| dc.description.abstracteng | Oil palm (Elaeis guineensis Jacq.) and rubber (Hevea brasiliensis Müll. Arg.) plantations cover large areas in tropical regions and may still expand further. In contrast, the area covered by natural forests has strongly declined. From environmental perspectives, this raises concerns not only with respect to biodiversity but also regarding the integrity of the hydrological cycle including potential changes in transpiration. For rubber plantations, high evapotranspiration rates were reported from mainland Asia and it was indicated that rubber tree transpiration responds sensitively to dynamics in environmental drivers. Prior to our studies, little scientifically based information was available for oil palm. The objective of this study was to test and if necessary adjust a sap flux measurement technique for the two species, to analyze influences of plantation age and environmental drivers on oil palm and rubber tree transpiration and to delineate differences between the two plantation types. The study was implemented in the lowlands of Sumatra, Indonesia. Methods include lab experiments with thermal dissipation probes and gravimetric reference measurements, three different sap flux methods applied in the field (thermal dissipation probes, heat field deformation and stem heat balance), and a monitoring of sap flux on a total of 25 plots in lowland Sumatra.
Thus far, sap flux measurements had not been applied to oil palms. We first calibrated the thermal dissipation probe method (TDP) after Granier (1985) and tested a field sampling scheme. The probes were inserted into leaf petioles, as most likely the density of vessels is higher and vessels are distributed more homogenously than in the trunk. In the laboratory, we tested this set-up against gravimetric measurements and derived new, oil palm specific parameters for the calibration equation. In the field, in a 12-year-old plantation, 56 leaves on 10 palms were equipped with TDP sensors. We did not find significant influences of a series of studied variables and thus took an un-stratified approach to determine an appropriate sampling scheme to estimate stand transpiration rates. The relative standard error of the mean was used as a measure for the potential estimation error of stand transpiration associated with sample size. It was 14% for a sample size of 13 leaves to determine the average leaf water use and four palms to determine the average number of leaves per palm. Increasing these sample sizes only led to minor further decreases of the relative standard errors of the mean.
Oil palm transpiration rates were then monitored on 15 on-farm plots using this sap flux technique and sampling scheme in conjunction with eddy covariance measurements at two sites. Oil palm water use and transpiration increased from two to about eight-year old plantations and then leveled off up to an age of 22 years. Among medium-aged, 10 to 18 years old plantations, substantial spatial heterogeneity was found, with the highest oil palm stand transpiration rates (2.5 mm d–1) in a highly intensively managed plantation being as high as or even higher than transpiration of forests in the same region. However, average transpiration across several small-holder properties was much lower. It was also indicated that the temporal dynamics of transpiration in oil palm are “buffered”, which means that day-to-day transpiration rates fluctuate much less than micrometeorological drivers.
In rubber trees, lab experiments did not indicate significant differences to the original calibration equation. In a field trial, the TDP sensors combined with heat field deformation sensors resulted in tree water use rates very similar to those obtained by the independent stem heat balance method. We monitored sap flux on 10 on-farm plots in mono-cultural rubber plantations between 2 and 17 years old. Tree water use increased from two to about eight-year old plantations and then leveled off. We observed pronounced differences in the relation tree diameter to water use: it increased steeply in young plantations up to an age of eight years; in 14 and 16 yr old plantations the increase was much less pronounced. In the older plantations, sap flux densities decreased with increasing tree diameter, while the opposite was true for the younger plantations. Rubber tree transpiration showed a pronounced seasonality due to partial leaf shedding. Overall, the magnitude of transpiration was substantially lower than previously reported for rubber plantations on the Asian mainland. Potential reasons include differences in methods, climate and management. Data from the Asian mainland were mainly derived by the eddy covariance approach, which often results in estimates higher than those from sap flux studies. Differences in sap flux derived rates from both regions are much less pronounced. The climate in is more continental on the Asian mainland, and maritime in lowland Sumatra. Additionally, the studies in mainland Asia were conducted on well-managed experimental farms and with modern rubber tree clones. Our study was conducted on small-holder properties in the hinterland of Jambi, which are less sophisticated and intensively managed, and an age-related decline of water use rates was observed.
Oil palm and rubber plantations differed from each other mainly in the magnitude of transpiration, which is on average by ca. 30% higher in oil palm, and regarding the seasonality in rubber plantations as induced by periodic leaf-shedding. These differences between species and the suggestions made regarding climatic and management related influences on transpiration may be of interest to eco-hydrological assessments of post-forest tropical plantation landscapes. | de |