés x Elaeis guineensis Jacq. (OxG)] has gained importance over the last decade due to the advantages that it offers, especially those related to disease resistance (Zambrano and Amblard, 2007). Despite the high yield potential of both the hybrid and E. guineensis (Bastidas et al., 2007), it has been determined that the oil production is associated with soil nutrient and climate conditions, which have fluctuations throughout the year. This variation is reflected in the concentration of most of the harvest in short periods of time, largely associated with the rainfall patterns (Henson and Chai, 1998).
Water is the main factor involved in the biochemical pro cesses of photosynthesis and plays an essential role in the absorption and transport of nutrients from the soil (Cao et al., 2011). Photosynthesis may be inhibited during dry peri ods due to an increased vapor pressure deficit that forces the stomata closing (Dufrene and Saugier, 1993). Additionally, dry seasons are associated with higher temperatures, which may induce a reduction in the photochemical efficiency (Corley, 1982). However, according to Corley and Tinker (2008) high yields can still be obtain from palms that grow in suboptimal conditions due to the flexibility of this spe cies to withstand adverse conditions over extended periods.
In the Eastern Plains of Colombia, almost the entire an nual rainfall occurs during seven to eight months and a four-month dry season receives less than 50 mm of monthly rainfall (Rippstein et al., 2001). Considering that an adequate water supply and temperature are the most important factors that determine the performance of oil palms in the tropics and subtropics (Barrios et al., 2003; Corley et al., 1971), low rainfall, affects the oil production due to its effects on the fruit filling, abortion of inflores cences and sexual differentiation, among other processes (Corley and Tinker, 2008; Henson et al., 2005).
To avoid the adverse effects of dry seasons, growers use a variety of irrigation methods. Empirical methods are the most widely used, although they are not the most advisable (Lascano 1998). There are structured irrigation systems (gated pipe, spray, drip irrigation, etc.) that could be more expensive (Monroy 2010). Additionally, the irrigation practices depend on the increasingly limited availability of water (Domínguez et al., 2008), which makes harder watering the plantations. These situations lead basically to two options: finding materials tolerant to water deficit or the implementation of efficient, low-cost irrigation systems. Therefore, it is necessary to understand the physiological response of different materials to water deficit events.
Material and methods
Location
The research was conducted in the Mecasaragua and Aurora fields in the Cuernavaca estate of Unipalma S.A. plantation. This farm is located in the foothills of the eas tern plains (4°19" N and 73°13" W), in the municipality of Paratebueno (Cundinamarca, Colombia), at an altitude of 245 m a.s.l. The annual average temperature is 28°C, the annual rainfall is 2,800 mm and the relative humidity is 80%.
Plant material
Nine-year old E. guineensis and OxG interspecific hybrid palms (planted in 2004) were planted with at 9x9 m sta ggered spacing. Palms planted in the Aurora field were flood-irrigated during the dry seasons. Palms planted on the Mecasaragua field were maintained under rain-fed only conditions and were never artificially irrigated. Evaluation was performed over 16 palms per material type in each field during three seasons (wet season, dry season and dry-to-wet transition season). For the dry season, the assessments were performed after 70 d with no rain.
Plant water potential
The leaf water potential was determined by using a pressure chamber (Scholander pump, PMS Instrument, Albany, OR) (Bayona et al., 2007). The operating principle consists of introducing leaflets in an airtight chamber immediately after being cut; the lid has a small opening where the base of the leaf central vein is exposed; pressure is applied inside the chamber by injecting nitrogen gas to counteract the ne gative pressure of the water column on the leaf until a drop is observed in the exposed base of the vein. The pressure is recorded on a built-in pressure gauge. This measurement was performed between 4:00 and 6:00 h (predawn) to verify whether the plants were stressed.
Gas-exchange measurements
To quantify the gas exchange (photosynthesis, transpira tion and stomatal conductance) a portable photosynthesis meter was used (LiCor 6400XT, LI-COR Biosciences, Lincoln, NE) with the following benchmark parameters: 400 ppm CO2 and 1,000 µmol m-2 s-1 PAR radiation. The maximum photosynthesis was measured between 8:30 and 11:30 h. And the Instantaneous water-use efficiency to photosynthesis (WUEp) was determined by the ratio of PN to E.
Chlorophyll fluorescence
The chlorophyll fluorescence was determined using a por table photosynthesis system with an integrated fluorescen ce chamber (6400-40) LiCor 6400XT (LI-COR Biosciences, Lincoln, NE); Suresh et al. (2010) methodology was used with some modifications over time to adapt to darkness and saturating light pulse. We recorded the parameters of the maximum quantum yield (Fv/Fm), the photochemical efficiency of photosystem II (cpPSII) and the electron trans fer rate (ETR). Three leaflets from leaf 17 were taken from each palm and adapted to the darkness for 2 h; the Fv/Fm was measured; the leaflet was then saturated with 3,000 φimol m-2 s-1 radiation for 1 min, after which time the ETR and φPSII data were recorded. The measurements were performed only during the dry and wet seasons.
Results
Water potential
Figure 1 shows the predawn water potential of the palms evaluated in different seasons. Significant differences were found between the dry season and the dry-to-wet transi tion season in the hybrid material planted in the field with irrigation; the recorded values in each season were -51±13 KPa and -118±35 KPa, respectively. In the field with no irrigation, the water potential pattern was different and showed more negative values during the dry and dry-to-wet transition seasons compared with the wet season. However, no statistically differences were found in this field, where the reported values were -143±100 KPa, -160±52 KPa and -84±53 KPa, respectively. E. guineensis palms recorded the most negative values during the dry season in the field with no irrigation (-509±95 KPa), which indicates a severe stress. These values showed significant differences with respect to the seasons and the other groups evaluated.
Gas exchange
Figure 2 shows the behavior of the materials evaluated in the two fields (irrigated and non-irrigated) over time during three different seasons (dry, transitional and wet seasons) with respect to photosynthesis (Asat) and water-use effi ciency to photosynthesis (WUEp). Significant differences in photosynthesis were found between the irrigated and non-irrigated treatments in the dry season for both E. guineensis and the hybrid. During the transitional season there were also differences in the hybrid but not in E. guineensis. No differences were found between the two treatments during the wet season. In a separate evaluation of the two fields, with an average Asat of 11.81±2.4 vmol CO2 m-2 s-1 for E. guineensis and 11.64±2.3 |imol CO2 mf2 s-1 for the hybrid, no significant differences between materials over time were found in the irrigated field. However, significant differences were found among seasons in the field without irrigation. The Asat averages reported for E. guineensis were 4.52±1.08; 10.41±0.86 and 13.14±0.99 µmol CO2 m-2 s-1, in the dry, transitional and wet seasons, respectively, and 6.52± 1.45; 9.44±1.07 and 12.98±0.89 on average for the hybrid during the same seasons. However, no statistical significant diffe rences were found between the two materials.
The transpiration rate (E) of the materials showed a similar pattern. Statistically significant differences were found for E. guineensis between the dry season and the transitional season, while for the hybrid, differences between fields were found only during the dry season. In a separate analysis of the two fields, no statistically significant differences were found between materials over time in the field with irrigation, with an average transpiration rate of 3.37±0.58 mmol H2O m-2 s-1 for E. guineensis and 3.34±0.33 mmol H2O m-2 s-1 for the hybrid. In the field without irrigation, no differences were found for E. guineensis between the dry and the transition seasons, where the transpiration rate raised to 1.67±0.23 mmol H2O m-2 s-1. Instead, dur ing the wet season, the transpiration rate was 3.54±0 64 mmol H2O m-2 s-1. The transpiration rate of the hybrid showed an interesting pattern because significant differ ences were found between the dry and wet seasons, but with an intermediate transpiration rate values during the dry-to-wet transition season. The values reported for the dry, transitional and wet seasons were 1.52±0.67; 2.62±0.70 and 3.53±0.48 mmol H2O m-2 s-1, respectively.
The ratio of the photosynthetic rate to the transpiration rate determines the efficient use of water by the process of photosynthesis (WUEp; Blum, 2009). This ratio is basically the ratio of the water released by the plant to fix a given amount of CO2. Therefore, a higher WUEp value means a greater efficiency in the use of water for photosynthesis.
In the field with irrigation, significant differences in both materials were found between the dry and wet seasons, with values of 4.47±0.68 and 2.96±0.60 µmol CO2 /mmol H2O, respectively. Instead, the transition season showed intermediate values. No significant differences were found between the materials, which during the dry season were more efficient in the use of water. The field without irriga tion showed different responses. No significant differences were found for the hybrid with respect to the WUEp over time, with an average value of 3.89±0.78 µmol CO2/mmol H2O. However, the WUEp of E. guineensis decreased sig nificantly in the wet season.
Chlorophyll fluorescence
The maximum quantum efficiency (Fv/Fm) showed no sig nificant differences in any of the comparisons (Fig. 3). The average value was 0.81±0.02, and no trends related to the field or the seasons were found. In the field with irrigation, the non-photochemical quenching (NPQ) of the palms showed no significant variations and was slightly higher in both materials during the wet season. However, the NPQ differences among seasons were found in the field with no irrigation for both materials and reached the highest values during the wet season. The photochemical efficiency of photosystem II (0PII) and the electron transfer rate (ETR) responded similarly. These variables are closely related and no significant differences were found among seasons for E. guineensis in the field with irrigation. However, the values for the wet season were lower than those found in the dry season.
Discussion
The carbon dioxide fixation by palms planted in Aurora field (irrigated) did not change with respect to the season of the year or the rainfall rate. Under optimal conditions for healthy, well-irrigated palms, the CO2 fixation rate showed a tendency to increase during the dry season, which suggests that the water supplied by irrigation not only offsets the lack of rain but also creates a synergy with the weather conditions that stimulate the gas exchange. Henson and Mohd Hanif (2005) confirmed that, during the dry season, the most immediate response is the sto-matal closure and thus the reduction of the transpiration rates. However, as shown in figure 2, although there is a decrease in the transpiration rates during the dry season, there were no significant differences among seasons for both E. guineensis and the hybrid. With an efficient ir rigation system, the plants had enough water available to keep the stomata open and prevent a reduction in the photosynthesis. This outcome is reinforced by the increase in the proportion of absorbed energy for the photosynthetic process with an increased Í>PSII (Fig. 3). Additionally, the maximum variable fluorescence of chlorophyll (Fv/Fm) re mained unchanged, which suggests that the dry season did not create stress on the palms in the field with irrigation. However, the availability of water during the dry season led to a significant increase in the efficient use of water for photosynthesis, which had previously been reported for other species in which the WUEp increases with irrigation during droughts (Avola et al., 2008; Kiziloglu et al., 2009; Singh and Reddy, 2011). Moreover, the water potential data showed that, with flood irrigation, the water potential re mained similar between materials and sampling periods, which suggest that, during the dry season, the plants did not experience water deficit stress.
In the field with no irrigation (Mecasaragua), the lack of water available during the dry season had a strong impact on the physiology of the palms. The gas exchange was heavily affected, and the leaf water potential significantly decreased, particularly in E. guineensis. The photosynthetic rates were reduced by 66% in the non-irrigated field com pared with the irrigated field, and the photosynthesis rate decreased by 70% compared with the palms of that same field during the wet season. Similar results had been ob tained in nursery plants under water deficit, in E. guineensis progenies (Jazayeri et al., 2015) and interspecific OxG hybrids (Rivera et al., 2013). However, the Fv/Fm had no significant changes, which suggest that the palms may have adaptive or early recovery mechanisms in the photosystem II, which was not affected by the water stress.
Furthermore, the amount of energy received during the dry season cannot be used by palms in the same way in which well-irrigated palms use it, especially because of the effects that the water stress can have on the cellular biochemistry (Silva et al., 2013). Therefore, there is a significant decrease in the i>PSII (Fig. 3) related to problems in the electron transfer rate (ETR). This decline is closely related to the stomatal closure because the transpiration rate went from 3.8 mmol m-2 s-1 in the wet season to 1.5 mmol m-2 s-1 dur ing the dry season.
The dry-to-wet transition season is a reflection of the metabolic plasticity of the palms in their rehydration process. The recovery of the photosynthesis and transpi ration levels in the field without irrigation suggests the presence of biochemical mechanisms that allow the cells to rapidly return to their normal state after water stress events. These results had been observed in young palms (Suresh et al., 2010) and other species such as olives (Sofo et al., 2009) that were subjected to water stress and then returned to an appropriate level of soil moisture. Under these conditions cases an excellent recovery of the plants after a severe stress was observed and the plants reached normal levels of the different variables measured in a short period of time.
These findings support the need for irrigation to prevent palms from being affected by the lack of rain. Kallarackal et al. (2004) found that regardless of the microclimate observed in three oil palm-growing regions of India, ir rigation prevented a negative effect on the production dur ing dry periods. However, determining the best irrigation system has been a widely discussed topic. For example, a review of 15 years of irrigation in southern Thailand (Tit-tinutchanon et al., 2008) showed no differences among four irrigation systems. Moreover, favoring any type of irrigation is associated with the cost of implementation. Hence, irrigation must always be aimed at increasing the production potential of oil palms (Mejia 2000).
Conclusions
The comparison of the gas exchange in the fields with and without irrigation in three contrasting seasons showed the drastic impact of the dry season on the photosynthesis rate in both the hybrids and E. guineensis, with a reduction of up to 70%. In addition, the capacity of the palms to recover after a period of water deficit stress was confirmed. Consid ering these results, it is recommended to supply irrigation to avoid a drastic reduction in the carbon sequestration. Future works may study the impact of irrigation on the bunch production to quantify the fruit loss caused by periods of water stress.